DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant's arguments filed 18 December 2025 have been fully considered but they are not persuasive.
The Applicant argues that Guner et al. fails to teach “labeling the respective unseen images according to a reservoir feature present in the respective unseen images using a computer vision model trained using images from a feature library.” The Examiner respectfully disagrees. Guner et al. discloses that the high-resolution resistivity image may depict boundaries of subsurface structures, such as a plurality of layers disposed in information 132. These formation images may be used in reservoir characterization. Formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs (paragraph [0029]). Furthermore, Fig. 15 shows a workflow 1500 for processing images using a regression function. In block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm. This operation may be performed in near real time, whenever enough depth samples needed for regression function are obtained. Real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log. Workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations. After the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use or may be used as a part of additional processing or calculation operations. The corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures. Workflow 1500 may be utilized to derive information on formation 132. This information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like (paragraph [0088]). As noted above, corrected data from workflow 1500 may be utilized for many other downhole operations, processing operations, and/or the like. For example, corrected data may be utilized in pad-to-pad level equalization and automatic dip picking from corrected images (paragraph [0090]). Dip picking is the process of identifying and characterizing planar geological structures, such as bedding planes or factures, from borehole images. By characterizing planar geological structures, the supervised machine learning algorithm is labeling different characteristics found in the corrected images. Therefore, Guner et al. meets the claimed limitations and the rejection is maintained. The Examiner suggests amending the claims to include that the labeling is done on a report output for display from the computer vision module rather than just being done internally in the system.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-7, 10-18, and 22 are rejected under 35 U.S.C. 103 as being unpatentable over Guner et al. (U.S. Patent Application Publication 2023/0245278) in view of Kowalik et al. (U.S. Patent Application Publication 2009/0055097).
Regarding claim 1, Guner et al. discloses a computer-implemented method for characterization of reservoir features, the method comprising: extracting, with one or more hardware processors, unseen images from a borehole image log (Figs. 1, 3, and 15; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like); labeling, with the one or more hardware processors, respective unseen image according to a reservoir feature present in the respective unseen images using a computer vision model trained using images from a feature library, wherein the computer vision mode is a convolutional neural network (CNN) that obtains the respective unseen images as input and outputs at least one reservoir feature detected in the respective unseen images (Figs. 1, 3, 13, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0073] – a supervised machine learning algorithm may utilize an artificial neural network to process measurements from measurement operations to form an image during processing operations – Fig. 13 illustrates an artificial neural network 1300 with one or more inputs 1302 and one or more outputs 1304; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like); and interpreting, with the one or more hardware processors, the borehole image log according to the labeled images (Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like). However, Guner et al. fails to explicitly disclose segmenting, with one or more hardware processors, unseen images, wherein respective unseen images are segmented from the borehole image log according to contrast variations in the borehole image log.
Referring to the Kowalik et al. reference, Kowalik et al. discloses a computer-implemented method for characterization of features, the method comprising: segmenting, with one or more hardware processors, unseen images, wherein respective unseen images are segmented from the borehole image log according to contrast variations in the borehole image log (Fig. 4 – illustration of image data and segmented image data representing resistivity measurements; Figs. 6 and 7; paragraph [0024] – the image data 30 is segmented based on, for example, a homogeneity of local areas to produce segmented image data 30’ or 30’’ – the segmentation may be performed by analyzing the image data using software that determines image homogeneity based at least in part on color and location of picture elements; paragraph [0027] – the data are transformed into image data at 36, and segmented at 38 to produce segmented image data 30’’ – a series of hierarchical classification rules is applied to the segments of the segmented image data 30’’, as illustrated in Fig. 7; paragraph [0033] – while bedded segments may be separated from non-bedded segments using texture and color of the resistivity data).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have had segmented the unseen images according to contrast variations as disclosed by Kowalik et al. in the method disclosed by Guner et al. in order to help detect and classify the resistivity data found in the images.
