Response to Argument/Amendment
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 .
Applicant’s response to the last Office Action dated 02/05/2026, as well as arguments and amendment to claims, filed 03/19/2026 have been entered and made of record.
Since Applicant has not provided any remarks with respect to the interpretation of claim 1 under 35 U.S.C. 112(f), presented in the Non-Final Office Action dated 02/05/2026, the interpretation of the claims under this section of the rules will be maintained.
Status of Claims
Claims 1-10 are pending, claims 9 and 10 are newly presented.
Response to Arguments
Applicant’s arguments with respect to the rejection of claims 4 and 8 under 35 U.S.C. 103 have been fully considered. Applicant argues that Mao (US 11, 481,862) simply teaches a hybrid loss function and Cai (US 2023/0144209) teaches only one loss function, and therefore the combination teaches away from the claims. Examiner respectfully disagrees with Applicant’s argument. Examiner respectfully submits that the hybrid loss function of Mao includes aspects from object detection and semantic segmentation, and could easily be separated into two different loss functions for each task. This assertion combined with the line segment detection loss function of Cai, would have been obvious. Therefore, Examiner does not find this argument to be persuasive, and maintains the rejections of claims 4 and 8 under 35 U.S.C 103.
Applicant’s arguments presented in Pages 8-9 of its reply with respect to the rejections of claims 1 and 5 under 35 U.S.C. 103 have been fully considered. Applicant’s arguments are merely directed to the amended portion of the claims, and the new analyses presented below render these arguments moot. Applicant’s amendment of independent claims 1 and 5 has altered the scope of the claims, and therefore, has initiated the following new ground(s) of rejection. THIS ACTION IS MADE FINAL.
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-2, 4, 5-6, and 8-10 are rejected under 35 U.S.C. 103 as being unpatentable over Mao et al. (US 11,481,862 B2), in view of Cai et al. (US 2023/0144209), in further view of Du (CN 113076904 A).
Regarding claim 1, Mao teaches “a driving assistance system characterized by utilizing a deep neural network architecture for performing object detection and semantic segmentation recognition, the system comprising: an image capture module configured to capture an image” (Mao, Column 6 Lines 37-38 discloses; “The image capturing device 250 is configured to capture an image of an environment or a scene”), “a processing module configured to execute the deep neural network architecture based on the block images to construct shared feature maps for multiple object detections and semantic segmentations” (Mao, Abstract discloses; “The computer executable code, when executed at the processor, is configured to: receive an image of a scene; process the image using a neural network backbone to obtain a feature map; process the feature map using an object detection module to obtain object detection result of the image; and process the feature map using a semantic segmentation module to obtain semantic segmentation result of the image”), “multiple object bounding boxes” (Mao, Column 9 Lines 33-35 discloses; “The detection results may include bounding boxes in the image and labels of the bounding boxes indicating different objects”), “an output module configured to output information for object detection and semantic segmentation detection” (Mao, Column 6 Lines 61-64 discloses; “ Examples of these hardware and software components may include, but not limited to, other required memory, interfaces, buses, Input/Output (I/O) modules or devices, network interfaces, and peripheral devices”). Mao does not explicitly teach an image segmentation module, multiple block line segment parameters, and filtering and merging. Since Mao does not explicitly disclose these limitations, Examiner relies on the teachings of Cai, in an analogous field of endeavor. Specifically, Cai discloses, “an image segmentation module configured to divide the image into multiple block images” (Cai, Page 1, Para. 0004 discloses; “Specifically, a to-be-detected image is input to a neural network for feature extraction, and then an extracted feature (each feature map is divided into a plurality of grids in advance)”), “and multiple block line segment parameters” (Cai, Page 1, Para. 0004 discloses; “Finally, the line clusters are sorted based on a value of a confidence level of each predicted lane line (which may also be referred to as a confidence level of the grid, where the confidence level reflects whether a lane line passes through the grid and a probability that a lane line passes through the grid, a grid whose confidence level is greater than a preset value is used to predict a lane line, and a grid whose confidence level is less than the preset value is considered as having no contribution to prediction)”) “and further configured to perform filtering and merging based on the block line segment parameters” (Cai, Page 7, Para. 0092 discloses; “These features are combined with Hough transform and Kalman filter to detect a lane line, and has robust detection performance in most cases”, Cai, Page 8, Para. 0100 further discloses; “The feature fusion model 102 is configured to perform feature fusion on the feature maps extracted by the neural network 101 in different layers, to obtain a fused feature map.”
Mao and Cai are both considered to be analogous to the claimed invention because they are in the same field of utilizing neural networks to create feature maps for driving assistance systems. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Mao to incorporate the teachings of Cai in order to segment the acquired image into multiple blocks and perform filtering and merging based on the block line segment parameters. One of ordinary skill in the art would have combined the above-described limitations within the Mao and Cai references by known methods and each element merely performs the same function as it does separately, and the combination would yield predictable results. Combining the object detection and semantic segmentation of Mao with the segmentation and line parameters of Cai into one neural network would yield predictable results based on these known methods in the art. Accordingly, it would have been obvious to combine Mao and Cai to obtain the above specified limitations.
