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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 6/22/2026 has been entered.
Response to Amendment
1. This office action is in response to communications filed 12/3/2025 Claims 1, 11 and 16 are amended. Claims 2-10, 12-15, and 17-20 are original.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1-20 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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.
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.
1. Claim(s) 1, 11 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application 2020/0175401, Shen in view of U.S. Patent 2018/0096487, Nash et al. (hereinafter Nash).
2. Regarding Claim 1, Shen discloses A system (Abstract, system) comprising:
a first camera sensor (Fig. 1: 114 camera(s), the first camera sensor capturing in front of a vehicle ([0010], “image sensors (e.g., cameras) may be positioned about a vehicle”);
a second camera sensor (Fig. 1: 114 camera(s));
a controller ([0035], “processing hardware's computation resources are limited (e.g., microcontrollers”), the controller including:
a processing unit ([0011] “The machine learning models may be implemented via a system of one or more processors”),
a plurality of machine learning models ([0011], “two or more machine learning models may be used to analyze images”), wherein a first machine learning model in the plurality of machine learning models is configured to receive a first image frame from the first camera sensor (Fig. 2: 202 [0011-[0012], “machine learning models may be used to analyze images obtained from image sensors positioned about a vehicle”), and detect a first detection of a first object within the first image frame ([0014], “The first machine learning model may analyze all, or a substantial portion, of the images from the image sensors”), and wherein a second machine learning model in the plurality of machine learning models is configured to receive a second image frame from the second camera sensor ([0012], “the second machine learning model may be a comparatively slower machine learning model capable of analyzing a subset of the obtained images”) and detect a second detection of a second object within the second image frame ([0013], “A detector, as an example, may be used to detect an object (e.g., classify an object, determine location information, and so on”)), and
a machine learning pipeline configured to receive the first detection and the second detection (Fig. 1; [0023]-[0025], “the image processing network 102 [i.e. machine learning pipeline] functions to facilitate communication between various components of the system (e.g., between the image processing engines, between the image processing engines and the endpoint, etc.” [0013], “the first machine learning model and the second machine learning model may be detectors. A detector, as an example, may be used to detect an object”) and transmit an instruction to the processing unit based on the first detection and second detection ([0031], “Client device(s) 112 are devices that send [i.e. transmit] information to the image processing network 102, receive information from the image processing network 102, or both”).
However, Shen may not explicitly disclose the first camera sensor capturing a wide-field of view;
the second camera sensor capturing a near-field of view in front of the vehicle, the near-field of view being a subset of the wide-field of view, wherein the first camera sensor and the second camera sensor are configured to simultaneously record images in front of the vehicle
Nash teaches the first camera sensor capturing a wide-field of view (Fig. 2: 202; [0046], “the image sensor(s) 104 may capture the one or more images (e.g., wide-angle image(s))… the electronic device 102 may include two lenses (e.g., a wide-angle lens)”);
the second camera sensor capturing a near-field of view in front of the vehicle ([0053], “the telephoto lens may have a narrower FOV (e.g., a lesser angular range) than the wide-angle lens.” [0040], “the electronic device 102 include cameras.” [0040], “electronic device 102 in which systems and methods for fusing images may be implemented…unmanned aerial vehicles” ), the near-field of view being a subset of the wide-field of view (Abstract, “obtaining a second image from a second camera, the second camera having a second focal length and a second field of view disposed within the first field of view” see also claim 1 and 20. Identifies for pairing a wide-angle sensor with a telephoto sensor whose FOV is nested within it.), wherein the first camera sensor and the second camera sensor are configured to simultaneously record images in front of the vehicle ([0052], “a wide-angle image and a telephoto image may be captured concurrently.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the multi-sensor, model-based architecture as taught in Shen to use the wide-angle and narrow-angle cameras as taught in Nash so that each camera provides complementary field-of-view image data to be processed by the respective machine learning models as taught by Shen. The motivation would have been to improve object detection accuracy and robustness in forward-scene monitoring by leveraging both wide-filed situational coverage and narrow-field detail capture simultaneously, a design choice that is predictable and consistent with well-established teachings in the vehicle perception arts (see MPEP 2143, KSR Int’l v. Teleflex, 550 U.S. 398 (2007)).
