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 .
Claim Status
Applicant’s amendment filed on January 15, 2026 is acknowledged. Currently claims 1-12, 16-17, and 20-23 are pending. Claims 1, 7, and 16 have been amended. Claims 13-15, 18-19, and 24 are cancelled.
Response to Arguments
Applicant's arguments filed January 15, 2026, have been fully considered but they are not persuasive.
On pages 9-10 of Applicant’s remarks, Applicant alleges that Eagelberg, Zhu, and Soni, taken alone or as a whole, fail to teach at least “the detection mark is located at an apex angular position on a side of the traffic marker closest to the image acquisition device” of claims 1, 7, and 16.
The examiner respectfully disagrees. Under the broadest reasonable interpretation, the detection mark is located at an apex angular position on a side of the traffic marker closest to the image acquisition device can be interpreted as a detection mark that is orientated angularly on the side of the land mark closest to the image acquisition device. Therefore, the examiner asserts that Eagelberg teaches the detection mark is located at an apex angular position on a side of the traffic marker closest to the image acquisition device ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) ([0049] “In some embodiments, image acquisition unit 120 may include one or more image capture devices (e.g., cameras, CCDs, or any other type of image sensor), such as image capture device 122, image capture device 124, and image capture device 126.” wherein an image acquisition device on a vehicle is one or more image capture devices) ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a detection mark is a lane mark characteristic) ([0118] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, hazardous objects, and any other feature associated with an environment of a vehicle.” wherein a side of the traffic marker adjacent to the vehicle is a set of features associated with an environment of a vehicle; it is obvious to one of ordinary skill in the art that the detection mark is located on a side of the traffic marker closest to the image acquisition device to ensure the traffic marker is being detected) ([0186] “For example, a lane mark may be adjoining (e.g., next to or joining) another lane mark, or the lane mark may be oriented in a particular direction (e.g., at a particular angle) in relation to another lane mark.”).
On pages 10-12 of Applicant’s remarks, Applicant further alleges that Eagelberg in view of Zhu, Soni and Maheshwari does not teach the following distinguishing features in the amended claim 1:
outputting the detection line for a plurality of traffic sign elements in response to determining that a pixel distance between any two adjacent traffic sign elements along an extending direction of the road is less than a preset threshold value;
outputting the detection point for a plurality of traffic sign elements in response to determining that the pixel distance between any two adjacent traffic sign elements along the extending direction of the road is larger than the preset threshold value.
The examiner respectfully disagrees. Under the broadest reasonable interpretation, outputting the detection line for a plurality of traffic sign elements in response to determining that a pixel distance between any two adjacent traffic sign elements along an extending direction of the road is less than a preset threshold value can be interpretated as outputting a lane marking characteristic on surrounding traffic elements when a pixel distance between two adjacent traffic elements is less than a preset threshold value. Further, under the broadest reasonable interpretation, outputting the detection point for a plurality of traffic sign elements in response to determining that the pixel distance between any two adjacent traffic sign elements along the extending direction of the road is larger than the preset threshold value can be interpretated as outputting a different lane mark characteristic on surrounding traffic elements when a pixel distance between two adjacent traffic elements is larger than a preset threshold value.
Therefore, the examiner asserts that Eagelberg in view of Zhu, Soni, and Maheshwari teaches outputting the detection line for a plurality of traffic sign elements in response to determining that a pixel distance between any two adjacent traffic sign elements along an extending direction of the road is less than a preset threshold value (Eagelberg - [0163] “The at least one processing device may be further programmed to detect one or more lane mark characteristics of the at least one identified lane mark, and use the one or more detected lane mark characteristics to determine a type of the at least one identified lane mark.”) (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark.”) (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like) (Soni - [Col.7, lines 1-8] “The road network module 37 may calculate the predetermined path geometry based on a predetermined distance. The road network module 37 may access the predetermined distance as set by user input. The road network module 37 may access a centerline or a path boundary for at least one path from the map data 31. The road network module 37 calculates the predetermined path geometry based on a predetermined distance to the centerline or path boundary.” wherein a pixel distance is a predetermined distance) (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a preset threshold value and a local maximum value are local maxima of the response map with confidence values above a threshold) (Maheshwari - [0100] “The lane detection module 104 may compare the measured distances (643, 644) to a threshold distance, and the threshold distance may be predetermined or dynamically determined as part of a machine learning model. If the lane detection module 104 determines that the distance 643 associated with the first pixel 641 is less than the threshold distance, the lane detection module 104 may determine that the first pixel 641 should be linked in a subsection 645a with the beginning pixel 637 and any other pixels in between that satisfy the distance threshold. By contrast, if the lane detection module 104 determines that the distance 644 associated with the second pixel 642 is greater than the threshold distance, the lane detection module 104 may include the second pixel 642 in a different subsection 645b than the first pixel 641.”);
