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 11/11/2025 have been fully considered but they are not persuasive. Regarding claims 1, 7, and 13, Applicant argues (pg. 7-8 of the Remarks) that Sibley fails to teach “automatically apply a programmable image-selection parameter to the subset of the images to select one or more of the images, from the subset of the images, to store for machine-learning training, the image- selection parameter applied while the system moves across the agricultural field, the image- selection parameter including one or more target image properties.” Examiner respectfully disagrees. Sibley teaches (¶0107, ¶0109, ¶0227) identifying, storing, and indexing of images of crops and weeds using machine learning feature extraction; (¶0383) The first portion, mainly used for feature extraction and object detection, or classification, or both using a trained ML model, may be continually updated upon further training and implemented by the onsite platform 10400. For example, a machine learning model can be configured to determine that certain image frames, or sequence of continuous frames, ingested from a plurality of continuous frames ingested in an observation and treatment trial on a geographic boundary do not detect any objects of interest for targeting or from omitting from targeting. For example, a vehicle can pass through a patch of dirt without, in reality, any weeds, plants, or crops for a few meters. Upon uploading a continuous set of image frames to a server for analysis and for labelling a machine learning algorithm can be applied to detected an optimal subset of frames for human labelling or quality control, including for example, excluding the sequence of image frames capturing the few meters of dirt without any weeds, plants, or crops (implies that the frames that have weeds, plants, or crops are included). Once common landmarks are identified, the system via active learning can produce a subset of image frames for human labelling or quality control by removing or reducing images that have common landmarks as that of other images to further reduce redundancy of image quality analysis. Applied to further provide resources as training datasets for training the machine learning model; Additionally/alternatively examiner notes that Sibley’s Key frames can also be considered as subset of images Sibley teaches (¶0207) Landmarks can be used to identify which frames are of interest to store, store as a keyframe (because one does not need so many frames at once all having the same fruits, or detected objects, from frame to frame); (¶0101-¶0102 and ¶0130) as the vehicle 310 passes by a particular agricultural object in the real world, the object determination and object treatment engine can capture images and reconstruct a digital or virtual geographic scene representing the geographic scene; (¶0421) a storage, and a treatment mechanism, comprising obtaining (152200), by the treatment system mountable on an agricultural vehicle and configured to implement a machine learning (ML) algorithm, one or more images of a region of an agricultural environment near the treatment system, wherein the one or more images are captured from the region of a real-world where agricultural target objects are expected to be present; determining (152400), by the treatment system, one or more parameters for use with the ML algorithm, wherein at least one of the one or more parameters is based on one or more ML models related to identification of an agricultural object; determining (152600), by the treatment system, a real-world target in the one or more images using the ML algorithm, wherein the ML algorithm is at least partly implemented using the one or more processors of the treatment system.
Applicant further argues (pg. 8 of the Remarks) that for Sibley the “labeling is not performed on subset of images” and “labeling is performed at the offsite computing resource” and not by a system as it moves across an agricultural field. Examiner respectfully disagrees. In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., “labeling is not performed on subset of images” and “labeling is performed …by a system as it moves across an agricultural field”) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993). However, Sibley teaches (¶0383) a machine learning algorithm can be applied to detected an optimal subset of frames for labelling; (¶0205 and ¶0108) a labelled image, either from real-time performed by an agricultural observation and treatment system on the vehicle, or offline at a server, by a human, by a machine learning algorithm, assisted by a machine learning algorithm, or a combination thereof.
Applicant further argues (pg. 8-9 of the Remarks) “in addition to these [the above mentioned] arguments, Applicant addresses the specific Sibley paragraph citations included in the Office Action” and proceeds to quote various portions of the office action and saying that Sibley does not teach what the examiner has cited. Examiner respectfully disagrees. In the office action, hereinbelow, the examiner has provided a paragraph number with either a direct excerpt or a summarized description/reading of each of the cited paragraphs of Sibley. If there is a disagreement as to how the cited portions of Sibley relate to the claims of the instant application, examiner urges applicant to please describe the differences.