3. Claim 11 is a method claim, rejected with respect to the same limitation rejected in the system claim 1.
4. Claim 16 is a non-transitory CRM rejected with respect to the same limitations rejected in System Claim 1.
5. Claim(s) 2, 3, 12 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Nash as applied to claim 1, 11 and 16 above, and further in view of U.S. Patent 11,532,170 Alakarhu et al. (hereinafter Alakarhu).
6. Regarding Claim 2, Shen in view of Nash discloses The system of claim 1,
However, Shen in view of Nash does not explicitly disclose wherein the first detection of the first object comprises a detection of a license plate.
Further, Alakarhu teaches wherein the first detection of the first object comprises a detection of a license plate (Abstract, “capturing license plate (LP) information of a vehicle in relative motion to a camera device”).
It would have been obvious to one of ordinary skill in the art to modify the system of Shen which uses multiple sensors and corresponding ML models, by employing the wide-angle and narrow -angle cameras of Nash as the sensors, and further incorporating the plate detection and alphanumeric recognition techniques as taught in Alakrhu into the models executed on those sensors. The motivation would have been to provide a more robust vehicle monitoring and enforcement system, wherein the wide-angle camera provides context and vehicle localization, the narrow-angle camera provides detailed plate imagery, and the ML pipeline performs plate detection and OCR to identify the vehicle. Such a combination would have been a predictable use of prior art elements according to their established functions (wide/narrow FOV capture, multi-model ML processing, and plate/OCR recognition) to yield improved accuracy in license plate recognition. See MPEP 2143, KSR Int’l V. Teleflex, 550 U.S. 398 (2007).
7. Regarding Claim 3, Shen in view of Nash further in view of Alakarhu discloses The system of claim 2,
Alakarhu discloses wherein the second detection of the second object comprises an identification of alphanumeric characters on the license plate (Figs. 8 and 10, Col. 3 lines 21-22 “The LPR system may further include a controller communicatively connected to the camera device” Col. 3 lines 54-59, “The LPR system may result, in some examples, where characters of the first license plate have a greater probability of being recognized by a computerized optical character recognition (OCR) platform in the first long-exposure image than in the first short-exposure image”).
It would have been obvious to one of ordinary skill in the art to modify the system of Shen which uses multiple sensors and corresponding ML models, by employing the wide-angle and narrow -angle cameras of Nash as the sensors, and further incorporating the plate detection and alphanumeric recognition techniques as taught in Alakrhu into the models executed on those sensors. The motivation would have been to provide a more robust vehicle monitoring and enforcement system, wherein the wide-angle camera provides context and vehicle localization, the narrow-angle camera provides detailed plate imagery, and the ML pipeline performs plate detection and OCR to identify the vehicle. Such a combination would have been a predictable use of prior art elements according to their established functions (wide/narrow FOV capture, multi-model ML processing, and plate/OCR recognition) to yield improved accuracy in license plate recognition. See MPEP 2143, KSR Int’l V. Teleflex, 550 U.S. 398 (2007).
8. Claim 12 is a method claim, rejected with respect to the same limitation rejected in the system claims 2 and 3.
9. Claim 17 is a non-transitory CRM rejected with respect to the same limitations rejected in System Claims 2 and 3.
10. Claim(s) 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Meyer as applied to claim 1 above, and further in view of U.S. Patent Application 2022/0076102 Abraham et al. (hereinafter Abraham).
11. Regarding Claim 4, Shen in view of Nash discloses The system of claim 1,
However, Shen in view of Nash does not explicitly disclose wherein the first machine learning model and the second machine learning model are executed independently.
