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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on October 22, 2025 in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement has been considered by the examiner
Status of Claims
Claims 1, 2, 5 – 18 and 21 – 24 remain pending.
Claims 1, 5, 8, 9, 10, 17 and 18 are Amended
Claims 3, 4, 19 and 20 are canceled
Claims 21 – 24 are new claims
Response to Arguments
Applicant's arguments filed October 22, 2025 with respect to claims 1, 2, 5 – 18 and 21 – 24 have been fully considered but they are not persuasive.
Applicant’s arguments regarding claim 9 – 11 for 112(b) rejection was found persuasive. Examiner withdrawn the 112 (b) rejection for claim 9 – 11.
Applicants arguments regarding claim 1, 2, 4, 17 – 20 for 101 rejection was found persuasive. Examiner withdrawn the 101 rejection for claim 1, 2, 4, 17 – 20.
Response to Remarks
Applicant argues that Krivokon in view of Manato and Watanabe is silent on the following limitations below. Examiner respectfully disagrees for the reasons provided below:
In the Remarks (p. 21) regarding claim 1, applicants assert, “Watanabe does not disclose "a trained second machine learning model." Accordingly, Watanabe does not disclose "determining a state change of the target signal light from a time point when the second image is captured to a time point when the first image is captured by processing the second image and the first image based on the trained second machine learning model" as recited in amended claim 1. Further, Watanabe does not disclose "determining, based on a target state of the target signal light at the time point when the second image is captured and the state change, the second state of the target signal light in the first image"
Examiner respectfully disagrees because Krivokon is relied upon to teach a second machine learning model not Watanabe. In Krivokon [0066], the recurrent neural network model acts as multiple machine learning models when different multiple image frames at different time points are subjected for analysis/learning. This point was made in the last Office Action. In contrast, Watanabe is relied upon for recognizing the difference between a region of first and second images with the implication of assessing differences in luminance. Thus, Watanabe would have rendered determining the change of luminance between the first and second image obvious.
In the Remarks (p. 26) regarding claim 5, applicants assert, Lei fails to disclose, inter alia, "obtaining, based on a confidence level corresponding to the result of state change, a second confidence level of the second state or an adjusted second confidence level; and determining, based on the first confidence level and one of the second confidence level and the adjusted second confidence level, the target state" recited in amended claim 5.
Examiner respectfully disagrees because in [Page – 2, Paragraph – 1] Lei discloses, “If not, the first confidence level of the current estimated state and the second confidence level of the control state are obtained; the control state is a state of a traffic light obtained according to a control signal in the target reference information” wherein “obtaining, based on a confidence level corresponding to the result of state change” equates to first confidence level of the current estimated state, “second confidence level of the second state” implies to second confidence level of the control state. Lei determines final traffic light state by comparing the confidence weight of the states.
In the Remarks (p. 26) regarding claim 7, applicants assert, “Lei fails to disclose, inter alia, "determining a product of the first confidence level and a preset first factor as the adjusted first confidence level" recited in claim 7”.
Examiner respectfully disagrees because in Lei in [Page – 8; Paragraph – 5] discloses, “You can increase the preset value based on the initial confidence; when the color of the traffic light in the current estimated state changes from red to yellow within the preset duration, the preset value can be reduced” wherein increasing or decreasing a value performs the same mathematical function as multiplying the value with a preset factor.
In the Remarks (p. 28) regarding claim 8, applicants assert, Lei fails to disclose, inter alia, "determining, based on the state change, a determination mode corresponding to one of the second confidence level and the adjusted second confidence level; and determining, based on a confidence level of a target state corresponding to the target signal light at a time point when the second image is captured, the confidence level corresponding to the state change, and the determination mode, the second confidence level or the adjusted second confidence level"
Examiner respectfully disagrees because in Lei in in [Page – 2; Paragraph – 3]; “the first confidence level is determined by a change in a current estimated state within a preset time period; the second confidence level is determined by a time difference between a transmission time of a control signal and a time stamp of a traffic light image” wherein “change in a current estimated state” equates to state change and in [Page – 8, Paragraph – 5] Lei discloses about selecting whether to increase or decrease confidence based on transition type, “the color of the traffic light changes from green to yellow in the current estimated state within a preset time period , You can increase the preset value based on the initial confidence; when the color of the traffic light in the current estimated state changes from red to yellow within the preset duration, the preset value can be reduced”.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 2, 15 – 18, are rejected under 35 U.S.C 103 as being unpatentable over Krivokon et al. US Patent Application Publication No. US-20200135030-A1 (hereinafter Krivokon) in view of Manato “Traffic light recognition using high-definition map features" (hereinafter Manato) and further in view of Watanabe Patent Application Publication No. JP-2015153312-A (hereinafter Watanabe).