Applicant further argues (pg. 10) that Sibley does not teach that the image selection parameter represents “a confidence level that the target plant is detected.” Examiner respectfully disagrees. Sibley teaches (¶0383) Additionally, active learning techniques can be applied such that one or more machine learning algorithms analyzes an entire set of ingested images and performed detections, classifications, labelling pixel classification, or bounding box labelling, such that detections above a certain threshold can be used as training data and those that do not meet a threshold can be sent to a human for labelling, classifying, or performing quality control. (¶0336) confidence level of the ML detection (i.e., object detection) and the tracking algorithm; (¶0386) objects are plant objects of a certain type.
Applicant’s other arguments do no apply to the rejection found hereinbelow.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 7, 13, 15 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Sibley.
Regarding claim 7 “A system comprising: an agricultural vehicle” Sibley teaches (¶0006, ¶0086, ¶0093, ¶0307) a system comprising an agricultural vehicle.
As to “one or more cameras in mechanical communication with the agricultural vehicle, the one or more cameras configured to capture images of an agricultural field in a direction of movement of the agricultural vehicle” Sibley teaches (¶0101-¶0102 and ¶0130) as the vehicle 310 passes by a particular agricultural object in the real world, the object determination and object treatment engine can capture images and reconstruct a digital or virtual geographic scene representing the geographic scene.
As to “one or more microprocessors in communication with the one or more cameras, the one or more microprocessors in mechanical communication with the agricultural vehicle” Sibley teaches (¶0090) image processing module; (¶0127) ECU to control cameras; (¶0126 and ¶0128) compute module to receive and process image data; (¶0303, ¶0080, ¶0130) embodiments implemented using computer system with processing unit, for autonomous driving system described.
As to “non-volatile computer memory operatively coupled to the one or more microprocessors, the non-volatile computer memory in mechanical communication with the agricultural vehicle, the non-volatile computer memory storing computer-readable instructions that, when executed by the computer, cause the one or more microprocessors to” Sibley teaches (¶0303) computer system includes processor and memory storing instructions for performing methods and steps described.
As to “automatically analyze each image for a presence of at least one target plant using a trained machine-learning model” Sibley teaches (¶0164, ¶0168, ¶0177, ¶0277) images captured are used to detect specific known objects using a machine learning model (e.g., weeds or carrots.)
As to “the trained machine-learning model having been trained with first images that include the at least one target plant and second images that do not include the at least one target plant” Sibley teaches (¶0308, ¶0328, ¶0331, ¶0368) ML algorithm is trained to identify crops and weeds; (¶0383) Active learning can also be applied for example, to determine common landmarks from frame to frame that are not necessarily plant objects of interest, such as target plants for treating. Machine learning algorithms analyzes an entire set of ingested images and performed detections, classifications, labelling pixel classification, or bounding box labelling, such that detections above a certain threshold can be used as training data and those that do not meet a threshold can be sent to a human for labelling, classifying, or performing quality control; (¶0384-¶0386, ¶0389, ¶0391, ¶0392) training data includes exclusion targets and inclusion targets and different types of datasets.
As to “automatically detect, using the trained machine-learning model, the at least one target plant in a subset of the images, automatically apply a programmable image-selection parameter to the subset of the images to select one or more of the images, from the subset of the images, to store for machine-learning training, the image-selection parameter applied while the system moves across the agricultural field” Sibley teaches (¶0107, ¶0109, ¶0227) identifying, storing, and indexing of images of crops and weeds using machine learning feature extraction; (¶0383) The first portion, mainly used for feature extraction and object detection, or classification, or both using a trained ML model, may be continually updated upon further training and implemented by the onsite platform 10400. For example, a machine learning model can be configured to determine that certain image frames, or sequence of continuous frames, ingested from a plurality of continuous frames ingested in an observation and treatment trial on a geographic boundary do not detect any objects of interest for targeting or from omitting from targeting. For example, a vehicle can pass through a patch of dirt without, in reality, any weeds, plants, or crops for a few meters. Upon uploading a continuous set of image frames to a server for analysis and for labelling a machine learning algorithm can be applied to detected an optimal subset of frames for human labelling or quality control, including for example, excluding the sequence of image frames capturing the few meters of dirt without any weeds, plants, or crops (implies that the frames that have weeds, plants, or crops are included). Once common landmarks are identified, the system via active learning can produce a subset of image frames for human labelling or quality control by removing or reducing images that have common landmarks as that of other images to further reduce redundancy of image quality analysis. Applied to further provide resources as training datasets for training the machine learning model; (¶0207) Landmarks can be used to identify which frames are of interest to store, store as a keyframe (because one does not need so many frames at once all having the same fruits, or detected objects, from frame to frame); (¶0192) each portion of the image that includes agricultural objects can be labeled and assigned a unique identifier to be indexed in a database; (¶0366, ¶0368) ML algorithm may be programmed to eliminate known objects from the images (e.g., carrots or another crop) and the remaining objects may be classified as being unwanted objects; (¶0358) weeds are indexed; (¶0105 and ¶0421) capture more than one view of object 353-a and store all of the different frames that include object.
