DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Status
Claims 1-8 are pending for examination in the application filed 09/04/2025. Claims 1-7 have been amended.
Priority
Acknowledgement is made of Applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d). The certified copies have been filed in parent application JP2021-028714 filed on 02/25/2021 and parent application JP2022-011323 filed on 01/27/2022.
Response to Arguments and Amendments
The 35 U.S.C. 112(f) interpretation of claims 1-8 has been withdrawn in light of the amendments.
Applicant's arguments filed 09/04/2025 have been fully considered but they are not persuasive. Applicant argues on pages 6-9 of the Remarks filed 09/04/2025 that Baek in view of Li and Assarsson do not teach the amended limitation “estimating, using stereo matching, three-dimensional positions of the surrounding points set and whose three-dimensional positions have not been estimated”. Examiner respectfully disagrees. As cited in the Non-Final Office Action,
Assarsson, in the same field of endeavor of stereo matching, teaches a second estimation unit configured to estimate, using stereo matching, three- dimensional positions of the surrounding points set by the surrounding point setting unit ([0076] The raw stereo matching is noisy, and one may want to enforce that a continuous surface is reconstructed. This can be done in a regularization step, where the current view ray is intersected with planes constructed from each of the neighboring points and normals, see FIG. 6. An edge-preserving weighted average of these intersections is computed as a new position along the ray. [0084] The next levels of the hierarchy are executed in the same manner, doubling the resolution (see above) for each iteration until the highest hierarchical level has finished. A few differences from the first hierarchical level should however be noted. There are now estimated normals in each step, and the filters used for stereo matching are thus oriented in the corresponding direction).
In the Non-Final Office Action, this limitation was mapped to Fig. 10 steps a-c, which include ray marching, silhouette bounded, and regularization. The steps following ray marching serve to refine the estimation of the 3D positions of the points. Now, with the newly added amendments, Assarsson teaches estimating, using stereo matching, three-dimensional positions of the surrounding points set and whose three-dimensional positions have not been estimated, as shown in Fig. 10 a) ray marching, as the three-dimensional positions of the surrounding points are estimated for the first time in this step. Assarsson explains: [0067] stereo matching is done by raymarching view rays from one camera in each pair and storing the results as view-space geometry buffers. The matching is divided both into multiple steps and in a hierarchical fashion, where the resolution is doubled in each dimension for each level of the hierarchy (see FIG. 1). [0069] In the first ray-marching step, a resolution N times lower than the defined geometric resolution is started with. I.e., a view ray is followed for each 2.sup.Nth vertical and 2.sup.Nth horizontal pixel for one of the cameras in each pair. Marching is performed along the ray for a fixed length starting from the conservative estimate of the geometry constructed as described above. For each step along the ray, a rectangular filter is constructed in world space and projected on both cameras. Colors are sampled with mipmapping in the corresponding color textures, and the distance between both patches is computed using a mean-subtracted Sum of Absolute Difference (SAD) cost function. Thus, in the raw stereo matching step of Assarsson, a selected pixel and corresponding pixel are used for first estimating and surrounding pixels are used for second estimating: [0020] For each selected pixel in the first image a corresponding pixel in the other, second, image of the pair of images may be determined. A corresponding pixel may mean that the two pixels picture about the same point for the same object. In the ray marching, a stepwise marching along a ray from the first camera's position in 3D space through the pixel in the first image into the virtual 3D scene. For each step one is at a certain 3D position, p, along the ray. This position p may then be projected on the second image and the pixels from the first and the second images are compared whether the same point that is intended to be 3D reconstructed is pictured. Also pixels around the identified pixels may be used in the comparison. The comparison may result in a score for how good the matching between the two pixels is. Therefore, the 35 U.S.C. § 103 rejections of claims 1-8 are maintained.
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 1-2, 4, and 7-8 are rejected under 35 U.S.C. 103 as being unpatentable over Baek (KR20170065893A) in view of Li (Li, Dawei, et al. "Multi-scale neighborhood feature extraction and aggregation for point cloud segmentation." IEEE Transactions on Circuits and Systems for Video Technology 31.6 (2020): 2175-2191) and Assarsson (US20240062463A1).
