DETAILED ACTION
Claims 1-12 and 14-21 are currently pending.
Response to Arguments
Applicant's arguments filed 8/7/26 have been fully considered but they are not persuasive.
The Applicant argues on page 8 of the response in essence that: Thus, Scime's input layer does not teach or suggest the claimed providing of the first and second features to a trained model, especially where the first feature "indicates a characteristic of ejecta projected from the volume of molten material generated by the additive manufacturing process of the part" (claim 1) and the second feature "indicates a temperature of the volume of molten material, the first feature and the second feature being different than the image" (id.).
Scime discloses that the powder bed build layer frame undergoes some pre-processing 312 before being input into the trained DSNN. The pre-processing generally includes calibration 314 and converting the calibrated powder bed build layer frame into the proper format as an input layer for the DSCNN 318 (paragraph 73). The DSCNN uses the image data to generate feature vectors that classify anomalies including ejecta (paragraph 28). Newly cited reference Kitchen discloses extracting, from the image (ii) a second feature that indicates a temperature of the volume of molten material (paragraph 86, Some implementations track temperature intensity, melt points, temperature and cooling profile, and/or weld beam movement. Some implementations detect features from NIR, high speed IR, and/or optical cameras).
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.
Claims 1, 5, 7-9, 11, 12, 16 and 18-21 are rejected under 35 U.S.C. 103 as being unpatentable over Scime et al. US Publication 2022/0134435 (hereafter “Scime”) and Kitchen et al. US Publication 2021/0318673 (hereafter “Kitchen”).
Referring to claims 1, 12, and 20, Scime discloses a method comprising:
obtaining, using a camera, an image representing a volume of molten material generated by an additive manufacturing process of a part (paragraph 65, The imaging system 250 can include one or more imaging devices (e.g., cameras or imaging sensors). Each imaging device can capture images of the powder bed during manufacture and provide them to the computer 202 for use in connection with the powder bed anomaly detection software), wherein the volume of molten material is distinguished from the part as a whole by reaction with a laser (paragraph 54, Laser Powder Bed Fusion (“L-PBF”) is where one or more focused laser beams are used to selectively melt powder);
extracting, from the image, (i) a first feature that indicates a characteristic of ejecta projected from the volume of molten material generated by the additive manufacturing process of the part and (ii) a second feature (paragraph 28, The software platform can monitor essentially any powder bed additive manufacturing process and provide segmentation and classification of anomalies it is trained to detect. A few examples of anomalies capable of being detected by some embodiments include recoating issues, part distortion, melt ejecta, printer malfunctions, and porosity), the first feature and the second feature being different than the image (paragraph 27, In addition, the output 614 of the DSCNN (i.e., pixel-wise feature vectors) can be fed into the DSCNN-Perceptron 228 to label the entire build layer);
providing the first feature and the second feature to a trained machine learning model (paragraph 73, The powder bed build layer frame undergoes some pre-processing 312 before being input into the trained DSNN. The pre-processing generally includes calibration 314 and converting the calibrated powder bed build layer frame into the proper format as an input layer for the DSCNN 318);
obtaining output from the trained machine learning model processing the first feature and the second feature (paragraph 105, The results of the classification can be returned to the software platform for presentation to the user, e.g., in raw form, in a visualized form, or in some other form 348);
determining, using the output of the trained machine learning model processing the first feature and the second feature (paragraph 58, In general, embodiments of the present disclosure utilize a dynamic segmentation convolutional neural network (“DSCNN”) that incorporates contextual information at multiple size scales in order to make semantically-relevant anomaly classifications), a predicted porosity of the part (paragraph 28, The software platform can monitor essentially any powder bed additive manufacturing process and provide segmentation and classification of anomalies it is trained to detect. A few examples of anomalies capable of being detected by some embodiments include recoating issues, part distortion, melt ejecta, printer malfunctions, and porosity) (paragraph 137, Certain embodiments provide anomaly classifications (i.e., model predictions)); and
adjusting, based on the predicted porosity of the part, manufacturing of the part to mitigate the predicted porosity of the part (paragraph 34, During a live analysis, the interface can execute customizable macros that can interface with a printer's user interface to effect process interventions).
While Scime discloses extracting features from the images and extracting a temperature of the volume of molten material, Scime does not disclose expressly extracting, from the image (ii) a second feature that indicates a temperature of the volume of molten material.