As to “and automatically store only the one or more of the images, selected using the image-selection parameter, for the machine-learning training in a computer storage device operably coupled to the one or more microprocessors.” Sibley teaches (¶0129) storing information related to agricultural objects, scenes, environments, images and videos related to agricultural objects and terrain, training data for machine learning algorithms, raw data captured by image capture devices or other sensing devices, processed data such as a repository of indexed images of agricultural objects; (¶0227) object/feature detection, used for creating a time lapse visualization, mapping the object, generating key frames with detections for indexing and storage, diagnosing and improving machine learning models, etc; (¶0383-¶0384) supervised and unsupervised learning; (¶0386) acquiring data for improving ML models during use.
Regarding claim 13, A system comprising: an agricultural vehicle” Sibley teaches (¶0006, ¶0086, ¶0093, ¶0307) a system comprising an agricultural vehicle.
As to “one or more cameras in mechanical communication with the agricultural vehicle, the one or more cameras configured to capture images of an agricultural field in a direction of movement of the agricultural vehicle” Sibley teaches (¶0101-¶0102 and ¶0130) as the vehicle 310 passes by a particular agricultural object in the real world, the object determination and object treatment engine can capture images and reconstruct a digital or virtual geographic scene representing the geographic scene.
As to “one or more microprocessors in communication with the one or more cameras, the one or more microprocessors in mechanical communication with the agricultural vehicle” Sibley teaches (¶0090) image processing module; (¶0127) ECU to control cameras; (¶0126 and ¶0128) compute module to receive and process image data; (¶0303, ¶0080, ¶0130) embodiments implemented using computer system with processing unit, for autonomous driving system described.
As to “and non-volatile computer memory operatively coupled to the one or more microprocessors, the non-volatile computer memory in mechanical communication with the agricultural vehicle, the non-volatile computer memory storing computer-readable instructions that, when executed by the one or more microprocessors, cause the one or more microprocessors to” Sibley teaches (¶0303) computer system includes processor and memory storing instructions for performing methods and steps described.
As to “automatically analyze each image for a presence of at least one target plant using a trained machine-learning model” Sibley teaches (¶0164, ¶0168, ¶0177, ¶0277) images captured are used to detect specific known objects using a machine learning model (e.g., weeds or carrots.)
As to “the trained machine-learning model having been trained with first images that include the at least one target plant and second images that do not include the at least one target plant” Sibley teaches (¶0308, ¶0328, ¶0331, ¶0368) ML algorithm is trained to identify crops and weeds; (¶0383) Active learning can also be applied for example, to determine common landmarks from frame to frame that are not necessarily plant objects of interest, such as target plants for treating. Machine learning algorithms analyzes an entire set of ingested images and performed detections, classifications, labelling pixel classification, or bounding box labelling, such that detections above a certain threshold can be used as training data and those that do not meet a threshold can be sent to a human for labelling, classifying, or performing quality control; (¶0384-¶0386, ¶0389, ¶0391, ¶0392) training data includes exclusion targets and inclusion targets and different types of datasets.