outputting the detection point for a plurality of traffic sign elements in response to determining that the pixel distance between any two adjacent traffic sign elements along the extending direction of the road is larger than the preset threshold value (Eagelberg - [0163] “The at least one processing device may be further programmed to detect one or more lane mark characteristics of the at least one identified lane mark, and use the one or more detected lane mark characteristics to determine a type of the at least one identified lane mark.”) (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark.”) (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like) (Soni - [Col.7, lines 1-8] “The road network module 37 may calculate the predetermined path geometry based on a predetermined distance. The road network module 37 may access the predetermined distance as set by user input. The road network module 37 may access a centerline or a path boundary for at least one path from the map data 31. The road network module 37 calculates the predetermined path geometry based on a predetermined distance to the centerline or path boundary.” wherein a pixel distance is a predetermined distance) (Eagelberg - [0144] “In one embodiment, processing unit 110 may calculate the distance between a snail trail and a road polynomial (e.g., along the trail). If the variance of this distance along the trail exceeds a predetermined threshold (for example, 0.1 to 0.2 meters on a straight road, 0.3 to 0.4 meters on a moderately curvy road, and 0.5 to 0.6 meters on a road with sharp curves), processing unit 110 may determine that the leading vehicle is likely changing lanes.” wherein the preset threshold value is a predetermined threshold) (Maheshwari - [0100] “The lane detection module 104 may compare the measured distances (643, 644) to a threshold distance, and the threshold distance may be predetermined or dynamically determined as part of a machine learning model. If the lane detection module 104 determines that the distance 643 associated with the first pixel 641 is less than the threshold distance, the lane detection module 104 may determine that the first pixel 641 should be linked in a subsection 645a with the beginning pixel 637 and any other pixels in between that satisfy the distance threshold. By contrast, if the lane detection module 104 determines that the distance 644 associated with the second pixel 642 is greater than the threshold distance, the lane detection module 104 may include the second pixel 642 in a different subsection 645b than the first pixel 641.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include a pixel distance of Soni that corresponds with a preset threshold value of Zhu, wherein the pixel distance is less than or larger than a preset threshold value, of Maheshwari, in the traffic marker detection method of Eagelberg to accurately classify traffic markers according to their position and distance relative to a vehicle.
The examiner respectfully suggests Applicant amend the independent claims to reflect that which is disclosed in paragraphs [0051] and [0053] of Applicant’s specifications to distinguish the usage of a detection point and detection line respectfully. Additionally the examiner suggests Applicant amend the independent claims to specify that the “apex angular position” as recited in the independent claims, enables the detection mark to move according to the position of the traffic marker relative to the vehicle, as disclosed in paragraph [0057] of Applicant’s specifications.
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.
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.
Claims 1-12, 16-17, and 20-23 are rejected under 35 U.S.C. 103 as being unpatentable of Eagelberg et al., US 20180120859 A1, (hereinafter “Eagelberg”) in view of Zhu et al., US 20040234136 A1, (hereinafter “Zhu”) in further view of Soni et al., US 11030457 B2, (hereinafter “Soni”) in further view of Maheshwari et al., US 20210209941 A1, (hereinafter “Maheshwari”).
Regarding claim 1, Eagelberg teaches a traffic marker detection method, comprising:
acquiring a target image containing a traffic marker in a road ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.); and
inputting the target image into a traffic marker detection model to obtain a detection mark ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics; a detection mark is a lane mark characteristic) corresponding to the traffic marker ([0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like);
wherein the detection mark comprises at least one of a detection point and a detection line for characterizing a position ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein the detection mark is a lane mark characteristic and a detection point is a reference point) of the traffic marker in the target image ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.);
wherein the traffic marker comprises a plurality of traffic sign elements ([0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like), and
wherein the inputting the target image into the traffic marker detection model to obtain the detection mark corresponding to the traffic marker comprises ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics; a detection mark is a lane mark characteristic) ([0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like):
outputting the detection line for a plurality of traffic sign elements (Eagelberg - [0163] “The at least one processing device may be further programmed to detect one or more lane mark characteristics of the at least one identified lane mark, and use the one or more detected lane mark characteristics to determine a type of the at least one identified lane mark.”) (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark.”) (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like);
outputting the detection point for the plurality of traffic sign elements in response to determining direction of a road (Eagelberg - [0163] “The at least one processing device may be further programmed to detect one or more lane mark characteristics of the at least one identified lane mark, and use the one or more detected lane mark characteristics to determine a type of the at least one identified lane mark.”) (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark.”) (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like);
wherein the target image is obtained by an image acquisition device on a vehicle, and the detection mark is located at an apex angular position on a side of the traffic marker closest to the image acquisition device ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) ([0049] “In some embodiments, image acquisition unit 120 may include one or more image capture devices (e.g., cameras, CCDs, or any other type of image sensor), such as image capture device 122, image capture device 124, and image capture device 126.” wherein an image acquisition device on a vehicle is one or more image capture devices) ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a detection mark is a lane mark characteristic) ([0118] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, hazardous objects, and any other feature associated with an environment of a vehicle.” wherein a side of the traffic marker adjacent to the vehicle is a set of features associated with an environment of a vehicle; it is obvious to one of ordinary skill in the art that the detection mark is located on a side of the traffic marker closest to the image acquisition device to ensure the traffic marker is being detected) ([0186] “For example, a lane mark may be adjoining (e.g., next to or joining) another lane mark, or the lane mark may be oriented in a particular direction (e.g., at a particular angle) in relation to another lane mark.”).