In further regard to claim 1, the combination of Mao in view of Cai does not explicitly teach, “through a non-maximum suppression method for filtering and merging line segments of the multiple block line segment parameters in each of the block images”. Since the combination of Mao and Cai does not explicitly disclose these limitations, Examiner relies on the teachings of Du in an analogous field of endeavor. Specifically, Du discloses, “through a non-maximum suppression method for filtering and merging line segments of the multiple block line segment parameters in each of the block images” (Du discloses; “and performing non-maximum suppression to filter the overlapped serious and short line segments;” It would have been obvious to combine the non-maximum suppression method of Du with the block line segment parameters in the block images of Cai.)
The combination of Mao, Cai, and Du are all considered to be analogous to the claimed invention because they are all in the same field of utilizing deep learning for driving assistance systems. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Mao and Cai to incorporate the teachings of Du in order to filter and merge the candidate line segments into a single line segment detection. One of ordinary skill in the art would have been motivated to combine the previously described system of the combination of Mao and Cai with the teachings of Du to eliminate the number of candidate line segments and get closer to the true line segment detection. Accordingly, it would have been obvious to combine the combination of Mao and Cai with Du to obtain the limitations of claim 1.
Regarding claim 2, the combination of Mao in view of Cai in further view of Du teaches the driving assistance system as claimed in claim 1, “wherein the image capture module comprises at least one RGB camera” (Mao, Column 6 Lines 34-37 discloses; “In certain embodiments, the image capturing device 250 may be a video camera, a gray scale camera, an RGB camera, an RGB depth camera, a depth camera, a LIDAR, or a combination thereof.”)
Regarding claim 4, the combination of Mao in view of Cai in further view of Du teaches the driving assistance system as claimed in claim 1, “wherein the deep neural network architecture includes implementing an algorithm, the algorithm comprising a loss function for line segment detection” (Cai Page 15, Para. 0148 discloses; “A proper loss function is constructed, so that a trained CNN can output the first confidence level that is of the first grid and that meets a requirement in this disclosure.”) “a loss function used in target detection, and a loss function for object bounding box regression” (Mao, Column 5 Lines 60-62 discloses; “A hybrid loss function is defined for both the object detection module 106 and the semantic segmentation module 108”). The proposed combination as well as the motivation for combining the Mao and Cai references presented in the rejection of claim 1, apply to claim 4 and are incorporated herein by reference. Thus, the system recited in claim 4 is met by Mao and Cai.
Claims 5-6, and 8 recite methods with steps corresponding to the elements of the systems recited in claims 1-2, and 4, respectively. Therefore, the recited steps of these claims are mapped to the proposed combination in the same manner as the corresponding elements in their corresponding system claims. Additionally, the rationale and motivation to combine the Mao, Cai, and Du references, presented in rejection of Claim 1, apply to these claims.
Claims 3 and 7 are rejected under 35 U.S.C. 103 as being unpatentable over Mao in view of Cai, in further view of Du, and still in view of Shao et al. (“Semantic Segmentation for Free Space and Lan Based on Grid-Based Interest Point Detection”)
Regarding claim 3, the combination of Mao, Cai, and Du teaches the driving assistance system utilizing a neural network for performing object detection and semantic segmentation recognition with an RGB camera, and the object detection module as disclosed in the rejections of claims 1 and 2. The combination of Mao, Cai, and Du also teaches “wherein the processing module includes an object detection module and a line segment detection module” (Cai, Abstract discloses; “This disclosure discloses lane line detection methods and devices”) The combination of Mao and Cai does not explicitly teach the evaluation methods including Intersection Over Union and TuSimple benchmarks. Since the combination of Mao, Cai, and Du does not explicitly disclose these limitations, Examiner relies on the teachings of Shao, in an analogous field of endeavor. Specifically, Shao discloses, “the object detection module configured to perform an evaluation method for object detection, which includes using the Intersection Over Union (IoU) method for evaluation” (Shao, Section IV. Evaluation Metrics and Experimental Settings discloses; “The most commonly used evaluation metric in semantic segmentation is the mean Intersection over Union, also known as the Jaccard index, which is used for the evaluation of free space. The mean IoU of an image is calculated by the average IoU of free space and background in our experiment”), “the line segment detection module configured to perform an evaluation method for line segment detection, which includes using TuSimple benchmarks for evaluation” (Shao, Section IV. Performance and Comparison discloses; “For the detection of lane, we evaluate the results on the TuSimple dataset”).