Abraham teaches wherein the first machine learning model and the second machine learning model are executed independently ([0021], “If the DNN models are executed independently of one another, then there is no dependency between the DNN models”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine teachings by employing the wide-angle and narrow-angle cameras as taught in Nash as the sensors in the system as taught in Shen, and further implementing the independent execution scheme as taught in Abraham’s application for the associated machine learning models. The motivation would have been to improve robustness and efficiency of vehicle perception: wide-angle and narrow angle cameras provide complementary field of view for scene analysis (as taught by Nash), while independent execution of the respective ML models (as taught by Abraham) reduces processing bottlenecks and allows each model to operate at its optimal frequency, thereby yielding predictable benefits in detection accuracy and system responsiveness. Such a combination represents the routine integration of known elements to achieve improved performance in multi-camera machine learning systems. See KSR Int’l V. Teleflex, 550 U.S. 398 (2007); MPEP 2143.
12. Claim(s) 5 and 6 are rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Nash as applied to claim 1 above, and further in view of U.S. Patent Application 2019/0251369, Popov et al. (hereinafter Popov).
13. Regarding Claim 5, Shen in view of Nash discloses The system of claim 1,
However, Shen in view of Nash do not explicitly disclose wherein the machine learning pipeline is configured to execute the second machine learning model after detecting the first detection.
Popov teaches wherein the machine learning pipeline is configured to execute the second machine learning model after detecting the first detection (Figs. 2-3, [0028]-[0030], a license plate detection module that first detects plate regions in an image, and after detection, a license plate recognition module (second ML model) is executed on those detected regions to identify alphanumeric characters).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to combine the multi-camera, multi-model architecture of Shen with the wide-angle/narrow-angle simultaneous capture as taught in Nash and the sequential pipeline execution as taught in Popov. The motivation would have been to improve forward-scene monitoring accuracy and robustness by (i) using wide and narrow field-of-view cameras to obtain complementary imagery as taught in Nash, (ii) assigning separate machine learning models to analyze the respective camera streams as taught in Shen, (and iii) structuring those models into a sequential pipeline so that detection by a first model (e.g. plate detection) triggers execution of a second model (e.g., character recognition) as taught in Popov. Such combination represents the predictable integration of known elements according to their established functions-multi-camera inputs, ML-based detection, and sequential detection-then-recognition pipelines- to yield improved vehicle perception and license plate recognition. See KSR Int’l Teleflex, 550 U.S. 398 (2007); MPEP 2143.
14. Regarding Claim 6, Shen in view of Nash further in view of Popov discloses The system of claim 5,
Popov discloses wherein the machine learning pipeline (Figs 2-3 illustrate the sequential ML pipeline) is configured to provide location information ([0028], describes that the license plate detection module first processes the captured image and determines the location of a license plate region (bounding box/ region of interest, ROI).detection module that outputs regions of interest (ROI) (essentially location information) to the second machine learning model ([0029]-[0030], the recognition module receives as input the detected license plate region from the detection module) when providing the second image frame (Figs. 2-3 illustrate that the recognition stage operates on the cropped plate image (a sub-frame/region derived from the first detection). This cropped region is effectively the “second frame” generated from the first detection) to the second machine learning model ([0031], The recognition module performs OCR on the provided plate region, i.e., the second ML model executes after receiving location information from the first).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to combine the multi-camera, multi-model architecture of Shen with the wide-angle/narrow-angle simultaneous capture as taught in Nash and the sequential pipeline execution as taught in Popov. The motivation would have been to improve forward-scene monitoring accuracy and robustness by (i) using wide and narrow field-of-view cameras to obtain complementary imagery as taught in Nash, (ii) assigning separate machine learning models to analyze the respective camera streams as taught in Shen, (and iii) structuring those models into a sequential pipeline so that detection by a first model (e.g. plate detection) triggers execution of a second model (e.g., character recognition) as taught in Popov. Such combination represents the predictable integration of known elements according to their established functions-multi-camera inputs, ML-based detection, and sequential detection-then-recognition pipelines- to yield improved vehicle perception and license plate recognition. See KSR Int’l Teleflex, 550 U.S. 398 (2007); MPEP 2143.
15. Claim(s) 7, 15 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Nash as applied to claim 1, 11 and 16 above, and further in view of U.S. Patent 12,417,636 Ramanathan et al. (hereinafter Ramanathan).