Eagelberg does not specifically disclose a preset threshold value.
However, Zhu teaches a preset threshold value ([0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a preset threshold value is a threshold).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include a preset threshold value of Zhu, in the traffic marker detection model of Eagelberg to better train the accuracy of the detection model.
Eagelberg in view of Zhu does not specifically disclose a pixel distance.
However, Soni teaches a pixel distance ([Col.7, lines 1-8] “The road network module 37 may calculate the predetermined path geometry based on a predetermined distance. The road network module 37 may access the predetermined distance as set by user input. The road network module 37 may access a centerline or a path boundary for at least one path from the map data 31. The road network module 37 calculates the predetermined path geometry based on a predetermined distance to the centerline or path boundary.” wherein a pixel distance is a predetermined distance) ([Col.6, lines 38-40] “For example, the width of a road in the road network has a geographic distance, which is converted to a number of pixels in the image data 33 using the scaling factor.” wherein geographic distance is converted to pixels in image data).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include a pixel distance of Soni, in the traffic marker detection model of Eagelberg in view of Zhu to better train the accuracy of the detection model.
Eagelberg in view of Zhu and Soni does not specifically disclose determining that a pixel distance between any two adjacent traffic sign elements along an extending direction of the road is less than or larger than a preset threshold value.
However, Maheshwari teaches a pixel distance between any two adjacent traffic sign elements along an extending direction of the road is less than or larger than a preset threshold value ([0100] “The lane detection module 104 may compare the measured distances (643, 644) to a threshold distance, and the threshold distance may be predetermined or dynamically determined as part of a machine learning model. If the lane detection module 104 determines that the distance 643 associated with the first pixel 641 is less than the threshold distance, the lane detection module 104 may determine that the first pixel 641 should be linked in a subsection 645a with the beginning pixel 637 and any other pixels in between that satisfy the distance threshold. By contrast, if the lane detection module 104 determines that the distance 644 associated with the second pixel 642 is greater than the threshold distance, the lane detection module 104 may include the second pixel 642 in a different subsection 645b than the first pixel 641.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the traffic marker detection model of Eagelberg in view of Zhu, Soni by identifying and classifying the varying distances between traffic sign elements of Maheshwari to more specifically and accurately characterize the positions of traffic markers located within a target image.
Regarding claim 2, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 1, wherein the inputting the target image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) into the traffic marker detection model to obtain the detection mark corresponding to the traffic marker (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics; a detection mark is a lane mark characteristic) comprises:
obtaining a response map (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.”) corresponding to the target image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.), wherein the response has a plurality of positions and a response value of each position in the response map is used for characterizing a confidence level (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.”) (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a confidence level is confidence values) that the traffic marker exists at each position (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics; a detection mark is a lane mark characteristic); and
generating the detection mark (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein the detection mark is a lane mark characteristic) based on the response value at each position in the response map (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to add a response map with response values that characterize a confidence level of Zhu, to the traffic marker detection model of Eagelberg in view of Soni and Maheshwari so said confidence levels can better train the accuracy of the detection model.
Regarding claim 3, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 2, wherein generating the detection mark (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein the detection mark is a lane mark characteristic) based on the response values in the response map (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) comprises:
performing a sliding window operation (Soni - [Col. 11, lines 15-22] “FIG. 13 illustrates a sliding section technique. In FIG. 13, a sliding window 81 is moved across the image patch 61. Rather than define entire sections of roadway, the predetermined path geometry for training the lane detection model 40 is limited to the sliding window 81. The sliding section approach improves the speed of training the lane detection model 40 because a smaller portion of the image patch 61 is analyzed.”) in the response map (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.”);
in response to the response value (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) of a target point of the response map in a sliding window area satisfies a predetermined condition (Soni - [Col. 9, lines 3-11] “The predetermined path geometry may be defined according to lane boundaries 56 that are spaced by the calculated width from a center line 44. The predetermined path geometry may be defined differently for each section of roadway as demarcated between shape points 42. The shape points 42 may be determined based on the turns in the road. When a road turns more than a set angle, or has more than a predetermined curvature, a new shape point 42 is added to define a new road section.” wherein a target point is predetermined path geometry) (Soni - [Col. 11, lines 15-22] “FIG. 13 illustrates a sliding section technique. In FIG. 13, a sliding window 81 is moved across the image patch 61. Rather than define entire sections of roadway, the predetermined path geometry for training the lane detection model 40 is limited to the sliding window 81. The sliding section approach improves the speed of training the lane detection model 40 because a smaller portion of the image patch 61 is analyzed.” wherein the sliding window area is the image patch 61) (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a predetermined condition is confidence values above a threshold), generating the detection mark (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein the detection mark is a lane mark characteristic) at the target point of the sliding window area (Soni - [Col. 9, lines 3-11] “The predetermined path geometry may be defined according to lane boundaries 56 that are spaced by the calculated width from a center line 44. The predetermined path geometry may be defined differently for each section of roadway as demarcated between shape points 42. The shape points 42 may be determined based on the turns in the road. When a road turns more than a set angle, or has more than a predetermined curvature, a new shape point 42 is added to define a new road section.” wherein a target point is predetermined path geometry) (Soni - [Col. 11, lines 15-22] “FIG. 13 illustrates a sliding section technique. In FIG. 13, a sliding window 81 is moved across the image patch 61. Rather than define entire sections of roadway, the predetermined path geometry for training the lane detection model 40 is limited to the sliding window 81. The sliding section approach improves the speed of training the lane detection model 40 because a smaller portion of the image patch 61 is analyzed.” wherein the sliding window area is the image patch 61);