Kitchen discloses extracting, from the image (ii) a second feature that indicates a temperature of the volume of molten material (paragraph 86, Some implementations track temperature intensity, melt points, temperature and cooling profile, and/or weld beam movement. Some implementations detect features from NIR, high speed IR, and/or optical cameras).
Before the effective filing date of the claimed invention, it would have obvious to a person of ordinary skill in the art to extract temperature from an image using a camera. The motivation for doing so would have been to reduce the necessity for the sensor to be physically close to the source in order to prevent interference with the manufacture and prevent damage to the sensor. Therefore, it would have been obvious to combine Kitchen with Scime to obtain the invention as specified in claims 1, 12, and 20.
Referring to claims 5 and 16, Scime discloses providing feedback to the additive manufacturing process, wherein the feedback is configured to adjust the manufacturing of the part (paragraph 107, The post processing can also include layer-wise flagging 352 where certain layers are flagged for the user to identify a particular feature or set of features. For example, if a particular layer has a significant number of anomalies or a significantly large anomaly, that layer can be flagged for user review).
Referring to claims 7 and 18, Scime discloses wherein determining the predicted porosity of the part comprises determining that continuing the additive manufacturing process will cause a defect in the part without adjustments to the additive manufacturing process (paragraph 34, The DSCNN-P can be trained to flag or label layers to focus the attention of an operator or to execute a process intervention).
Referring to claims 8 and 19, Scime discloses wherein at least one of the first feature or the second feature indicates one or more features include one or more of a value indicating a length of the volume, a value indicating a spread of ejecta, a value indicating a temperature of ejecta, or a value indicating a temperature of the volume (paragraph 42, The visualizations can illustrate not only image data and classification data, but also other characteristics and information associated with the additive manufacturing process such as recoater traverse speed, measured bed temperature, build height).
Referring to claim 9, Scime discloses wherein at least one of the first feature or the second feature indicates a shape of the volume of molten material (paragraph 42, The visualizations can illustrate not only image data and classification data, but also other characteristics and information associated with the additive manufacturing process such as recoater traverse speed, measured bed temperature, build height).
Referring to claim 11, Scime discloses providing an indication that a second part has no defect (paragraph 27, The labels assigned by the DSCNN-P can be combined with global heuristics 712 and presented to the user as powder bed layer flags 714).
Referring to claim 21, Scime discloses extracting, from the image, (i) the first feature that indicates a characteristic of ejecta projected from the volume of molten material generated by the additive manufacturing process and (ii) the second feature, and (iii) a third feature that indicates a shape of the volume of molten material (paragraph 28, The software platform can monitor essentially any powder bed additive manufacturing process and provide segmentation and classification of anomalies it is trained to detect. A few examples of anomalies capable of being detected by some embodiments include recoating issues, part distortion, melt ejecta, printer malfunctions, and porosity) (paragraph 42, The visualizations can illustrate not only image data and classification data, but also other characteristics and information associated with the additive manufacturing process such as recoater traverse speed, measured bed temperature, build height, and layer times to name a few), the first, second and third feature being different than the image (paragraph 27, In addition, the output 614 of the DSCNN (i.e., pixel-wise feature vectors) can be fed into the DSCNN-Perceptron 228 to label the entire build layer);
providing the first feature, the second feature, and the third feature to a trained machine learning model to generate the output used to determine the predicted porosity of the part (paragraph 129, The additive manufacturing apparatus 5 acquires monitoring information during additive manufacturing while additively manufacturing the modeled object by using this recipe (step S33).
Kitchen discloses extracting, from the image (ii) the second feature that indicates a temperature of the volume of molten material (paragraph 86, Some implementations track temperature intensity, melt points, temperature and cooling profile, and/or weld beam movement. Some implementations detect features from NIR, high speed IR, and/or optical cameras).
Claims 2-4, 6, 10, 14, 15 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Scime et al. US Publication 2022/0134435 and Kitchen et al. US Publication 2021/0318673 as applied to claims 1, 5, 12 and 16 above, and further in view of Kawanaka et al. US Publication 2025/0050586 (hereafter “Kawanaka”).
Referring to claim 2, Scime discloses determining the predicted porosity of the part comprises using the output of the trained machine learning model, but does not disclose expressly determining, for a future point in time, a severity of porosity of the part.