Regarding claim 1, Krivokon discloses a method for detecting a signal light, comprising: obtaining a first image and a second image previous to the first image in time sequence, wherein both the first image and the second image include a same target signal light (Krivokon in [0022] discloses, “the model may be trained using labeled images of the same traffic light captured over time” wherein capturing images overtime equates to capturing first and second image); determining, based on the first image, a first state of the target signal light in the first image; determining, based on the second image and the first image, a second state of the target signal light in the first image (Krivokon in [0066] discloses, “the model may be a recurrent neural network or a long short term memory neural network. For instance, over time, the state of the traffic light 720 would change and appear different in images captured at different times. This, as well as some heuristics about traffic light patterns, may also be used to determine or confirm a detected state of a traffic light (and lane)”), wherein the determining, based on the second image and the first image, a second state of the target signal light in the first image includes: determining the second state by processing, based on a trained second machine learning model, the second image and the first image, wherein the trained second machine learning model determines the second state based on information in a time domain of the target signal light in the first image and in the second image (Krivokon in [0066] discloses, “In addition or alternatively, the model may be trained using images of the same traffic light captured over time. In this regard, the model may be a recurrent neural network or a long short term memory neural network. For instance, over time, the state of the traffic light 720 would change and appear different in images captured at different times. This, as well as some heuristics about traffic light patterns, may also be used to determine or confirm a detected state of a traffic light (and lane)”);
Krivokon doesn’t disclose about the following limitation as further recited in the claim.
Manato discloses about determining, based on the first state and the second state, the target state of the target signal light at the time point when the first image is captured (Manato in [Section – 4.3, Paragraph – 2] discloses, “The inter-frame filter returns the current color state by comparing the previous recognition result to the recognition result for the current frame”).
It would have been obvious to one with one having an ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Manato into the system of Krivokon because it would make the output result of the system more accurate in different scenarios.
Krivokon and Manato in the combination doesn’t disclose about the following limitation as further recited in the claim.
Watanabe discloses the determining the second state by processing, based on the trained (Watanabe in [0020] discloses, “the lighting state of the traffic light changes from red to blue between the imaging time of the first image and the imaging time of the second image, an area corresponding to the red traffic light and the green traffic light in the difference image”. Furthermore, Watanabe recognizing the difference of a region of first and second images would imply knowing the difference in luminance. Therefore, would have render determining the change of luminance of between the first and second image obvious).
It would have been obvious to one with one having an ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Watanabe into the system of Krivokon in view of Manato because it would increase the detection accuracy and false positive of the system by using the historical data and learned transition pattern.
Summary of Citations (Watanabe)
Paragraph [0020]; “the lighting state of the traffic light changes from red to blue between the imaging time of the first image and the imaging time of the second image, an area corresponding to the red traffic light and the green traffic light in the difference image”.
Summary of Citations (Krivokon)
Paragraph [0022]; “the model may be trained using labeled images of the same traffic light captured over time”.
Paragraph [0059]; “The area of this projection 810 in the image may then be processed to determine the state of the traffic light in the image 600”.
Paragraph [0066]; ““In addition or alternatively, the model may be trained using images of the same traffic light captured over time. In this regard, the model may be a recurrent neural network or a long short term memory neural network. For instance, over time, the state of the traffic light 720 would change and appear different in images captured at different times. This, as well as some heuristics about traffic light patterns, may also be used to determine or confirm a detected state of a traffic light (and lane). For instance, if a traffic light is green, the next color the traffic light could be would not be red. Thus, lane states may be determined such that temporal consistency between traffic light states is enforced. This may also help the model to provide information about the dynamic states of lights, such as flashing red or yellow lights”.