Regarding claim 2, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 1 including that wherein the computer vision model is trained using images stored in the feature library and obtained from published borehole image logs (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0070] – in a supervised machine learning algorithm, an optimal function that relates a given set of inputs and outputs function that relates a given set of inputs and outputs is found where the outputs corresponding to given inputs are already known (published information) – a regression function is found from a dataset with known inputs and outputs – this dataset is denoted as the training, validation and testing dataset in this disclosure; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 3, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 1 including that wherein the computer vision model is trained using images stored in the feature library and obtained from available field historical data (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 4, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claims 1 and 3 including that wherein the images stored in the feature library and obtained from available field historical data are preprocessed (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 5, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 1 including that wherein the respective unseen images from the borehole image log are preprocessed to increase a resolution of the respective unseen images, resize the respective unseen images, or increase a quality of the respective unseen images (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 6, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 1 including that wherein interpreting the borehole image log according to the labeled images comprises characterization of subterranean features of the reservoir (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 7, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 1 including that wherein the segmenting, labeling, and interpreting occurs in real time or utilizing a memory gauge (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like; Kowalik et al.: Fig. 4 – illustration of image data and segmented image data representing resistivity measurements; Figs. 6 and 7; paragraph [0024] – the image data 30 is segmented based on, for example, a homogeneity of local areas to produce segmented image data 30’ or 30’’ – the segmentation may be performed by analyzing the image data using software that determines image homogeneity based at least in part on color and location of picture elements; paragraph [0027] – the data are transformed into image data at 36, and segmented at 38 to produce segmented image data 30’’ – a series of hierarchical classification rules is applied to the segments of the segmented image data 30’’, as illustrated in Fig. 7; paragraph [0033] – while bedded segments may be separated from non-bedded segments using texture and color of the resistivity data).
Regarding claim 10, Guner et al. discloses an apparatus comprising a non-transitory, computer readable, storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising: extracting unseen images from a borehole image log (Figs. 1, 3, and 15; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like); labeling a respective unseen image according to a reservoir feature present in the respective unseen image using a computer vision model trained using images from a feature library, wherein the computer vision model is a convolutional neural network (CNN) that obtains the respective unseen images as input and outputs at least one reservoir feature detected in the respective unseen images (Figs. 1, 3, 13, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0073] – a supervised machine learning algorithm may utilize an artificial neural network to process measurements from measurement operations to form an image during processing operations – Fig. 13 illustrates an artificial neural network 1300 with one or more inputs 1302 and one or more outputs 1304; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like); and interpreting the borehole image log according to the labeled image (Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like). However, Guner et al. fails to explicitly disclose segmenting, with one or more hardware processors, unseen images, wherein respective unseen images are segmented from the borehole image log according to contrast variations in the borehole image log.
Referring to the Kowalik et al. reference, Kowalik et al. discloses an apparatus comprising a non-transitory, computer readable, storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising: segmenting, with one or more hardware processors, unseen images, wherein respective unseen images are segmented from the borehole image log according to contrast variations in the borehole image log (Fig. 4 – illustration of image data and segmented image data representing resistivity measurements; Figs. 6 and 7; paragraph [0024] – the image data 30 is segmented based on, for example, a homogeneity of local areas to produce segmented image data 30’ or 30’’ – the segmentation may be performed by analyzing the image data using software that determines image homogeneity based at least in part on color and location of picture elements; paragraph [0027] – the data are transformed into image data at 36, and segmented at 38 to produce segmented image data 30’’ – a series of hierarchical classification rules is applied to the segments of the segmented image data 30’’, as illustrated in Fig. 7; paragraph [0033] – while bedded segments may be separated from non-bedded segments using texture and color of the resistivity data).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have had segmented the unseen images according to contrast variations as disclosed by Kowalik et al. in the apparatus disclosed by Guner et al. in order to help detect and classify the resistivity data found in the images.
Regarding claim 11, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 10 including that wherein the computer vision model is trained using images stored in the feature library and obtained from published borehole image logs (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0070] – in a supervised machine learning algorithm, an optimal function that relates a given set of inputs and outputs function that relates a given set of inputs and outputs is found where the outputs corresponding to given inputs are already known (published information) – a regression function is found from a dataset with known inputs and outputs – this dataset is denoted as the training, validation and testing dataset in this disclosure; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 12, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 10 including that wherein the computer vision model is trained using images stored in the feature library and obtained from available field historical data (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 13, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claims 10 and 12 including that wherein the images stored in the feature library and obtained from available field historical data are preprocessed (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 14, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 10 including that wherein the respective unseen images from the borehole image log are preprocessed to increase a resolution of the respective unseen images, resize the respective unseen images, or increase a quality of the respective unseen images (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 15, Guner et al. discloses a system, comprising: one or more memory modules; one or more hardware processors communicably coupled to the one or more memory modules, the one or more hardware processors configured to execute instructions stored on the one or more memory models to perform operations comprising: extracting unseen images from a borehole image log (Figs. 1, 3, and 15; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like); labeling respective unseen images according to a reservoir feature present in the respective unseen images using a computer vision model trained using images from a feature library, wherein the computer vision model is a convolutional neural network (CNN) that obtains the respective unseen images as inputs and outputs at least one reservoir feature detected in the respective unseen images (Figs. 1, 3, 13, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0073] – a supervised machine learning algorithm may utilize an artificial neural network to process measurements from measurement operations to form an image during processing operations – Fig. 13 illustrates an artificial neural network 1300 with one or more inputs 1302 and one or more outputs 1304; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like); and interpreting the borehole image log according to the labeled images (Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like). However, Guner et al. fails to explicitly disclose segmenting, with one or more hardware processors, unseen images, wherein respective unseen images are segmented from the borehole image log according to contrast variations in the borehole image log.