Regarding claim 1, Baek teaches an information processing apparatus (parking assistance apparatus 200) comprising: one or more processors; and at least one memory storing executable instructions ([145] the parking assisting device 200 may include an interface unit 210, a memory 220, a processor 230, and a power supply unit 240. [150] Such memory 220 may include at least one of various hardware storage media such as ROM, RAM, EPROM, flash drive, hard drive, etc. to store the various data. [151] The processor 230 can control the overall operation of each component included in the parking assist device 200), which when executed by the one or more processors, cause the information processing apparatus to perform a method comprising: acquiring a first image from a first optical system (camera 310) and a second image from a second optical system (camera 320), the first image and the second image being acquired from an imaging apparatus that includes the first optical system and the second optical system ([85] Such a camera 161 can acquire a stereo image with respect to the front of the vehicle from the first and second cameras 310 and 320), which are arranged in a device so that an imaging field of view of the first optical system is at least partially overlapped with an imaging field of view of the second optical system ([85] Such a camera 161 can acquire a stereo image with respect to the front of the vehicle from the first and second cameras 310 and 320. (E.g., a pedestrian, a traffic light, a road, a lane, another vehicle) appearing in at least one stereo image based on the disparity information, based on the stereo image, After the object is detected, the movement of the object can be continuously tracked);
performing stereo matching of feature points ([108] The disparity calculator 520 receives the image signal processed by the image preprocessing unit 510, performs stereo matching on the received images) of a first number that is smaller than a number of pixels in the first image in the first image and the second image ([109] At this time, the stereo matching may be performed on a pixel-by-pixel basis of the stereo images or on a predetermined block basis) to estimate three-dimensional positions of the feature points with respect to the imaging apparatus ([122] sequentially, the object in the obtained stereo images is identified, the motion or motion vector of the identified object is calculated, and the motion of the object or the like is tracked based on the calculated motion or motion vector);
setting the feature point determined to be acquired from a three-dimensional space, in which an object to be detected is located (parking zone), set in a field of view of the imaging apparatus based on the three-dimensional position, among the feature points of the first number, as a target point ([211] specific feature point) ([157] The processor 230 may classify at least some of the plurality of feature points into a plurality of parking lines having a straight line shape. At this time, the processor can classify the plurality of feature points into the plurality of parking lines by using the predetermined clustering technique. [211] For example, the processor 230 uses the DBSCAN (Density Based Spatial Clustering of Application with Noise) to calculate the number of neighboring feature points within a unit area epsilon (e.g., a circle of a predetermined radius) And when the predetermined number of neighboring feature points are included, the specific feature point may be labeled as a core object);
setting surrounding points of a second number ([211] For example, the processor 230 uses the DBSCAN (Density Based Spatial Clustering of Application with Noise) to calculate the number of neighboring feature points within a unit area epsilon (e.g., a circle of a predetermined radius));
determining the target point is the feature point indicating a feature of an object existing in the three-dimensional space based on a number of surrounding points whose distance between the three-dimensional position of the target point estimated and the three- dimensional positions of the surrounding points estimated is within a predetermined distance exceeding a threshold, and determine the target point is noise based on the number of surrounding points being at or below the threshold ([211] For example, the processor 230 uses the DBSCAN (Density Based Spatial Clustering of Application with Noise) to calculate the number of neighboring feature points within a unit area epsilon (e.g., a circle of a predetermined radius) And when the predetermined number of neighboring feature points are included, the specific feature point may be labeled as a core object, and in the opposite case, it may be labeled as a noise object).
Baek does not teach that a density of the surrounding points is higher than the feature points for which the three-dimensional positions are estimated in an image region between the target point and the feature points estimated; estimating, using stereo matching, three- dimensional positions of the surrounding points set and whose three-dimensional positions have not been estimated.