16. Regarding Claim 7, Shen in view of Nash discloses The system of claim 1,
However, Shen in view of Nash does not explicitly disclose wherein the machine learning pipeline is configured to input the first detection and second detection into a third machine learning model to detect an event.
Ramanathan teaches wherein the machine learning pipeline (Fig. 2: image analysis engine 206 with multiple ML engines (object detection 222, activity classification 224, event engine 226) is configured to input the first detection and second detection (Col. 4 lines 56-58, “an output of one or more of the engines 222-226 can be fed as an input to one or more of the engines 222-226”- detection results (first and second ) are used as inputs downstream) into a third machine learning model (i.e., event engine 226) to detect an event (Col. 6 Lines 38-39, “the image analysis engine 106 can use the third model, e.g., the event engine 226, to determine the event”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the teachings of Shen in view of Nash with Ramanathan. The motivation would have been to improve overall vehicle perception and event recognition accuracy by (i) capturing complementary fields of view for robustness, (ii) applying independent ML models to teach stream for specialized detection, and (iii) feeding both detections into a third ML model to infer higher-level events, thereby yielding predictable benefits in safety automation. The combination represents routine integration of know elements according to their established functions-multi-camera perception (Shen, Nash) and hierarchical ML pipelines with event detection (Ramanathan) to achieve improved system functionality. See KSR Int’l v. Teleflex, 550 U.S. 398 (2007); MPEP 2143.
17. Claim 15 is a method claim, rejected with respect to the same limitations rejected in System Claim 7.
18. Claim 20 is a non-transitory CRM claim, rejected with respect to the same limitations rejected in System Claim 7.
19. Claim(s) 8 is rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Nash as applied to claim 1 above, and further in view of U.S. Patent 9,152,853 El Dokor.
20. Regarding Claim 8, Shen in view of Nash discloses The system of claim 1,
However, Shen in view of Nash does not explicitly disclose further comprising an inward-facing camera communicatively coupled to the controller, wherein the controller is configured to execute an inward machine learning model, the inward machine learning model configured to receive third image frames from the inward-facing camera and detect a gesture occurring within the third image frames, wherein the controller is configured to transmit a second instruction to the processing unit based on a type of the gesture.
El Dokor teaches an inward-facing camera communicatively coupled to the controller (in vehicle camera sensor (e.g., TOF or RGB camera ) mounted inside the vehicle cabin to capture occupant hand/body movements), see abstract, wherein the controller is configured to execute an inward machine learning model (a processor/controller executes gesture recognition algorithms (machine learning classifiers) to analyze the occupant frames, see Col. 4 lines 1-3, “Gesture recognition 230, Interface implementation 240, and Algorithm parameter adjustment/learning 250”), the inward machine learning model configured to receive third image frames from the inward-facing camera (abstract, “receiving one or more raw frames from the TOF sensor, performing clustering to locate one or more body part clusters of the vehicle occupant”) and detect a gesture occurring within the third image frames, wherein the controller is configured to transmit a second instruction to the processing unit based on a type of the gesture (abstract, “performing clustering to locate one or more body part clusters of the vehicle occupant, calculating the location of the tip of the hand of the vehicle occupant, determining whether the hand has performed a dynamic or a static gesture, retrieving a command corresponding to one of the determined static or dynamic gestures, and executing the command”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the outward multi-camera, multi-model system of (Shen/Nash) references with the inward gesture-recognition system as taught in El Dokor. The motivation would have been to enhance vehicle perception and human-machine interaction by (i) using wide-narrow cameras for robust external scene understanding (Shen/Nash), and (ii) incorporating an inward camera and gesture model (El Dokor) so that the driver or passenger could issue control instructions via gestures detected in cabin frames. Such integration of inward and outward perception pipelines represents a predictable use of known elements according to their established functions: outward cameras for environment monitoring and inward cameras for driver input. The result would improve system usability and safety by allowing gesture-based controls without requiring manual contact, consistent with the reasoning in KSR Int’l v. Teleflex, 550 U.S. 398 (2007) and MPEP 2143.