wherein the predetermined condition (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a predetermined condition is confidence values above a threshold) is that the response value (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) of the target point of the sliding window area (Soni - [Col. 9, lines 3-11] “The predetermined path geometry may be defined according to lane boundaries 56 that are spaced by the calculated width from a center line 44. The predetermined path geometry may be defined differently for each section of roadway as demarcated between shape points 42. The shape points 42 may be determined based on the turns in the road. When a road turns more than a set angle, or has more than a predetermined curvature, a new shape point 42 is added to define a new road section.” wherein a target point is predetermined path geometry) (Soni - [Col. 11, lines 15-22] “FIG. 13 illustrates a sliding section technique. In FIG. 13, a sliding window 81 is moved across the image patch 61. Rather than define entire sections of roadway, the predetermined path geometry for training the lane detection model 40 is limited to the sliding window 81. The sliding section approach improves the speed of training the lane detection model 40 because a smaller portion of the image patch 61 is analyzed.” wherein the sliding window area is the image patch 61) is greater than or equal to a preset threshold value and is a local maximum value (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a preset threshold value and a local maximum value are local maxima of the response map with confidence values above a threshold) (Maheshwari - [0100] “The lane detection module 104 may compare the measured distances (643, 644) to a threshold distance, and the threshold distance may be predetermined or dynamically determined as part of a machine learning model. If the lane detection module 104 determines that the distance 643 associated with the first pixel 641 is less than the threshold distance, the lane detection module 104 may determine that the first pixel 641 should be linked in a subsection 645a with the beginning pixel 637 and any other pixels in between that satisfy the distance threshold. By contrast, if the lane detection module 104 determines that the distance 644 associated with the second pixel 642 is greater than the threshold distance, the lane detection module 104 may include the second pixel 642 in a different subsection 645b than the first pixel 641.”) of the sliding window area (Soni - [Col. 11, lines 15-22] “FIG. 13 illustrates a sliding section technique. In FIG. 13, a sliding window 81 is moved across the image patch 61. Rather than define entire sections of roadway, the predetermined path geometry for training the lane detection model 40 is limited to the sliding window 81. The sliding section approach improves the speed of training the lane detection model 40 because a smaller portion of the image patch 61 is analyzed.” wherein the sliding window area is the image patch 61).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine a sliding window operation, of Soni, with the traffic marker detection model, of Eagelberg in view of Zhu, to improve the speed of training the model by using a smaller portion of the image during analysis.
Regarding claim 4, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 2, wherein the traffic marker comprises a plurality of types of traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like), and the obtaining the response map (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.”) corresponding to the target image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) comprises: obtaining a plurality of sub-response maps (Zhu - [0061] “a low-cost variance check is performed at all spatial locations before computing response maps to rule out low variance areas” wherein a plurality of sub-response maps are computed response maps) corresponding to the plurality of types of traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like), wherein the response value of each position (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) in each of the plurality of sub-response maps (Zhu - [0061] “a low-cost variance check is performed at all spatial locations before computing response maps to rule out low variance areas” wherein a plurality of sub-response maps are computed response maps) is used to characterize the confidence level (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a confidence level is confidence values) that a corresponding type of traffic sign element exists at the corresponding position (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 2.
Regarding claim 5, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 4, wherein the detection mark comprises a plurality of types of identifications (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a detection mark is a lane mark characteristic), and the generating the detection mark (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein the detection mark is a lane mark characteristic) based on the response value in the response map (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) comprises:
for each of the plurality of sub-response maps (Zhu - [0061] “a low-cost variance check is performed at all spatial locations before computing response maps to rule out low variance areas” wherein a plurality of sub-response maps are computed response maps), generating a corresponding type of identification for the corresponding type of traffic sign element (Eagelberg - [0179] “processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature. Alternatively or in addition, the reference point may include another landmark (e.g., a tree, building, bridge, overpass, parked vehicle, body of water, etc.).” wherein detection mark identification is one or more lane mark characteristics of the at least one identified mark for the corresponding set of features associated with the reference point; wherein traffic sign element is the set of features associated with the reference point) based on the response value (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) in the sub-response map (Zhu - [0061] “a low-cost variance check is performed at all spatial locations before computing response maps to rule out low variance areas” wherein a plurality of sub-response maps are computed response maps).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 2.