Li, in the same field of endeavor of point cloud target detection, teaches that a density of the surrounding points (dots in Fig. 3) is higher than the feature points (x's in Fig. 3) for which the three-dimensional positions are estimated in an image region between the target point and the feature points estimated ([pg. 2177 para. 3] The proposed Multi-scale Neighborhood Feature Extraction and Aggregation Model (MNFEAM) contains two modules—the Multi-scale Feature Extraction Module (MFEM) and the Locality Feature Aggregation Module (LFAM). The main purpose of our model is to effectively learn high-level feature representations from the original world space features (such as XYZ, colors, and normals). The overview of MNFEAM is shown in Fig. 1. In Section III-A, we first introduce the MFEM part, in which we identify k neighboring points for each point at three different radius-defined scales, respectively. Upon the three neighborhood groups, we extract the multi-scale local features by passing through several MLPs in order to expand the receptive field and improve the abstraction of local information. [pg. 2179 para. 3] We identify n-nearest neighbors for each point in the multi-scale locality feature space (embedding) by L1-distance. For each point in the embedding a local feature grouping of n×m is formed).
PNG
media_image1.png
524
868
media_image1.png
Greyscale
Therefore, it would have been obvious to a person of ordinary skill in the art at the time that the invention was made to modify the apparatus of Baek with the teachings of Li to use surrounding points with a higher density than the feature points because "a single point in a 3D scene does not carry much useful information, and it only makes sense when associated with its neighboring points. Therefore, the first and the most crucial step of point cloud learning is to define the neighbors of each point" [Li pg. 2177 para. 4].
Assarsson, in the same field of endeavor of stereo matching, teaches estimating, using stereo matching, three- dimensional positions of the surrounding points set and whose three-dimensional positions have not been estimated ([0020] For each selected pixel in the first image a corresponding pixel in the other, second, image of the pair of images may be determined. A corresponding pixel may mean that the two pixels picture about the same point for the same object. In the ray marching, a stepwise marching along a ray from the first camera's position in 3D space through the pixel in the first image into the virtual 3D scene. For each step one is at a certain 3D position, p, along the ray. This position p may then be projected on the second image and the pixels from the first and the second images are compared whether the same point that is intended to be 3D reconstructed is pictured. Also pixels around the identified pixels may be used in the comparison. The comparison may result in a score for how good the matching between the two pixels is. [0067] stereo matching is done by raymarching view rays from one camera in each pair and storing the results as view-space geometry buffers. The matching is divided both into multiple steps and in a hierarchical fashion, where the resolution is doubled in each dimension for each level of the hierarchy (see FIG. 1). [0069] In the first ray-marching step, a resolution N times lower than the defined geometric resolution is started with. I.e., a view ray is followed for each 2.sup.Nth vertical and 2.sup.Nth horizontal pixel for one of the cameras in each pair. Marching is performed along the ray for a fixed length starting from the conservative estimate of the geometry constructed as described above. For each step along the ray, a rectangular filter is constructed in world space and projected on both cameras. Colors are sampled with mipmapping in the corresponding color textures, and the distance between both patches is computed using a mean-subtracted Sum of Absolute Difference (SAD) cost function).
Therefore, it would have been obvious to a person of ordinary skill in the art at the time that the invention was made to modify the apparatus of Baek with the teachings of Assarsson to perform stereo matching of the surrounding points for "real-time generation of 3D imaging that [is] fast and computational efficient while providing acceptable image quality" and "to be able to easily estimate and utilize normal information, stereo matching is done by raymarching view rays from one camera in each pair and storing the results as view-space geometry buffers" [Assarsson 0067].