Mao, Cai, Du and Shao are all considered to be analogous to the claimed invention because they are all in the same field of utilizing neural networks to obtain feature maps for a driving assistance system. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of the combination of Mao, Cai, and Du to incorporate the teachings of Shao in order to perform an evaluation for object detection and line segment detection. One of ordinary skill in the art would have been motivated to combine the previously described methods of combination of Mao, Cai, and Du with the teachings of Shao to enhance the accuracy and robustness of the object detection system (See Shao, Section IV; “experiments are conducted on different datasets to verify the accuracy and robustness of our method”). Accordingly, it would have been obvious to combine Mao, Cai, Du, and Shao to obtain the invention in claim 3.
Claim 7 recites a method with steps corresponding to the elements of the system recited in claim 3. Therefore, the recited steps of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding system claim. Additionally, the rationale and motivation to combine the Mao, Cai, Du, and Shao references, presented in rejection of Claim 3, apply to this claim.
Claims 9 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Mao in view of Cai, in further view of Du, and still in view of Bao et al. (CN 113920319 A)
Regarding newly added claim 9, the combination of Mao, Cai, and Du do not explicitly teach, “The system of claim 1, wherein the non-maximum suppression method further comprises: sorting the line segments of the multiple block line segment parameters in each of the block images according to their confidences; for each of the sorted line segments in each of the block images, performing following steps beginning with the line segment with the highest confidence as a target line segment: aligning pairs of each one of the line segments and the target line segment according to their lower upper limit and higher lower limit; calculating a normalized cosine similarity value of the pairs of the line segments; and filtering out the pair with the normalized cosine similarity value above a set threshold.” Since the combination of Mao, Cai, and Du does not explicitly disclose these limitations, Examiner relies on the teachings of Bao in an analogous field of endeavor. Specifically, Bao teaches, “The system of claim 1, wherein the non-maximum suppression method further comprises: sorting the line segments of the multiple block line segment parameters in each of the block images according to their confidences;” (Bao discloses; “The maximum coexisting candidate lane line with the highest total confidence level is subset as the target coexisting candidate lane line subset” Examiner interprets this disclosure to show that the candidate lane lines were sorted by confidence level, with the highest being the target.) “for each of the sorted line segments in each of the block images, performing following steps beginning with the line segment with the highest confidence as a target line segment:” (Bao discloses; “The maximum coexisting candidate lane line with the highest total confidence level is subset as the target coexisting candidate lane line subset”) “aligning pairs of each one of the line segments and the target line segment according to their lower upper limit and higher lower limit;” (Bao discloses; “determining a comparison area according to the starting point and the ending point of the two candidate lane lines;” It would be obvious to make one of the candidates the target line.) “calculating a normalized cosine similarity value of the pairs of the line segments;” (Bao discloses; “calculating the parallel similarity and the coexisting relationship between two candidate lane lines;” Examiner interprets the parallel similarity and the cosine similarity to be the same thing since they are both calculating the differences between line segment angles.) “and filtering out the pair with the normalized cosine similarity value above a set threshold.” (Bao discloses; “if the normalized loss value loss is less than the preset higher parallel similarity threshold value that is, the parallel similarity of the two candidate lane line is high; otherwise, if the normalization loss is greater than or equal to the preset high parallel similarity threshold value and less than the preset medium parallel similarity threshold that is, the parallel similarity of the two candidate lane line is medium; otherwise, the parallel similarity of the two candidate lane lines is low, i.e., when the parallel similarity of two candidate lane line is low;” It would have been obvious to filter out the candidates of Bao that are above the set threshold. Bao also discloses; “An embodiment of the present invention provides a method for filtering a lane interference noise line, comprising: obtaining multiple candidate lane line, and calculating the confidence of each candidate lane line; calculating the parallel similarity and the coexisting relationship between two candidate lane lines;” As shown in this Bao disclosure, the entire invention is for filtering for lane line. Thus, it would have been obvious to filter out the candidates of Bao that are above the set threshold.)
The combination of Mao, Cai, Du, and Bao are all considered to be analogous to the claimed invention because they are all in the same field of utilizing driving assistance systems for detecting objects/lane lines. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Mao, Cai, and Du to incorporate the teachings of Bao in order to calculate confidence and similarity values to filter out candidate line detections. One of ordinary skill in the art would have been motivated to combine the previously described system of the combination of Mao, Cai, and Du with the teachings of Bao to ensure the system uses the line detection that is closest to the true lane line orientation. Accordingly, it would have been obvious to combine the combination of Mao, Cai, and Du with Bao to obtain the limitations of claim 9.
Newly added claim 10 recites a method with steps corresponding to the elements of the system recited in claim 9. Therefore, the recited steps of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding system claim. Additionally, the rationale and motivation to combine the Mao, Cai, Du, and Bao references, presented in rejection of Claim 9, apply to this claim.
Conclusion
THIS ACTION IS MADE FINAL. 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 JUSTIN M. OAKES whose telephone number is (571)272-9379. The examiner can normally be reached 7:30am-5pm.
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, Amandeep Saini can be reached at (571) 272-3382. 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.
/JUSTIN M OAKES/Examiner, Art Unit 2662
/Siamak Harandi/Primary Examiner, Art Unit 2662
Prosecution Insights