Summary of Citations (Manato)
[Section – 4.3, Paragraph – 2]; “The inter-frame filter returns the current color state by comparing the previous recognition result to the recognition result for the current frame”.
Regarding claim 2, 15, 16 the grounds of rejection and motivation to combine from the last Office Action with respect to Krivokon in view of Manato and Watanabe apply here.
Regarding claim 17, apparatus claim 17 corresponds to method claim 1. Therefore, the rejection analysis and motivation to combine of claim 1 is applicable to claim 17.
Regarding claim 18, is a non-transitory computer readable storage medium claim
corresponds to method claim 1. Therefore, the rejection analysis of claim 1 is applied in claim 18.
Claims 5 – 8 and 23 are rejected under 35 U.S.C 103 as being unpatentable over Krivokon in view of Manato and Watanabe and further in view of Lei Patent Application Publication No. CN-110619307-A (hereinafter Lei).
Krivokon, Manato and Watanabe in the combination fails to teach the limitations as recited in claims 5 – 8 respectively. However, Lei does. The grounds of rejection and motivation to combine from the last Office Action with respect to Lei apply here.
Regarding claim 23, Lei in the combination discloses the method of claim 5, wherein the obtaining, based on a confidence level corresponding to the result of state change, a second confidence level of the second state or an adjusted second confidence level includes (Lei in [Page – 13; Paragraph – 5] discloses, “the first confidence level is determined by a change in a current estimated state within a preset time period; the second confidence level is determined by a time difference between a transmission time of the control signal and a time stamp of the traffic light image, and The working status of the traffic light is determined”): in response to determining that the target signal light is on or when the target signal light is off (Lei in [Page – 8; Paragraph – 5] discloses, “when the color of the traffic light changes from green to yellow in the current estimated state within a preset time period”), determining the confidence level corresponding to the state change as the adjusted second confidence level or the second confidence level (Lei discloses about second confidence level in [Page – 8; Paragraph – 6]).
Summary of Citations (Lei)
[Page – 8; Paragraph – 5]; “when the color of the traffic light changes from green to yellow in the current estimated state within a preset time period”.
[Page – 8; Paragraph – 6]; “The second confidence level is determined by the time difference between the time when the control signal is sent and the time stamp of the traffic light image, and the working state of the traffic light. The electronic device can set an initial value for the current estimated state, and then adjust it based on the above time difference and the working state of the traffic light. For example, when the time difference is relatively small, a preset value may be added to the initial confidence level. When a traffic light abnormality report is received, the preset value may be reduced based on the initial confidence level”.
[Page – 13; Paragraph – 5]; “the first confidence level is determined by a change in a current estimated state within a preset time period; the second confidence level is determined by a time difference between a transmission time of the control signal and a time stamp of the traffic light image, and The working status of the traffic light is determined”.
Claims 9 – 14 and 24 are rejected under 35 U.S.C 103 as being unpatentable over Krivokon in view of Manato, Watanabe and Lei and further in view of Zhuang Patent Application Publication No. CN-110532903-A (hereinafter Zhuang).
Regarding claim 9, Lei in the combination discloses the method of claim 5, wherein the determining, based on the first confidence level and one of the second confidence level and the adjusted second confidence level, the target state includes (Lei in [Page – 8; Paragraph – 7] discloses, “Determine the current actual state of the traffic light according to the first confidence level and the second confidence level”. Furthermore, in [Page – 8; Paragraph – 6] discloses about adjusting second confidence level): determining, based on the first comparison result, whether the first confidence level needs to be corrected (Lei in [Page – 2; Paragraph – 8] discloses about whether the confidence level need to change based on abnormality, “determining whether the traffic light is abnormal according to the received traffic light abnormality report or control signal”. Additionally, Lei in [Page – 2; Paragraph – 5] discloses about first confidence level, “the first initial confidence of each field in the current estimated state”): in response to determining that the first confidence needs to be corrected, correcting the first confidence level to determine an adjusted first confidence level (Lei in [Page – 8; Paragraph – 5] discloses, “the first confidence level is determined by a change in the current estimated state within a preset time period ... You can increase the preset value based on the initial confidence; when the color of the traffic light in the current estimated state changes from red to yellow within the preset duration, the preset value can be reduced”), and determining, based on the adjusted first confidence level and one of the second confidence level and the adjusted second confidence level, the target state (Lei in [Page – 8; Paragraph – 7] discloses, “Determine the current actual state of the traffic light according to the first confidence level and the second confidence level”. Additionally, in [Page – 8; Paragraph – 6] discloses about adjusting second confidence level, “The second confidence level ... is determined by the time difference between the time when the control signal is sent and the time stamp of the traffic light image, and the working state of the traffic light. ... adjust it based on the above time difference and the working state of the traffic light”).