As to “automatically detect, using the trained machine-learning model, the at least one target plant in a subset of the images, automatically apply a programmable image-selection parameter to the subset of the images to select one or more of the images, from the subset of images, to store for machine-learning training, the image-selection parameter applied while the system moves across the agricultural field, the image-selection parameter representing one or more outputs from the trained machine-learning model” Sibley teaches (¶0107, ¶0109, ¶0227) identifying, storing, and indexing of images of crops and weeds using machine learning feature extraction; (¶0383) The first portion, mainly used for feature extraction and object detection, or classification, or both using a trained ML model, may be continually updated upon further training and implemented by the onsite platform 10400. For example, a machine learning model can be configured to determine that certain image frames, or sequence of continuous frames, ingested from a plurality of continuous frames ingested in an observation and treatment trial on a geographic boundary do not detect any objects of interest for targeting or from omitting from targeting. For example, a vehicle can pass through a patch of dirt without, in reality, any weeds, plants, or crops for a few meters. Upon uploading a continuous set of image frames to a server for analysis and for labelling a machine learning algorithm can be applied to detected an optimal subset of frames for human labelling or quality control, including for example, excluding the sequence of image frames capturing the few meters of dirt without any weeds, plants, or crops (implies that the frames that have weeds, plants, or crops are included). Once common landmarks are identified, the system via active learning can produce a subset of image frames for human labelling or quality control by removing or reducing images that have common landmarks as that of other images to further reduce redundancy of image quality analysis. Applied to further provide resources as training datasets for training the machine learning model; (¶0207) Landmarks can be used to identify which frames are of interest to store, store as a keyframe (because one does not need so many frames at once all having the same fruits, or detected objects, from frame to frame); (¶0192) each portion of the image that includes agricultural objects can be labeled and assigned a unique identifier to be indexed in a database; (¶0366, ¶0368) ML algorithm may be programmed to eliminate known objects from the images (e.g., carrots or another crop) and the remaining objects may be classified as being unwanted objects; (¶0358) weeds are indexed; (¶0105 and ¶0421) capture more than one view of object 353-a and store all of the different frames that include object.
As to “and automatically store only the one or more of the images, selected using the image-selection parameter, for the machine-learning training in a computer storage device operably coupled to the one or more microprocessors.” Sibley teaches (¶0129) storing information related to agricultural objects, scenes, environments, images and videos related to agricultural objects and terrain, training data for machine learning algorithms, raw data captured by image capture devices or other sensing devices, processed data such as a repository of indexed images of agricultural objects; (¶0227) object/feature detection using machine learning, used for creating a time lapse visualization, mapping the object, generating key frames with detections for indexing and storage, diagnosing and improving machine learning models, etc; (¶0383-¶0384) supervised and unsupervised learning; (¶0386) acquiring data for improving ML models during use.
Regarding claim 15, “The system of claim 13, wherein the one or more outputs from the trained machine-learning model include a confidence level that the target plant is detected.” Sibley teaches (¶0383) Additionally, active learning techniques can be applied such that one or more machine learning algorithms analyzes an entire set of ingested images and performed detections, classifications, labelling pixel classification, or bounding box labelling, such that detections above a certain threshold can be used as training data and those that do not meet a threshold can be sent to a human for labelling, classifying, or performing quality control. (¶0336) confidence level of the ML detection (i.e., object detection) and the tracking algorithm; (¶0386) objects are plant objects of a certain type.
Claim Rejections - 35 USC § 103
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.
Claim(s) 1-4, 6 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sibley et al. (US 20220183208, hereinafter Sibley) in view of Singh (US 20230115704, hereinafter Singh.)
Regarding claim 1, “A system comprising: an agricultural vehicle” Sibley teaches (¶0006, ¶0086, ¶0093, ¶0307) a system comprising an agricultural vehicle.
As to “one or more cameras in mechanical communication with the agricultural vehicle, the one or more cameras configured to capture images of an agricultural field in a direction of movement of the agricultural vehicle” Sibley teaches (¶0101-¶0102 and ¶0130) as the vehicle 310 passes by a particular agricultural object in the real world, the object determination and object treatment engine can capture images and reconstruct a digital or virtual geographic scene representing the geographic scene.