Regarding claim 6, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claims 1, wherein the target image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) is obtained by an image acquisition device on a vehicle (Eagelberg - [0049] “In some embodiments, image acquisition unit 120 may include one or more image capture devices (e.g., cameras, CCDs, or any other type of image sensor), such as image capture device 122, image capture device 124, and image capture device 126.” wherein an image acquisition device on a vehicle is one or more image capture devices), and the detection mark (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein the detection mark is a lane mark characteristic) is located on a side of the traffic marker adjacent to the image acquisition device (Eagelberg - [0118] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, hazardous objects, and any other feature associated with an environment of a vehicle.” wherein a side of the traffic marker adjacent to the vehicle is a set of features associated with an environment of a vehicle; it is obvious to one of ordinary skill in the art that the detection mark is located on a side of the traffic marker adjacent to the vehicle to ensure the traffic marker is being detected in reference to the vehicle).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 7, Eagelberg teaches a training method for a traffic marker detection model ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics), comprising:
acquiring a training image ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) and a labeled image ([0197] “The neural network may be trained using labeled images and/or lane mark identifiers stored in association with one or more images. Labeled images or images in associated with labels and/or lane mark identifiers may be received by the neural network from a variety of resources, such as one or more databases.”) corresponding to the training image, wherein the training image comprises a traffic marker ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.);
the labeled image ([0197] “The neural network may be trained using labeled images and/or lane mark identifiers stored in association with one or more images. Labeled images or images in associated with labels and/or lane mark identifiers may be received by the neural network from a variety of resources, such as one or more databases.”) comprises a labeled mark ([0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) corresponding to the traffic marker in the training image ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.);
the labeled mark is used for marking a position ([0241] “An example labeled image 1406 for use by, for example, a neural network is shown in FIG. 14B. As shown, labeled image 1406 includes a representation of road 1402 with the lane merge, and lane marks 1404A-1404E have been labeled, as indicated by the differing line patterns. For example, the dot-dash lines on lane marks 1404A, 1404B, and 1404E may indicate “regular” lane marks that do not form part of the lane merge, the small-dash line on lane mark 1404C may indicate the inner portion of a merge lane, and the solid line on lane mark 1404D may indicate the outer portion of a merge lane.” wherein the labeled mark is lane marks that have been labeled for the position of the lanes) of the traffic marker in the corresponding training image ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.);
the labeled mark comprises a labeled point and a labeled line ([0241] “An example labeled image 1406 for use by, for example, a neural network is shown in FIG. 14B. As shown, labeled image 1406 includes a representation of road 1402 with the lane merge, and lane marks 1404A-1404E have been labeled, as indicated by the differing line patterns. For example, the dot-dash lines on lane marks 1404A, 1404B, and 1404E may indicate “regular” lane marks that do not form part of the lane merge, the small-dash line on lane mark 1404C may indicate the inner portion of a merge lane, and the solid line on lane mark 1404D may indicate the outer portion of a merge lane.” wherein the labeled mark is lane marks that have been labeled and a labeled line is the small-dash lines and solid lines), wherein the traffic marker comprises a plurality of traffic sign elements, the labeled line is determined ([0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like) ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein the detection mark is a lane mark characteristic and a detection point is a reference point) in response to determining that a ([0241] “An example labeled image 1406 for use by, for example, a neural network is shown in FIG. 14B. As shown, labeled image 1406 includes a representation of road 1402 with the lane merge, and lane marks 1404A-1404E have been labeled, as indicated by the differing line patterns. For example, the dot-dash lines on lane marks 1404A, 1404B, and 1404E may indicate “regular” lane marks that do not form part of the lane merge, the small-dash line on lane mark 1404C may indicate the inner portion of a merge lane, and the solid line on lane mark 1404D may indicate the outer portion of a merge lane.” wherein the labeled point is lane marks that have been labeled);
inputting the training image ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) into a traffic marker detection model ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics) to obtain([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.); and
training the traffic marker detection model ([0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics) based on ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) and the labeled mark ([0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) of the corresponding labeled image ([0197] “The neural network may be trained using labeled images and/or lane mark identifiers stored in association with one or more images. Labeled images or images in associated with labels and/or lane mark identifiers may be received by the neural network from a variety of resources, such as one or more databases.”);
wherein the training image is obtained by an image acquisition device on a vehicle, and the labeled mark is located on a side of the traffic marker adjacent to the image acquisition device ([0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.”) ([0049] “In some embodiments, image acquisition unit 120 may include one or more image capture devices (e.g., cameras, CCDs, or any other type of image sensor), such as image capture device 122, image capture device 124, and image capture device 126.” wherein an image acquisition device on a vehicle is one or more image capture devices) ([0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) ([0118] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, hazardous objects, and any other feature associated with an environment of a vehicle.” wherein a side of the traffic marker adjacent to the vehicle is a set of features associated with an environment of a vehicle; it is obvious to one of ordinary skill in the art that the detection mark is located on a side of the traffic marker adjacent to the vehicle to ensure the traffic marker is being detected in reference to the vehicle).
Eagelberg does not specifically disclose a preset threshold value and a predicted mark.
However, Zhu teaches a preset threshold value ([0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a preset threshold value is a threshold) and a predicted mark ([0042] “In accordance with the present invention, the boosted training method 204 identifies the features that best separate vehicle patterns from non-vehicle patterns. These features are chosen to define the classifiers 206. Based on the feature values, the classifier outputs a confidence score that indicates the confidence in the pattern being a vehicle.” wherein a predicted mark is the identified features that define the classifiers which output a confidence score).
Further, Eagelberg in view of Zhu does not specifically disclose a pixel distance.