21. Claim(s) 9 is rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Nash as applied to claim 1 above, and further in view of U.S. Patent 8,214,219 Prieto et al. (hereinafter Prieto).
22. Regarding Claim 9, Shen in view of Nash discloses The system of claim 1,
However, Shen in view of Nash does not explicitly disclose further comprising a microphone configured to record audio samples within the vehicle, wherein the controller is configured to execute voice model to reduce noise within the audio samples and detect a spoken command within the audio samples, wherein the controller is configured to transmit a third instruction to the processing unit based on the spoken command.
Further Prieto teaches a microphone configured to record audio samples within the vehicle (abstract, “a microphone system provided in the vehicle interior in order to detect audio information”), wherein the controller is configured to execute voice model to reduce noise within the audio samples (abstract, “An acoustic echo canceller eliminates portions of the audio information detected by the microphone system”) and detect a spoken command within the audio samples (Col. 8 lines48-49, “speech recognizer 4 to better detect the presence of voice audio information coming from the user 3”), wherein the controller is configured to transmit a third instruction to the processing unit based on the spoken command (Abstract, “An interaction manager provides grammar information to a speech recognizer. The speech recognizer provides speech recognition results to the interaction manager.” The recognized speech command is output to an interaction manager/vehicle control system which executes the corresponding instruction).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the outward multi-camera machine-learning system as taught in Shen as modified by the wide/narrow-filed arrangement as taught in Nash, with the in-cabin spoken command system as taught in Prieto. The motivation would have been to enhance overall vehicle perception and user interaction by integrating external scene understating (multi-camera ML detection) with in-vehicle voice command input, thereby allowing the occupant to issue spoken instructions while the system simultaneously perceives the environment. Such an integration represents a predictable use of known elements (outward cameras + ML perception, wide/narrow FOV arrangement, and in-cabin microphone with voice model) according to their established functions to improve usability, safety, and control of vehicle systems. See KSR Int’l v. Teleflex, 550 U.S. 398 (2007); MPEP 2143.
23. Claim(s) 10 is rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Nash as applied to claim 1 above, and further in view of U.S. Patent 10,757,271, Gonzalez et al. (hereinafter Gonzalez).
24. Regarding Claim 10, Shen in view of Nash discloses The system of claim 1,
However, Shen in view of Nash does not explicitly disclose further comprising a wireless network interface, wherein the controller is configured to establish a mesh network with at least one other dashcam using the wireless network interface.
Gonzalez teaches a wireless network interface, wherein the controller is configured to establish a mesh network with at least one other dashcam using the wireless network interface (Abstract, “A mesh network adapter is designed to communicate with a camera”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the multi-camera, model-based vehicle system as taught in (Shen in view of Nash) to further include the wireless network interface taught in Gonzalez, enabling the controller to establish a mesh network with one or more dashcams. The motivation would have been to improve scalability and reliability of video data collection and sharing among vehicle cameras, leveraging mesh networking to extend coverage, provide redundancy, and reduce single points of failure. This represents the predictable integration of known technologies-multi-camera ML perception (Shen, and Nash) and camera mesh networking (Gonzalez) -to enhance overall system robustness and functionality. See KSR Int’l v. Teleflex, 550 U.S. 398 (2007); MPEP 2143.
25. Claim(s) 13, 14, 18 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Nash as applied to claim 11 and 16 above, further in view of U.S. Patent Application 2022/0076102 Abraham et al. (hereinafter Abraham) and further in view of U.S. Patent Application 2019/0251369, Popov et al. (hereinafter Popov).
26. Regarding Claim 13, Shen in view of Nash discloses The method of claim 11,
However, Shen in view of Nash does not explicitly disclose further comprising executing the first machine learning model and the second machine learning model independently.