Referring to the Kowalik et al. reference, Kowalik et al. discloses a system, comprising: segmenting, with one or more hardware processors, unseen images, wherein respective unseen images are segmented from the borehole image log according to contrast variations in the borehole image log (Fig. 4 – illustration of image data and segmented image data representing resistivity measurements; Figs. 6 and 7; paragraph [0024] – the image data 30 is segmented based on, for example, a homogeneity of local areas to produce segmented image data 30’ or 30’’ – the segmentation may be performed by analyzing the image data using software that determines image homogeneity based at least in part on color and location of picture elements; paragraph [0027] – the data are transformed into image data at 36, and segmented at 38 to produce segmented image data 30’’ – a series of hierarchical classification rules is applied to the segments of the segmented image data 30’’, as illustrated in Fig. 7; paragraph [0033] – while bedded segments may be separated from non-bedded segments using texture and color of the resistivity data).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have had segmented the unseen images according to contrast variations as disclosed by Kowalik et al. in the system disclosed by Guner et al. in order to help detect and classify the resistivity data found in the images.
Regarding claim 16, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 15 including that wherein the computer vision model is trained using images stored in the feature library and obtained from published borehole image logs (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0070] – in a supervised machine learning algorithm, an optimal function that relates a given set of inputs and outputs function that relates a given set of inputs and outputs is found where the outputs corresponding to given inputs are already known (published information) – a regression function is found from a dataset with known inputs and outputs – this dataset is denoted as the training, validation and testing dataset in this disclosure; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 17, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 15 including that wherein interpreting the borehole image log according to the labeled images comprises characterization of subterranean features of the reservoir (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Regarding claim 18, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 15 including that wherein the segmenting, labeling, and interpreting occurs in real time or utilizing a memory gauge (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like; Kowalik et al.: Fig. 4 – illustration of image data and segmented image data representing resistivity measurements; Figs. 6 and 7; paragraph [0024] – the image data 30 is segmented based on, for example, a homogeneity of local areas to produce segmented image data 30’ or 30’’ – the segmentation may be performed by analyzing the image data using software that determines image homogeneity based at least in part on color and location of picture elements; paragraph [0027] – the data are transformed into image data at 36, and segmented at 38 to produce segmented image data 30’’ – a series of hierarchical classification rules is applied to the segments of the segmented image data 30’’, as illustrated in Fig. 7; paragraph [0033] – while bedded segments may be separated from non-bedded segments using texture and color of the resistivity data).