However, Soni teaches a pixel distance ([Col.7, lines 1-8] “The road network module 37 may calculate the predetermined path geometry based on a predetermined distance. The road network module 37 may access the predetermined distance as set by user input. The road network module 37 may access a centerline or a path boundary for at least one path from the map data 31. The road network module 37 calculates the predetermined path geometry based on a predetermined distance to the centerline or path boundary.” wherein a pixel distance is a predetermined distance) ([Col.6, lines 38-40] “For example, the width of a road in the road network has a geographic distance, which is converted to a number of pixels in the image data 33 using the scaling factor.” wherein geographic distance is converted to pixels in image data).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Eagelberg in view of Zhu and Soni does not specifically disclose determining that a pixel distance between any two adjacent traffic sign elements along an extending direction of the road is less than or larger than a preset threshold value.
However, Maheshwari teaches a pixel distance between any two adjacent traffic sign elements along an extending direction of the road is less than or larger than a preset threshold value ([0100] “The lane detection module 104 may compare the measured distances (643, 644) to a threshold distance, and the threshold distance may be predetermined or dynamically determined as part of a machine learning model. If the lane detection module 104 determines that the distance 643 associated with the first pixel 641 is less than the threshold distance, the lane detection module 104 may determine that the first pixel 641 should be linked in a subsection 645a with the beginning pixel 637 and any other pixels in between that satisfy the distance threshold. By contrast, if the lane detection module 104 determines that the distance 644 associated with the second pixel 642 is greater than the threshold distance, the lane detection module 104 may include the second pixel 642 in a different subsection 645b than the first pixel 641.”).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 8, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 7, wherein the traffic marker comprises a plurality of first traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a traffic marker comprising first traffic sign elements are a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like), the pixel distance between any two adjacent (Eagelberg - [0179] “For instance, referring to FIG. 10A, the detected lane mark characteristic(s) may include a distance between lane marks 1004A and 1004C or between lane marks 1004B and 1004C.” wherein a pixel distance is a distance between lane marks; It is obvious to one of ordinary skill in the art that the distance of Eagelberg, is a pixel distance because pixels comprise the images that are being processed in the method disclosed by Eagelberg) first traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein first traffic sign elements are lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like) of the plurality of first traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the plurality of first traffic sign elements are lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like) is greater than or equal to a preset pixel distance (Maheshwari - [0100] “The lane detection module 104 may compare the measured distances (643, 644) to a threshold distance, and the threshold distance may be predetermined or dynamically determined as part of a machine learning model. If the lane detection module 104 determines that the distance 643 associated with the first pixel 641 is less than the threshold distance, the lane detection module 104 may determine that the first pixel 641 should be linked in a subsection 645a with the beginning pixel 637 and any other pixels in between that satisfy the distance threshold. By contrast, if the lane detection module 104 determines that the distance 644 associated with the second pixel 642 is greater than the threshold distance, the lane detection module 104 may include the second pixel 642 in a different subsection 645b than the first pixel 641.”), and the labeled mark (Eagelberg - [0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) comprises a labeled point (Eagelberg - [0241] “An example labeled image 1406 for use by, for example, a neural network is shown in FIG. 14B. As shown, labeled image 1406 includes a representation of road 1402 with the lane merge, and lane marks 1404A-1404E have been labeled, as indicated by the differing line patterns. For example, the dot-dash lines on lane marks 1404A, 1404B, and 1404E may indicate “regular” lane marks that do not form part of the lane merge, the small-dash line on lane mark 1404C may indicate the inner portion of a merge lane, and the solid line on lane mark 1404D may indicate the outer portion of a merge lane.” wherein the labeled mark is lane marks that have been labeled and a labeled line is the small-dash lines and solid lines) corresponding to each first traffic sign element (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein first traffic sign elements are lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 9, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 7, wherein the traffic marker comprises a plurality of second traffic sign elements (Eagelberg - [0120] “As described in connection with FIG. 6 below, stereo image analysis module 404 may include instructions for detecting a set of features within the first and second sets of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, hazardous objects, and the like.” wherein a plurality of second traffic sign elements are a set of features within the second set of images), the pixel distance between any two adjacent (Eagelberg - [0179] “For instance, referring to FIG. 10A, the detected lane mark characteristic(s) may include a distance between lane marks 1004A and 1004C or between lane marks 1004B and 1004C.” wherein a pixel distance is a distance between lane marks; It is obvious to one of ordinary skill in the art that the distance of Eagelberg, is a pixel distance because pixels comprise the images that are being processed in the method disclosed by Eagelberg) second traffic sign elements of the plurality of second traffic sign elements (Eagelberg - [0120] “As described in connection with FIG. 6 below, stereo image analysis module 404 may include instructions for detecting a set of features within the first and second sets of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, hazardous objects, and the like.” wherein a plurality of second traffic sign elements are a set of features within the second set of images) is less than the preset pixel distance (Maheshwari - [0100] “The lane detection module 104 may compare the measured distances (643, 644) to a threshold distance, and the threshold distance may be predetermined or dynamically determined as part of a machine learning model. If the lane detection module 104 determines that the distance 643 associated with the first pixel 641 is less than the threshold distance, the lane detection module 104 may determine that the first pixel 641 should be linked in a subsection 645a with the beginning pixel 637 and any other pixels in between that satisfy the distance threshold. By contrast, if the lane detection module 104 determines that the distance 644 associated with the second pixel 642 is greater than the threshold distance, the