Kawanaka discloses determining, for a future point in time, a severity of porosity of the part
wherein determining the predicted porosity of the part comprises using the output of the trained machine learning model comprises:
determining, for a future point in time, a severity of porosity of the part (paragraph 139, The additive manufacturing condition search device 2 causes the generation unit 42 to generate the prediction model for predicting the input parameter that is the solution that satisfies the target value of the modeling result of the standard sample 1 (step S13). Specifically, the additive manufacturing condition search device 2 generates, as the prediction model, a function indicating a relationship between pieces of input and output data of the additive manufacturing apparatus 5 by using data (for example, initial data) stored in the storage unit 22).
Before the effective filing date of the claimed invention, it would have obvious to a person of ordinary skill in the art to determine a severity of porosity of the part for a future point in time. The motivation for doing so would have been to reduce defects in the manufacture. Therefore, it would have been obvious to combine Kawanaka with Scime to obtain the invention as specified in claim 2.
Referring to claims 3 and 14, Scime discloses determining the predicted porosity of the part comprises using the output of the trained machine learning model, but does not disclose expressly determining, for a future point in time, a type of defect.
Kawanaka discloses wherein determining the predicted porosity of the part comprises using the output of the trained machine learning model comprises:
determining, for a future point in time, a type of defect (paragraph 124, The data processing unit 67 performs machine learning by using, as train data, past monitoring information of the monitoring information database 661 and a past defect determination result of the defect determination result database 662, and creates a model that predicts the defect determination result in a case where the monitoring information is used as an input).
Before the effective filing date of the claimed invention, it would have obvious to a person of ordinary skill in the art to determine a type of defect of the part for a future point in time. The motivation for doing so would have been to reduce defects in the manufacture. Therefore, it would have been obvious to combine Kawanaka with Scime to obtain the invention as specified in claims 3 and 14.
Referring to claims 4 and 15, Kawanaka discloses wherein the type of defect is lack of fusion, conduction, or keyholing (paragraph 67, When powder is irradiated with a beam, in a case where there is no material to be bonded to the melted powder and in a case where heat cannot be dissipated quickly due to heat conduction, the melted powder shrinks into a spherical shape to form a relatively large spherical lump on the powder. Thus, in the “Down-skin” that forms the overhang, a condition for suppressing energy is selected).
Referring to claims 6 and 17, Scime discloses adjusting the manufacturing of the part, but does not disclose expressly adjusting one or more of a laser power, scanning speed, or delay.
Kawanaka discloses wherein adjusting the manufacturing of the part comprises adjusting one or more of a laser power, scanning speed, or delay (paragraph 72, Here, the control factors an output and a scanning speed of a light beam with which the contour line is irradiated).
Before the effective filing date of the claimed invention, it would have obvious to a person of ordinary skill in the art to adjust one or more of a laser power, scanning speed, or delay. The motivation for doing so would have been to reduce defects in the manufacture. Therefore, it would have been obvious to combine Kawanaka with Scime to obtain the invention as specified in claims 6 and 17.
Referring to claim 10, Scime discloses the trained machine learning mode, but does not disclose expressly wherein the trained machine learning model includes one or more of a Logistic Regression (LR) model, a Support Vector Machine (SVM), or K-Nearest Neighbors (KNN) algorithm.
Kawanaka discloses wherein the trained machine learning model includes one or more of a Logistic Regression (LR) model, a Support Vector Machine (SVM), or K-Nearest Neighbors (KNN) algorithm (paragraph 105, The generation unit 42 generates a prediction model indicating a relationship between the setting value of the condition within the search region and the actual measurement value of the output by statistical analysis such as regression analysis capable of coping with multiple-input and multiple-output such as a neural network and a support vector machine, correlation analysis, principal component analysis, or multiple regression analysis).
Before the effective filing date of the claimed invention, it would have obvious to a person of ordinary skill in the art for a machine learning model to include a support vector machine. The motivation for doing so would have been to utilize a machine learning model that is efficient at learning while reducing memory usage. Therefore, it would have been obvious to combine Kawanaka with Scime to obtain the invention as specified in claim 10.
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 PETER K HUNTSINGER whose telephone number is (571)272-7435. The examiner can normally be reached Monday - Friday 8:30 - 5:00.
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, Benny Q Tieu can be reached at 571-272-7490. 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.
/PETER K HUNTSINGER/Primary Examiner, Art Unit 2682
Prosecution Insights