Regarding claim 2, Baek, Li, and Assarsson teach the apparatus of claim 1. Baek further teaches wherein the determining determines whether the target point is the feature point indicating a feature of an object existing in the three-dimensional space based on a number of the surrounding points having the values of distances from the imaging apparatus to positions indicated by the surrounding points, which are similar to the value of a distance from the imaging apparatus to a position indicated by the target point ([211] For example, the processor 230 uses the DBSCAN (Density Based Spatial Clustering of Application with Noise) to calculate the number of neighboring feature points within a unit area epsilon (e.g., a circle of a predetermined radius) And when the predetermined number of neighboring feature points are included, the specific feature point may be labeled as a core object, and in the opposite case, it may be labeled as a noise object. If the specific feature point is a core object, the process of expanding the cluster is repeated depending on whether other feature points in the unit area are also labeled as a core object by the same method, and all the detected feature points are grouped into at least one. [158] For example, based on DBSCAN (Density Based Spatial Clustering of Application with Noise), the processor 230 may consider at least some of the plurality of feature points as a plurality of parking It can be classified into lines. [215] The processor 230 may calculate a straight line equation for each of the sorted first to seventh parking lines 1011 to 1017 and calculate an intersection J between the calculated straight line equations. In addition, the processor 230 can determine whether the distance between two intersections J adjacent to each other among the calculated intersections J satisfies the specification of the predetermined parking zone. If the distance between the two intersections J satisfies the specification of the predetermined parking zone, a line connecting the two intersection points can be set as the entry line of the vehicle 100), or a ratio of the number of the surrounding points having the values of the distances from the imaging apparatus to the positions indicated by the surrounding points, which are similar to the value of the distance from the imaging apparatus to the position indicated by the target point, with respect to the number of the surrounding points.
Regarding claim 4, Baek, Li, and Assarsson teach the apparatus of claim 1. Baek further teaches wherein setting the surrounding points sets the surrounding points in the first image such that the surrounding points are set within a predetermined spacing around the target point ([158] For example, based on DBSCAN (Density Based Spatial Clustering of Application with Noise), the processor 230 may consider at least some of the plurality of feature points as a plurality of parking It can be classified into lines. Accordingly, even if two contour lines having a slope difference equal to or greater than a predetermined value are connected to each other, the processor can classify the two contour lines into different parking lines. Any one of the plurality of classified parking lines may be spaced apart from the rest of the parking lines by a predetermined distance or inclined by a predetermined angle or more).
Regarding claim 7, Baek teaches an information processing method ([127] 6A and 6B are diagrams for explaining the operation method of the controller 170 of FIG. 5 based on the stereo image obtained in the first and second frame periods, respectively) comprising:
first estimating a three-dimensional position of a feature point, which includes a distance in a real space from a certain position of an imaging apparatus to a position indicated by the feature point, using a result of stereo matching ([108] The disparity calculator 520 receives the image signal processed by the image preprocessing unit 510, performs stereo matching on the received images. [122] sequentially, the object in the obtained stereo images is identified, the motion or motion vector of the identified object is calculated, and the motion of the object or the like is tracked based on the calculated motion or motion vector) of a first image and a second image acquired from the imaging apparatus including a first optical system and a second optical system ([85] Such a camera 161 can acquire a stereo image with respect to the front of the vehicle from the first and second cameras 310 and 320), which are arranged in a device so that an imaging field of view of the first optical system is at least partially overlapped with an imaging field of view of the second optical system ([85] Such a camera 161 can acquire a stereo image with respect to the front of the vehicle from the first and second cameras 310 and 320. (E.g., a pedestrian, a traffic light, a road, a lane, another vehicle) appearing in at least one stereo image based on the disparity information, based on the stereo image, After the object is detected, the movement of the object can be continuously tracked), the stereo matching being performed at the feature points of a first number smaller than a number of pixels in the first image ([109] At this time, the stereo matching may be performed on a pixel-by-pixel basis of the stereo images or on a predetermined block basis);
setting the feature point determined to be acquired from a three-dimensional space, in which an object to be detected is located (parking zone), set in a field of view of the imaging apparatus based on the three-dimensional position, among the feature points of the first number, as a target point ([211] specific feature point) ([157] The processor 230 may classify at least some of the plurality of feature points into a plurality of parking lines having a straight line shape. At this time, the processor can classify the plurality of feature points into the plurality of parking lines by using the predetermined clustering technique. [211] For example, the processor 230 uses the DBSCAN (Density Based Spatial Clustering of Application with Noise) to calculate the number of neighboring feature points within a unit area epsilon (e.g., a circle of a predetermined radius) And when the predetermined number of neighboring feature points are included, the specific feature point may be labeled as a core object);
setting surrounding points of a second number ([211] For example, the processor 230 uses the DBSCAN (Density Based Spatial Clustering of Application with Noise) to calculate the number of neighboring feature points within a unit area epsilon (e.g., a circle of a predetermined radius));
determining the target point is the feature point indicating a feature of an object existing in the three-dimensional space based on a number of surrounding points whose distance between the three-dimensional position of the target point and the three-dimensional positions of the surrounding points is within a predetermined distance exceeding a threshold, and determining the target point is noise based on the number of surrounding points being at or below the threshold ([211] For example, the processor 230 uses the DBSCAN (Density Based Spatial Clustering of Application with Noise) to calculate the number of neighboring feature points within a unit area epsilon (e.g., a circle of a predetermined radius) And when the predetermined number of neighboring feature points are included, the specific feature point may be labeled as a core object, and in the opposite case, it may be labeled as a noise object).