lane detection module 104 may include the second pixel 642 in a different subsection 645b than the first pixel 641.”), and the labeled mark (Eagelberg - [0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) comprises labeled lines corresponding (Eagelberg - [0241] “An example labeled image 1406 for use by, for example, a neural network is shown in FIG. 14B. As shown, labeled image 1406 includes a representation of road 1402 with the lane merge, and lane marks 1404A-1404E have been labeled, as indicated by the differing line patterns. For example, the dot-dash lines on lane marks 1404A, 1404B, and 1404E may indicate “regular” lane marks that do not form part of the lane merge, the small-dash line on lane mark 1404C may indicate the inner portion of a merge lane, and the solid line on lane mark 1404D may indicate the outer portion of a merge lane.” wherein the labeled mark is lane marks that have been labeled and a labeled line is the small-dash lines and solid lines) to the plurality of second traffic sign elements (Eagelberg - [0120] “As described in connection with FIG. 6 below, stereo image analysis module 404 may include instructions for detecting a set of features within the first and second sets of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, hazardous objects, and the like.” wherein a plurality of second traffic sign elements are a set of features within the second set of images).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 10, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 7, wherein
the inputting the training image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) into the traffic marker detection model (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics) to obtain the predicted mark (Zhu - [0042] “In accordance with the present invention, the boosted training method 204 identifies the features that best separate vehicle patterns from non-vehicle patterns. These features are chosen to define the classifiers 206. Based on the feature values, the classifier outputs a confidence score that indicates the confidence in the pattern being a vehicle.” wherein a predicted mark is the identified features that define the classifiers which output a confidence score) of the traffic marker (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like) in the training image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) comprises:
obtaining by the traffic marker detection model (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics) a predicted response map (Zhu - [0060] “Principal component detection is performed in the following way: at every spatial location, a confidence response is computed through component detection. The corresponding component classifier is used to evaluate the response. Thus, a complete response map is obtained for the principal component.” wherein a predicted response map is a response map obtained for the principal component) corresponding to the training image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.), wherein the predicted response map comprises predicted response values of various positions (Zhu - [0060] “Principal component detection is performed in the following way: at every spatial location, a confidence response is computed through component detection. The corresponding component classifier is used to evaluate the response. Thus, a complete response map is obtained for the principal component.” wherein a predicted response map is a response map obtained for the principal component at every spatial location); and wherein
training the traffic marker detection model (Eagelberg- [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics) based on the predicted mark (Zhu - [0042] “In accordance with the present invention, the boosted training method 204 identifies the features that best separate vehicle patterns from non-vehicle patterns. These features are chosen to define the classifiers 206. Based on the feature values, the classifier outputs a confidence score that indicates the confidence in the pattern being a vehicle.” wherein a predicted mark is the identified features that define the classifiers which output a confidence score) of the training image (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image is the set of images with a set of features such as lane markings, vehicles, road signs, etc.) and the labeled mark (Eagelberg - [0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) of the labeled image ([0197] “The neural network may be trained using labeled images and/or lane mark identifiers stored in association with one or more images. Labeled images or images in associated with labels and/or lane mark identifiers may be received by the neural network from a variety of resources, such as one or more databases.”) comprises:
obtaining a truth-value response map (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a corresponding truth-value response map is a response map with confidence values above a threshold) of the labeled image (Eagelberg - [0197] “The neural network may be trained using labeled images and/or lane mark identifiers stored in association with one or more images. Labeled images or images in associated with labels and/or lane mark identifiers may be received by the neural network from a variety of resources, such as one or more databases.”), the truth- value response map comprising truth-value response values of various positions (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a corresponding truth-value response map is a response map with confidence values above a threshold; truth-value response values of various positions are confidence values above a threshold at corresponding locations); and
training the traffic marker detection model (Eagelberg - [0179] “Returning to FIG. 9, at step 908 processing unit 110 may detect one or more lane mark characteristics of the at least one identified mark. The detected lane mark characteristic(s) may include, for example, a distance of the at least one identified lane mark to a reference point. The reference point may include, for example, a sign, lamppost, traffic signal, curb, or other road feature.” wherein a traffic marker detection model is a processing unit that detects one or more lane mark characteristics) based on the predicted response value of the predicted response map (Zhu - [0060] “Principal component detection is performed in the following way: at every spatial location, a confidence response is computed through component detection. The corresponding component classifier is used to evaluate the response. Thus, a complete response map is obtained for the principal component.” wherein a predicted response map is a response map obtained for the principal component at every spatial location) and the truth-value response value of the truth-value response map (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a corresponding truth-value response map is a response map with confidence values above a threshold; truth-value response values of various positions are confidence values above a threshold at corresponding locations).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 11, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 10, wherein, in the truth-value response map,
the truth-value response value at each position within a predetermined range (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a corresponding truth-value response map is a response map with confidence values above a threshold; truth-value response values of various positions are confidence values above a threshold at corresponding locations; the predetermined range is the values above a threshold) of the labeled mark (Eagelberg - [0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) exhibits a predetermined distribution; or,