Krivokon, Manato, Watanabe and Lei in the combination doesn’t disclose the limitations as further recited in the claim.
Zhuang discloses obtaining a first position of the target signal light in the first image through a trained first machine learning model (Zhuang in [0180] discloses, “determining a first position area of each traffic light contained in the first rectangular frame through a neural network model and acquiring the color state of the traffic light when the traffic light image is acquired”), determining a first comparison result by comparing the first position and a reference position (Zhuang in [0083] discloses, comparing the position area of the red light output by the neural network model with the reference position of the traffic light determined by the previous N frames of snap images”); the reference position including reference coordinates or a reference box of the target signal light in the first image and the second image(Zhuang in [0192] discloses, performing weighted calculation on a first position area corresponding to the traffic light with the color state of red light output by the neural network model when the traffic light with the color state of red light is positioned in the first N frames of traffic light images, so as to obtain a second position area corresponding to the traffic light” wherein second position area (reference coordinate), and the disclosed current frame and previous N frame equates to first and second image).
It would have been obvious to one with one having an ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Zhuang into the system of Krivokon in view of Manato, Watanabe and Lei because it would allow the system to determine and penalize the detection that appear outside the signal light region resulting improving the reliability of the confidence level and output.
Summary of Citations (Lei)
[Page – 2; Paragraph – 5]; “the first initial confidence of each field in the current estimated state”.
[Page – 2; Paragraph – 8]; “determining whether the traffic light is abnormal according to the received traffic light abnormality report or control signal”.
[Page – 8; Paragraph – 5]; “the first confidence level is determined by a change in the current estimated state within a preset time period ... You can increase the preset value based on the initial confidence; when the color of the traffic light in the current estimated state changes from red to yellow within the preset duration, the preset value can be reduced”.
[Page – 8; Paragraph – 6]; “The second confidence level is used to characterize the credibility of the control state. The second confidence level is determined by the time difference between the time when the control signal is sent and the time stamp of the traffic light image, and the working state of the traffic light. The electronic device can set an initial value for the current estimated state, and then adjust it based on the above time difference and the working state of the traffic light”.
[Page – 8; Paragraph – 7]; “Determine the current actual state of the traffic light according to the first confidence level and the second confidence level”.
Summary of Citations (Zhuang)
Paragraph [0083]; comparing the position area of the red light output by the neural network model with the reference position of the traffic light determined by the previous N frames of snap images”.
Paragraph [0129]; “calculating the position coordinate and the score of the traffic light with the highest confidence coefficient by adopting a non-maximum suppression algorithm, obtaining the color state of the traffic light according to the score”.
Paragraph [0180]; “determining a first position area of each traffic light contained in the first rectangular frame through a neural network model and acquiring the color state of the traffic light when the traffic light image is acquired”.
Paragraph [0192]; “performing weighted calculation on a first position area corresponding to the traffic light with the color state of red light output by the neural network model when the traffic light with the color state of red light is positioned in the first N frames of traffic light images, so as to obtain a second position area corresponding to the traffic light”.
Regarding claim 10, Lei in the combination discloses the method of claim 5, further comprising wherein the determining, based on the first confidence level and one of the second confidence level and the adjusted second confidence level, the target state includes (Lei in [Page – 8; Paragraph – 7] discloses, “Determine the current actual state of the traffic light according to the first confidence level and the second confidence level”. Furthermore, in [Page – 8; Paragraph – 6] discloses about adjusting second confidence level): determining, based on the second comparison result, whether the second confidence level needs to be corrected in response to determining that the second confidence needs to be corrected, correcting the second confidence level to determine a corrected second confidence level (Lei in [Page – 8; Paragraph – 6] discloses, “The second confidence level is determined by the time difference between the time when the control signal is sent and the time stamp of the traffic light image, and the working state of the traffic light. The electronic device can set an initial value for the current estimated state, and then adjust it based on the above time difference and the working state of the traffic light. For example, when the time difference is relatively small, a preset value may be added to the initial confidence level. When a traffic light abnormality report is received, the preset value may be reduced based on the initial confidence level”), and determining, based on the first confidence level and the corrected second confidence level, the target state (Lei in [Page – 8; Paragraph – 7] discloses, “Determine the current actual state of the traffic light according to the first confidence level and the second confidence level”).