As to “one or more microprocessors in communication with the one or more cameras, the one or more microprocessors in mechanical communication with the agricultural vehicle” Sibley teaches (¶0090) image processing module; (¶0127) ECU to control cameras; (¶0126 and ¶0128) compute module to receive and process image data; (¶0303, ¶0080, ¶0130) embodiments implemented using computer system with processing unit, for autonomous driving system described.
As to “non-volatile computer memory operatively coupled to the one or more microprocessors, the non-volatile computer memory in mechanical communication with the agricultural vehicle, the non-volatile computer memory storing computer-readable instructions that, when executed by the one or more microprocessors, cause the one or more microprocessors to” Sibley teaches (¶0303) computer system includes processor and memory storing instructions for performing methods and steps described.
As to “automatically analyze each image for a presence of at least one target plant using a trained machine-learning model” Sibley teaches (¶0164, ¶0168, ¶0177, ¶0277) images captured are used to detect specific known objects using a machine learning model (e.g., weeds or carrots.)
As to “the trained machine-learning model having been trained with first images that include the at least one target plant and second images that do not include the at least one target plant” Sibley teaches (¶0308, ¶0328, ¶0331, ¶0368) ML algorithm is trained to identify crops and weeds; (¶0383) Active learning can also be applied for example, to determine common landmarks from frame to frame that are not necessarily plant objects of interest, such as target plants for treating. Machine learning algorithms analyzes an entire set of ingested images and performed detections, classifications, labelling pixel classification, or bounding box labelling, such that detections above a certain threshold can be used as training data and those that do not meet a threshold can be sent to a human for labelling, classifying, or performing quality control; (¶0384-¶0386, ¶0389, ¶0391, ¶0392) training data includes exclusion targets and inclusion targets and different types of datasets.
As to “automatically detect, using the trained machine-learning model, the at least one target plant in a subset of the images, automatically apply a programmable image-selection parameter to the subset of the images to select one or more of the images, from the subset of the images, to store for machine-learning training, the image-selection parameter applied while the system moves across the agricultural field” Sibley teaches (¶0107, ¶0109, ¶0227) identifying, storing, and indexing of images of crops and weeds using machine learning feature extraction; (¶0383) The first portion, mainly used for feature extraction and object detection, or classification, or both using a trained ML model, may be continually updated upon further training and implemented by the onsite platform 10400. For example, a machine learning model can be configured to determine that certain image frames, or sequence of continuous frames, ingested from a plurality of continuous frames ingested in an observation and treatment trial on a geographic boundary do not detect any objects of interest for targeting or from omitting from targeting. For example, a vehicle can pass through a patch of dirt without, in reality, any weeds, plants, or crops for a few meters. Upon uploading a continuous set of image frames to a server for analysis and for labelling a machine learning algorithm can be applied to detected an optimal subset of frames for human labelling or quality control, including for example, excluding the sequence of image frames capturing the few meters of dirt without any weeds, plants, or crops (implies that the frames that have weeds, plants, or crops are included). Once common landmarks are identified, the system via active learning can produce a subset of image frames for human labelling or quality control by removing or reducing images that have common landmarks as that of other images to further reduce redundancy of image quality analysis. Applied to further provide resources as training datasets for training the machine learning model; (¶0207) Landmarks can be used to identify which frames are of interest to store, store as a keyframe (because one does not need so many frames at once all having the same fruits, or detected objects, from frame to frame); (¶0192) each portion of the image that includes agricultural objects can be labeled and assigned a unique identifier to be indexed in a database; (¶0366, ¶0368) ML algorithm may be programmed to eliminate known objects from the images (e.g., carrots or another crop) and the remaining objects may be classified as being unwanted objects; (¶0358) weeds are indexed; (¶0105 and ¶0421) capture more than one view of object 353-a and store all of the different frames that include object.
As to “and automatically store only the one or more of the images, selected using the image-selection parameter, for the machine-learning training in a computer storage device operably coupled to the one or more microprocessors.” Sibley teaches (¶0129) storing information related to agricultural objects, scenes, environments, images and videos related to agricultural objects and terrain, training data for machine learning algorithms, raw data captured by image capture devices or other sensing devices, processed data such as a repository of indexed images of agricultural objects; (¶0227) object/feature detection, used for creating a time lapse visualization, mapping the object, generating key frames with detections for indexing and storage, diagnosing and improving machine learning models, etc; (¶0383-¶0384) supervised and unsupervised learning; (¶0386) acquiring data for improving ML models during use.