Regarding claim 22, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 1 including that storing the labeled images in the feature library; and updated the computer vision model by training the computer vision model on the labeled images stored in the feature library (Guner et al.: Figs. 1, 3, and 15; paragraph [0029] – these formation images may be used in reservoir characterization – formation images with high resolution may allow accurate identification of thin beds and other fine features such as fractures, clasts, and vugs – these formation images may provide information about the sedimentology, lithology, porosity, and permeability of formation 132; paragraph [0070] – in a supervised machine learning algorithm, an optimal function that relates a given set of inputs and outputs function that relates a given set of inputs and outputs is found where the outputs corresponding to given inputs are already known (published information) – a regression function is found from a dataset with known inputs and outputs – this dataset is denoted as the training, validation and testing dataset in this disclosure; paragraph [0077] – additionally, information handling system 114 may utilize communication protocol 1406 to access processed measurements, troubleshooting findings, historical run data, and/or the like; paragraph [0084] – the dataset is obtained from traditional processing techniques, samples from a section or sections of a log or logs with high image quality after the processing may be selected – high image quality is defined as a section of the image where over eighty-fiver percent of the artifacts and noise in the image are processed out successfully such that the true underlying features are easily discernible; paragraph [0086] – neural network parameters may be updated in order to improve results; paragraph [0088] – Fig. 15 shows a workflow 1500 for processing images using a regression function – workflow 1500 may begin block 1502 in which measurements for a depth, or a plurality of depths, are obtained from a downhole tool 102 (e.g., referring to Fig. 1) during measurement operations – the measurements from block 1502 are send to an information handling system 114 (e.g., referring to Fig. 1) for further processing in block 1504 – in block 1504, one or more corrections, such as calibration and scaling with a tool constant, may be applied to the measurements from block 1502 to produce a raw image – in block 1506, the raw image, is then used in a regression function obtained by training a supervised machine learning algorithm - this operation may be performed in near real time, whenever enough depth samples needed for the regression function are obtained - real time logging operations refer to cases where processing is performed in a short enough time frame to enable making decisions on how the logging should proceed without performing costly and time-consuming actions to return the tool string to a previous position or starting over with a new log - workflow 1500 is performed and repeated as new data arrives for additional measurements taken during measurement operations - note that, if the number of azimuthal points in an image is less than the total number of azimuthal measurements, multiple images from a single depth point may be processed in parallel - for example, as discussed previously, images may only contain data from a single pad 134 of downhole tool 102 (e.g., referring to FIG. 1) and measurements from each pad 134 may be processed in parallel - processing may also be performed after the logging (i.e., measurement operations) is completed by dividing the measurements into images as performed by the algorithm - after the processing, an image (i.e., a corrected image) may be displayed directly to a user, it may be stored for later use (for example on information handling system 114) or may be used as a part of additional processing or calculation operations - the corrected image may be utilized to identify formation properties and/or geology of the formation such as formation resistivity, formation permittivity, standoff, layering of the formation, presence of washouts and fractures - workflow 1500 may be utilized to derive information on formation 132 (e.g., referring to FIG. 1) - this information derived from the corrected image guide logging decisions such as adjusting logging speed, determining whether a relog needs to be made, adjusting pad pressure and operating frequencies of the tool, modifying power level, and/or the like).
Claims 9 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Guner et al. in view of Kowalik et al. as applied to claims 1 and 15 above, and further in view of Ruel et al. (U.S. Patent Application Publication 2023/0122264).
Regarding claim 9, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 1, but fails to disclose that wherein labeling the respective unseen images using the trained computer vision model further comprises classifying the segments of the borehole image log using a you only look once (YOLO) network.
Referring to the Ruel et al. reference, Ruel et al. discloses a computer-implemented method for characterization of reservoir features, the method comprising: classifying the segments of the borehole image log using a you only look once (YOLO) network (paragraph [0090] – DNN models are available for retraining (mobile-vetv2, YOLO, etc.…), which means the DNN is structured in a way that it knows how to efficiently extract and organize the features found in an image).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have had used a you only look once (YOLO) network as disclosed by Ruel et al. in the method disclosed by Guner et al. in view of Kowalik et al. in order to efficiently extract and organize the features found in an image.
Regarding claim 20, Guner et al. in view of Kowalik et al. discloses all of the limitations as previously discussed with respect to claim 15, but fails to disclose that wherein labeling the respective unseen images using the trained computer vision model further comprises classifying the segments of the borehole image log using a you only look once (YOLO) network.
Referring to the Ruel et al. reference, Ruel et al. discloses a system for characterization of reservoir features, the system comprising: classifying the segments of the borehole image log using a you only look once (YOLO) network (paragraph [0090] – DNN models are available for retraining (mobile-vetv2, YOLO, etc.…), which means the DNN is structured in a way that it knows how to efficiently extract and organize the features found in an image).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have had used a you only look once (YOLO) network as disclosed by Ruel et al. in the system disclosed by Guner et al. in view of Kowalik et al. in order to efficiently extract and organize the features found in an image.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HEATHER R JONES whose telephone number is (571)272-7368. The examiner can normally be reached Mon. - Fri.: 9:00am - 5:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, William Vaughn can be reached on (571)272-3922. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/HEATHER R JONES/Primary Examiner, Art Unit 2481
April 13, 2026
Prosecution Insights