Krivokon further discloses second machine learning model (Krivokon in [0058] discloses about second model).
Zhuang further discloses obtaining a second position of the target signal light in the first image through the trained (Zhuang in [0106] discloses about minimum bounding box (second position), “a minimum bounding rectangle of the traffic light is determined, and coordinates of a center point of the minimum bounding rectangle in the first rectangle and length and width values of the minimum bounding rectangle are output”); determining a second comparison result by comparing the second position and a reference position (Zhuang in [0033 – 0034] discloses, “determining a second position area corresponding to the traffic light with the color state of red light output by the neural network model when the color state of the traffic light with the color state of red light is red light in the first N frames of traffic light images; comparing the first position area with the second position area of the same traffic light”); the reference position including reference coordinates or a reference box of the target signal light in the first image and the second image (Zhuang in [0192] discloses, performing weighted calculation on a first position area corresponding to the traffic light with the color state of red light output by the neural network model when the traffic light with the color state of red light is positioned in the first N frames of traffic light images, so as to obtain a second position area corresponding to the traffic light” wherein second position area (reference coordinate), and the disclosed current frame and previous N frame equates to first and second image).
Summary of Citations (Lei)
[Page – 8; Paragraph – 6]; “The second confidence level is determined by the time difference between the time when the control signal is sent and the time stamp of the traffic light image, and the working state of the traffic light. The electronic device can set an initial value for the current estimated state, and then adjust it based on the above time difference and the working state of the traffic light. For example, when the time difference is relatively small, a preset value may be added to the initial confidence level. When a traffic light abnormality report is received, the preset value may be reduced based on the initial confidence level”.
[Page – 8; Paragraph – 7]; “Determine the current actual state of the traffic light according to the first confidence level and the second confidence level”.
Summary of Citations (Krivokon)
Paragraph [0058]; “existing models or image processing techniques may be used to label images of traffic lights as well as the state of the traffic lights. As noted above, the vehicle 100 may utilize a traffic light detection system software module configured to detect the states of known traffic signals based on a priori locations, such as those identified in the map information 200”.
Summary of Citations (Zhuang)
Paragraph [0033 – 0034]; “determining a second position area corresponding to the traffic light with the color state of red light output by the neural network model when the color state of the traffic light with the color state of red light is red light in the first N frames of traffic light images; comparing the first position area with the second position area of the same traffic light”.
Paragraph [0106]; “a minimum bounding rectangle of the traffic light is determined, and coordinates of a center point of the minimum bounding rectangle in the first rectangle and length and width values of the minimum bounding rectangle are output”.
Paragraph [0192]; “performing weighted calculation on a first position area corresponding to the traffic light with the color state of red light output by the neural network model when the traffic light with the color state of red light is positioned in the first N frames of traffic light images, so as to obtain a second position area corresponding to the traffic light”.
Regarding claim 11, Zhuang in the combination further discloses the method of claim 9, further comprising: in response to a determination that the first confidence level needs to be corrected, correcting the first confidence level based on an intersection-over-union between the first position and the reference position; or in response to a determination that the second confidence level needs to be corrected, correcting the second confidence level based on an intersection-over-union between the second position and the reference position (Zhuang in [0089] discloses, “the first position area corresponding to the traffic light detected by the neural network model is compared with the second position area corresponding to the traffic light when the color state of the traffic light in the previous N frames traffic light image is the red light ... if the difference does not exceed the second preset threshold, the pixel to be processed is determined through the first position area, so that the image processing precision is improved, and image error processing caused by error position information output by the neural network model is avoided”).
Summary of Citations (Zhuang)
Paragraph [0089]; “the first position area corresponding to the traffic light detected by the neural network model is compared with the second position area corresponding to the traffic light when the color state of the traffic light in the previous N frames traffic light image is the red light ... if the difference does not exceed the second preset threshold, the pixel to be processed is determined through the first position area, so that the image processing precision is improved, and image error processing caused by error position information output by the neural network model is avoided”.