Sibley does not teach “the image-selection parameter controlling a volume of collected images” However, Singh teaches (¶0041) The storage policy may include a rule such as to store image frame(s) from an image stream based on a sample rate (e.g., store n images out of every consecutive m images; store j images every k seconds). Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the system as taught by Sibley with the storage policy based on a sample rate as taught by Singh for the benefit of saving memory/storage space.
Regarding claim 2, “The system of claim 1, wherein: the subset is a first subset, the one or more images are one or more first images, and the computer-readable instructions, when executed by the computer, further cause the one or more microprocessors to: automatically apply the image-selection parameter to a second subset of the images to select one or more second images, from the second subset of the images, to store for the machine-learning training, the trained machine-learning model not detecting the at least one target plant in the second subset of the images, and automatically store only the one or more second images, selected using the image-selection parameter, for the machine-learning training in the computer storage device.” Sibley teaches (¶0129) storing information related to agricultural objects, scenes, environments, images and videos related to agricultural objects and terrain, training data for machine learning algorithms, raw data captured by image capture devices or other sensing devices, processed data such as a repository of indexed images of agricultural objects; (¶0308, ¶0331) captured area includes desirable vegetation (e.g., crop being grown) and undesirable vegetation (e.g., weeds); (¶0358) system performs different runs (e.g., a month later); (¶0381, ¶0383) determining if a identified weed appears in subsequent images; (¶0383) a machine learning model can be configured to determine that certain image frames, or sequence of continuous frames, ingested from a plurality of continuous frames ingested in an observation and treatment trial on a geographic boundary do not detect any objects of interest for targeting or from omitting from targeting. For example, a vehicle can pass through a patch of dirt without, in reality, any weeds, plants, or crops for a few meters; (¶0389) use of different ML algorithms/models simultaneously to detect different types of objects (e.g., apples, pears); (¶0391) a library of possible objects detected in a farm may be maintained. New objects (e.g., previously unseen weeds) may be added to this library and used for ML training.
Regarding claim 3, “The system of claim 1, wherein the image-selection parameter comprises a maximum number of the collected images from each camera.” Singh teaches (¶0041) The storage policy may include a rule such as to store image frame(s) from an image stream based on a sample rate (e.g., store n images out of every consecutive m images; store j images every k seconds).
Regarding claim 4, “The system of claim 1, further comprising a spray boom attached to the agricultural vehicle, the one or more cameras mounted on the spray boom.” Sibley teaches (¶0155, ¶0199, and Fig. 6) agricultural vehicle system includes individual image capture devices embedded in each component spray system; (¶0157, ¶0175, Fig. 9A-9B) each modular spray subsystem or component treatment module including a structural mechanism, a compute unit, one or more sensors, one or more treatment units, and one or more illumination devices, can perform VSLAM and receive other non-visual based sensor readings, and continuously generate its own localized pose estimation, the pose being relative to specific objects detected by each of the component treatment modules, which can include agricultural objects including target objects or nearby objects or patterns, shapes, points, or a combination thereof that are of a similar size to that of the target objects; (¶0236, Figs. 19A-19B, 20A-20B) various horizontal/vertical variations.
Regarding claim 6, “The system of claim 1, wherein the image-selection parameter comprises a sampling rate of the subset of the images.” Singh teaches (¶0041) The storage policy may include a rule such as to store image frame(s) from an image stream based on a sample rate (e.g., store n images out of every consecutive m images; store j images every k seconds).
Claim(s) 5, is/are rejected under 35 U.S.C. 103 as being unpatentable over Sibley and Singh in view of Young (US 20210103728).