the truth-value response value (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a truth-value response values of various positions are confidence values above a threshold at corresponding locations) at the labeled mark (Eagelberg - [0197] “The labeling of the stored images be done by, for example, storing a label (e.g., a merge or split label) in associated with an identified lane mark.” wherein a labeled mark is a label associated with an identified lane mark) is a first value, and the truth-value response value (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a truth-value response values of various positions are confidence values above a threshold at corresponding locations) at a non-labeled mark is a second value different from the first value (Zhu - [0037] “At time t, L.sub.t gives an accumulated confidence score of how likely the trajectory under detection is caused by a vehicle. If this accumulated score is sufficiently high, the confidence level is high enough to claim that a vehicle appears in the sequences. If the accumulated score is sufficiently low, it is very likely that the current trajectory is not caused by a vehicle and can be safely discarded.” wherein a first value is a high confidence score and a second value for a non-labeled mark is a low confidence score).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 12, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 10, wherein the traffic marker comprises a plurality of types of traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like), and wherein the obtaining the truth-value response map corresponding to the labeled image (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a corresponding truth-value response map is a response map with confidence values above a threshold; truth-value response values of various positions are confidence values above a threshold at corresponding locations) comprises:
obtaining a plurality of truth-value sub-response maps (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a corresponding truth-value response map is a response map with confidence values above a threshold; truth-value response values of various positions are confidence values above a threshold at corresponding locations; It is obvious to one of ordinary skill in the art to obtain a plurality of truth-value sub-response maps because said maps are generated for a plurality of corresponding locations of Zhu) corresponding to the plurality of types of traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like), wherein, for each truth-value sub-response map of the plurality of truth-value sub-response maps, the truth-value response value of each position in the truth-value sub-response map (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a corresponding truth-value response map is a response map with confidence values above a threshold; truth-value response values of various positions are confidence values above a threshold at corresponding locations; It is obvious to one of ordinary skill in the art to obtain a plurality of truth-value sub-response maps because said maps are generated for a plurality of corresponding locations of Zhu) is used to characterize a confidence level (Zhu - [0060] “Values in the response map indicate the confidence that a principal component appears at the corresponding locations. Local maxima of the response map with confidence values above a threshold are considered potential candidates for the principal components.” wherein a confidence level is confidence values) that a corresponding type of traffic sign element exists at a respective position (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 16, the claim recites similar limitations to claim 1 but in the form of an electronic apparatus. Therefore, claim 16 recites similar limitations to claim 1 and is rejected for similar rationale and reasoning (see the analysis for claim 1 above).
Regarding claim 17, Eagelberg in view of Zhu, Soni, and Maheshwari teaches a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used for enabling the computer to perform a method of claim 1 (Eagelberg - [0010] “Consistent with other disclosed embodiments, non-transitory computer-readable storage media may store program instructions, which are executed by at least one processing device and perform any of the methods described herein.”).
Regarding claim 20, the claim recites similar limitations to claim 7 but in the form of an electronic apparatus. Therefore, claim 16 recites similar limitations to claim 7 and is rejected for similar rationale and reasoning (see the analysis for claim 7 above).
Regarding claim 21, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the electronic apparatus of claim 16, wherein the plurality of traffic sign elements comprise at least one of a traffic cone, a water horse, a utility pole, a lane line, or a tree (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like) (Eagelberg - [0179] “Alternatively or in addition, the reference point may include another landmark (e.g., a tree, building, bridge, overpass, parked vehicle, body of water, etc.).”).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 22, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the electronic apparatus of claim 16, wherein the traffic marker comprises different types of traffic sign elements (Eagelberg - [0124] “a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein the traffic marker is a set of features such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like), and different types of traffic sign elements have multiple color types or shape types of detection marks (Eagelberg - [0186] “For example, a lane mark characteristic may include one or more of a size, shape, and color of a lane mark.”).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 1.
Regarding claim 23, Eagelberg in view of Zhu, Soni, and Maheshwari teaches the method of claim 2, wherein the response map corresponds to a region in the target image where the traffic markers are located (Zhu - [0060] “Thus, a complete response map is obtained for the principal component. Values in the response map indicate the confidence that a principal component appears at the corresponding locations.” wherein a response value is values in the response map at the corresponding locations) (Eagelberg - [0124] “By performing the analysis, processing unit 110 may detect a set of features within the set of images, such as lane markings, vehicles, pedestrians, road signs, highway exit ramps, traffic lights, and the like.” wherein a target image containing a traffic marker is the set of images with a set of features such as lane markings, vehicles, road signs, etc.).
The motivation for combining Eagelberg, Zhu, Soni, and Maheshwari is the same motivation as used for claim 2.
Conclusion
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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AMANDA PEARSON whose telephone number is (703)-756-5786. The examiner can normally be reached Monday - Friday 8:00 - 5:30.
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, Emily Terrell can be reached on (571)- 270-3717. 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.
/AMANDA H PEARSON/Examiner, Art Unit 2666
/MING Y HON/Primary Examiner, Art Unit 2666