Regarding claim 12, Zhuang in the combination further discloses the method of claim 9, wherein the reference position is determined by operations including: obtaining a sample set including positions of the target signal light in a plurality of sample images captured within a preset time period, the preset time period being before a time point when the second image is captured (Zhuang in [0151 – 0152] discloses, “determining the corresponding position area of the traffic light 1 when the color state of the traffic light 1 in the first N frames of traffic light images is red light, and performing weighted fusion on the position areas corresponding to the traffic light 1 in the first N frames to obtain a second position area corresponding to the traffic light 1. Such as: assuming that N is 3, the current frame traffic light image is the 6 th frame, and it is determined through the neural network model that the color states of the traffic light 1”); obtaining, based on a clustering algorithm, a clustering result by clustering the positions of the target signal light in the plurality of sample images in the sample set; and determining, based on the clustering result, the reference position (Zhuang in [0097] discloses, “clustering the input traffic light candidate frames under the condition that the distance between any two traffic light candidate frames is calculated, and dividing the two traffic light candidate frames ... the distance between the two points is the distance between the two traffic light candidate frames. Such as: the distance between two traffic light candidate frames is calculated from the coordinates (x _ top, y _ top) of the upper left corners of any two traffic light candidate frames”).
Summary of Citations (Zhuang)
Paragraph [0151 – 0152]; “determining the corresponding position area of the traffic light 1 when the color state of the traffic light 1 in the first N frames of traffic light images is red light, and performing weighted fusion on the position areas corresponding to the traffic light 1 in the first N frames to obtain a second position area corresponding to the traffic light 1. Such as: assuming that N is 3, the current frame traffic light image is the 6 th frame, and it is determined through the neural network model that the color states of the traffic light 1”.
Paragraph [0097]; “clustering the input traffic light candidate frames under the condition that the distance between any two traffic light candidate frames is calculated, and dividing the two traffic light candidate frames with the distance not exceeding a first preset threshold value into the same first rectangular frame. That is, a point at the same position on the two traffic light candidate frames is selected, and the distance between the two points is the distance between the two traffic light candidate frames. Such as: the distance between two traffic light candidate frames is calculated from the coordinates (x _ top, y _ top) of the upper left corners of any two traffic light candidate frames”.
Regarding claim 13, Zhuang in the combination further discloses the obtaining, based on a clustering algorithm, a clustering result by clustering the positions of the target signal light in the plurality of sample images in the sample set includes: determining, based on the sample set, an initialization reference position set and a radius, the initialization reference position set including one or more initialization reference positions (Zhuang in [0097] discloses, “clustering the input traffic light candidate frames under the condition that the distance between any two traffic light candidate frames is calculated” wherein the distance between the two traffic light is the radius); determining, based on the initialization reference position set and the radius, an updated reference position set through a means clustering algorithm (Zhuang in [0094] discloses, “The traffic light candidate frames are divided through a clustering algorithm to obtain N first rectangular frames with preset specifications ... The following process of dividing the traffic light candidate frame by the clustering algorithm to obtain N first rectangular frames”); determining whether a termination condition is satisfied (Zhuang in [0097] discloses, “clustering the input traffic light candidate frames under the condition that the distance between any two traffic light candidate frames is calculated, and dividing the two traffic light candidate frames with the distance not exceeding a first preset threshold value into the same first rectangular frame”); and in response to a determination that the termination condition is satisfied, obtaining the clustering result, wherein the clustering result includes the updated reference position set (Zhuang in fig. 4 and Fig.9 disclosed about updated reference position set).
Summary of Citations (Zhuang)
Paragraph [0094]; “The traffic light candidate frames are divided through a clustering algorithm to obtain N first rectangular frames with preset specifications, namely the first rectangular frames are rectangles with the same length and width values, at least one traffic light candidate frame is contained in each first rectangular frame, and one traffic light candidate frame can only exist in one first rectangular frame. The following process of dividing the traffic light candidate frame by the clustering algorithm to obtain N first rectangular frames”.