Abraham teaches further comprising executing the first machine learning model and the second machine learning model independently ([0021], “If the DNN models are executed independently of one another, then there is no dependency between the DNN models”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine teachings by employing the wide-angle and narrow-angle facing cameras as taught in Nash as the sensors in the system as taught in Shen, and further implementing the independent execution scheme as taught in Abraham’s application for the associated machine learning models. The motivation would have been to improve robustness and efficiency of vehicle perception: wide-angle and narrow angle cameras provide complementary field of view for scene analysis (as taught by Nash), while independent execution of the respective ML models (as taught by Abraham) reduces processing bottlenecks and allows each model to operate at its optimal frequency, thereby yielding predictable benefits in detection accuracy and system responsiveness. Such a combination represents the routine integration of known elements to achieve improved performance in multi-camera machine learning systems. See KSR Int’l V. Teleflex, 550 U.S. 398 (2007); MPEP 2143.
However, Popov does not explicitly disclose executing, using the machine learning pipeline, the second machine learning model after detecting the first object within the first image frame.
Further, Popov teaches executing, using the machine learning pipeline, the second machine learning model after detecting the first object within the first image frame (Figs. 2-3, [0028]-[0030], a license plate detection module that first detects plate regions in an image, and after detection, a license plate recognition module (second ML model) is executed on those detected regions to identify alphanumeric characters).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to combine the multi-camera, multi-model architecture of Shen with the wide-angle/narrow-angle simultaneous capture as taught in Nash and the sequential pipeline execution as taught in Popov. The motivation would have been to improve forward-scene monitoring accuracy and robustness by (i) using wide and narrow field-of-view cameras to obtain complementary imagery as taught in Nash, (ii) assigning separate machine learning models to analyze the respective camera streams as taught in Shen, (and iii) structuring those models into a sequential pipeline so that detection by a first model (e.g. plate detection) triggers execution of a second model (e.g., character recognition) as taught in Popov. Such combination represents the predictable integration of known elements according to their established functions-multi-camera inputs, ML-based detection, and sequential detection-then-recognition pipelines- to yield improved vehicle perception and license plate recognition. See KSR Int’l Teleflex, 550 U.S. 398 (2007); MPEP 2143.
27. Regarding Claim 14, Shen in view of Nash further in view of Abraham further in view of Popov discloses The method of claim 13,
Popov discloses further comprising providing, using the machine learning pipeline(Figs 2-3 illustrate the sequential ML pipeline), location information of the first object ([0028], describes that the license plate detection module first processes the captured image and determines the location of a license plate region (bounding box/ region of interest, ROI).detection module that outputs regions of interest (ROI) (essentially location information) to the second machine learning model ([0029]-[0030], the recognition module receives as input the detected license plate region from the detection module) when providing the second image frame (Figs. 2-3 illustrate that the recognition stage operates on the cropped plate image (a sub-frame/region derived from the first detection). This cropped region is effectively the “second frame” generated from the first detection) to the second machine learning model ([0031], The recognition module performs OCR on the provided plate region, i.e., the second ML model executes after receiving location information from the first).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to combine the multi-camera, multi-model architecture of Shen with the wide-angle/narrow-angle simultaneous capture as taught in Nash and the sequential pipeline execution as taught in Popov. The motivation would have been to improve forward-scene monitoring accuracy and robustness by (i) using wide and narrow field-of-view cameras to obtain complementary imagery as taught in Meyer, (ii) assigning separate machine learning models to analyze the respective camera streams as taught in Shen, (and iii) structuring those models into a sequential pipeline so that detection by a first model (e.g. plate detection) triggers execution of a second model (e.g., character recognition) as taught in Popov. Such combination represents the predictable integration of known elements according to their established functions-multi-camera inputs, ML-based detection, and sequential detection-then-recognition pipelines- to yield improved vehicle perception and license plate recognition. See KSR Int’l Teleflex, 550 U.S. 398 (2007); MPEP 2143.
28. Claim 18 is a non-transitory CRM rejected with respect to the same limitations rejected in method Claim 13.
29. Claim 19 is a non-transitory CRM rejected with respect to the same limitations rejected in method Claim 14.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to OMER KHALID whose telephone number is (571)270-5997. The examiner can normally be reached Monday- Friday 9am-7pm.
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, John Miller can be reached at (571) 272-7353. 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.
/OMER KHALID/Examiner, Art Unit 2422
/JOHN W MILLER/Supervisory Patent Examiner, Art Unit 2422
Prosecution Insights