Regarding claim 5, “The system of claim 1, wherein the one or more microprocessors is/are in network communication … configured to: send a control signal to set the image-selection parameter, and receive the one or more images to store in a cloud storage.” Sibley teaches; (Fig. 1, ¶0084-¶0086) client 141 is a vehicle that communicates with server 150 via network 145, to store a file; (¶0088, ¶0093, ¶0129) for communicating captured data to edge and cloud computing devices; (¶0131) the navigation unit 430 can include a communications module 436 to send and receive signals from other components of the agricultural treatment system 400 such as with the compute unit 420 or to send and receive signals from other computing devices and databases off the vehicle including remote computing devices over the network 520.
Sibley and Singh do not teach “a gateway.” However, Young teaches (¶0004, ¶0033, ¶0046) an agricultural vehicle for field crop inspection/weed analysis; (¶0157, ¶0160) vehicle communicates via a network; (¶0144) network 840 broadly represents one or more wireless networks, internetworking elements such as routers or switches, gateways and hubs, and/or internetworks using WiFi, near-field radio, WiLAN, satellite or other communications technologies. Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the system that communicates via a network to a server as taught by Sibley and Singh to have gateways in the network as taught by Young for the benefit allowing data to flow from one discrete network to another and for communicating using multiple protocols.
Claim(s) 8-12, is/are rejected under 35 U.S.C. 103 as being unpatentable over Sibley and Singh in view of Takahashi et al. (US 20210365731, hereinafter Takahashi.)
Regarding claim 8, Sibley and Singh do not teach “The system of claim 7, wherein the one or more target image properties include a target weather condition.” However, Takahashi teaches (¶0146-¶0147) weather during image capture of each image selected in step S70 is within a preset predetermined range from the time or the weather included in log information 100; (¶0148) the images selected in step S70 include an image that is similar in time period or weather during image capture of the image within a preset predetermined range, this image is preferentially selected; (¶0084-¶0087) images are stored. Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the system as taught by Sibley and Singh with the selection of images based on time or weather as taught by Takahashi for the benefit of improving the machine learning algorithm.
Regarding claim 9, Sibley and Singh do not teach “The system of claim 7, wherein the one or more target image properties include a target time of day.” However, Takahashi teaches (¶0146-¶0147) weather during image capture of each image selected in step S70 is within a preset predetermined range from the time or the weather included in log information 100; (¶0148) the images selected in step S70 include an image that is similar in time period or weather during image capture of the image within a preset predetermined range, this image is preferentially selected; (¶0084-¶0087) images are stored. Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the system as taught by Sibley and Singh with the selection of images based on target time or weather as taught by Takahashi for the benefit of improving the machine learning algorithm.
Regarding claim 10, “The system of claim 7, wherein the one or more target image properties include a target date and/or a target month.” However, Takahashi teaches (¶0141) the images selected in step S70 include an image (with a different date and time) whose image-capturing-location information is similar within a preset predetermined range, this image is preferentially selected; (¶0084-¶0087) images are stored. Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the system as taught by Sibley and Singh with the selection of images based on target date as taught by Takahashi for the benefit of improving the machine learning algorithm.
Regarding claim 11, Sibley and Singh do not teach “The system of claim 7, wherein the one or more target image properties include a target brightness, a target gain, and/or a target contrast.” However, Takahashi teaches (¶0153-¶0154) the images selected in step S70 include an image that is similar in luminance or brightness within a predetermined range, this image is preferentially selected; (¶0084-¶0087) images are stored. Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the system as taught by Sibley and Singh with the selection of images based on target brightness as taught by Takahashi for the benefit of improving the machine learning algorithm.
Regarding claim 12, Sibley and Singh do not teach “The system of claim 7, wherein the one or more target image properties include one or more target ambient conditions” However, Takahashi teaches (¶0153-¶0154) the images selected in step S70 include an image that is similar in luminance or brightness within a predetermined range, this image is preferentially selected; (¶0084-¶0087) images are stored. Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the system as taught by Sibley and Singh with the selection of images based on target brightness as taught by Takahashi for the benefit of improving the machine learning algorithm.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to FRANK J JOHNSON whose telephone number is (571)272-9629. The examiner can normally be reached 9:00AM-5:00PM EST.
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, Brian T. Pendleton can be reached on 571-272-7527. 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.
/Frank Johnson/Primary Examiner, Art Unit 2425