Paragraph [0097]; “clustering the input traffic light candidate frames under the condition that the distance between any two traffic light candidate frames is calculated, and dividing the two traffic light candidate frames with the distance not exceeding a first preset threshold value into the same first rectangular frame”.
Regarding claim 14, the grounds of rejection and motivation to combine from the last Office Action with respect to Krivokon, Manato, Watanabe, Lei and Zhuang in the combination apply here.
Regarding claim 24, Zhuang in the combination further discloses the trained second machine learning model includes a plurality of structural layers, a convolutional layer, an activation layer, a pooling layer, an upsampling layer, and a cascade operation (Zhuang in [0113] discloses, ““the neural network model ... composed of a plurality of convolutional layers, pooling layers, upsampling layers, and cascade layers”. Additionally, Zhuang in [0122] discloses, “Performing pooling treatment on the data after the activation treatment of the result of the first layer of convolution operation”).
Summary of Citations (Zhuang)
Paragraph [0113]; “the neural network model in the embodiment of the present invention is composed of a plurality of convolutional layers, pooling layers, upsampling layers, and cascade layers”.
Paragraph [0122]; “Performing pooling treatment on the data after the activation treatment of the result of the first layer of convolution operation”.
Claim 21 is rejected under 35 U.S.C 103 as being unpatentable over Krivokon in view of Manato, Watanabe and Lei and further in view of Hashimoto US Patent Application Publication No. US-20210312198-A1 (hereinafter Hashimoto).
Regarding claim 21, Lei in the combination discloses the method of claim 5.
Krivokon, Manato, Watanabe and Lei in the combination doesn’t disclose about the following limitation as further recited in the claim.
Hashimoto discloses the obtaining, based on a confidence level corresponding to the state change, a second confidence level of the second state or an adjusted second confidence level includes: determining a product of a confidence level of the target state corresponding to the target signal light at the time point when the second image is captured and the confidence level corresponding to the state change as the second confidence level of the second state (Hashimoto in [0013] discloses, “the processor calculates, for each of the candidate states of the signal light, a corrected confidence score by multiplying the confidence score of the candidate state by the probability of transition from the preceding state of the signal light to the candidate state, and identifies the state of the signal light as one of the candidate states of the signal light having a maximum corrected confidence score” wherein corrected confidence score equates to second confidence level); and obtaining the adjusted second confidence level by adjusting the second confidence level based on a preset second factor (Hashimoto in [0087] discloses, “probabilities of transition between the candidate states are prestored in the memory”).
It would have been obvious to one with one having an ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Hashimoto into the system of Krivokon in view of Manato, Watanabe and Lei because it would allow the system to improve signal light state determination based on both prior state confidence level and state change confidence level.
Summary of Citations (Hashimoto)
Paragraph [0013]; “the processor calculates, for each of the candidate states of the signal light, a corrected confidence score by multiplying the confidence score of the candidate state by the probability of transition from the preceding state of the signal light to the candidate state, and identifies the state of the signal light as one of the candidate states of the signal light having a maximum corrected confidence score”.
Paragraph [0087]; “probabilities of transition between the candidate states are prestored in the memory”.
Allowable Subject Matter
Claim 22 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter.
Regarding claim 22, the prior art references taken individually or in combination fail to particularly disclose, fairly suggest, or render obvious the limitations as further recited. The applied prior arts Manato, Watanabe and Lei and Hashimoto doesn’t disclose the limitation, determining the state change is that a color of the target signal light changes from the time point when the second image is captured to the time point when the first image is captured, determining an average value of the confidence level corresponding to the state change and the confidence level of the target state corresponding to the target signal light at the time point when the second image is captured as the adjusted second confidence level. Hashimoto discloses about a product of a confidence level of the target state and confidence level corresponding to the state change as the second confidence level of the second state but didn’t disclose about average value of the confidence level corresponding to the state change and the confidence level of the target state corresponding to the target signal light
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.
Contact
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ZAID MUHAMMAD SALEH whose telephone number is (703)756-1684. The examiner can normally be reached M-F 8 am - 5 pm ET.
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, Vu Le can be reached on (571)272-7332. 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.
/ZAID MUHAMMAD SALEH/
Examiner, Art Unit 2668
02/01/2026
/VU LE/Supervisory Patent Examiner, Art Unit 2668