Baek does not teach that a density of the surrounding points is higher than the feature points estimated in the estimating in an image region between the target point and the feature points estimated in the estimating; second estimating, using stereo matching, three-dimensional positions of the surrounding points whose three-dimensional positions have not been estimated by the first estimating.
Li, in the same field of endeavor of point cloud target detection, teaches that a density of the surrounding points (dots in Fig. 3) is higher than the feature points (x's in Fig. 3) estimated in the estimating in an image region between the target point and the feature points estimated in the estimating ([pg. 2177 para. 3] The proposed Multi-scale Neighborhood Feature Extraction and Aggregation Model (MNFEAM) contains two modules—the Multi-scale Feature Extraction Module (MFEM) and the Locality Feature Aggregation Module (LFAM). The main purpose of our model is to effectively learn high-level feature representations from the original world space features (such as XYZ, colors, and normals). The overview of MNFEAM is shown in Fig. 1. In Section III-A, we first introduce the MFEM part, in which we identify k neighboring points for each point at three different radius-defined scales, respectively. Upon the three neighborhood groups, we extract the multi-scale local features by passing through several MLPs in order to expand the receptive field and improve the abstraction of local information. [pg. 2179 para. 3] We identify n-nearest neighbors for each point in the multi-scale locality feature space (embedding) by L1-distance. For each point in the embedding a local feature grouping of n×m is formed).
PNG
media_image1.png
524
868
media_image1.png
Greyscale
Therefore, it would have been obvious to a person of ordinary skill in the art at the time that the invention was made to modify the method of Baek with the teachings of Li to use surrounding points with a higher density than the feature points because "a single point in a 3D scene does not carry much useful information, and it only makes sense when associated with its neighboring points. Therefore, the first and the most crucial step of point cloud learning is to define the neighbors of each point" [Li pg. 2177 para. 4].
Assarsson, in the same field of endeavor of stereo matching, teaches second estimating, using stereo matching, three-dimensional positions of the surrounding points whose three-dimensional positions have not been estimated by the first estimating ([0020] For each selected pixel in the first image a corresponding pixel in the other, second, image of the pair of images may be determined. A corresponding pixel may mean that the two pixels picture about the same point for the same object. In the ray marching, a stepwise marching along a ray from the first camera's position in 3D space through the pixel in the first image into the virtual 3D scene. For each step one is at a certain 3D position, p, along the ray. This position p may then be projected on the second image and the pixels from the first and the second images are compared whether the same point that is intended to be 3D reconstructed is pictured. Also pixels around the identified pixels may be used in the comparison. The comparison may result in a score for how good the matching between the two pixels is. [0067] stereo matching is done by raymarching view rays from one camera in each pair and storing the results as view-space geometry buffers. The matching is divided both into multiple steps and in a hierarchical fashion, where the resolution is doubled in each dimension for each level of the hierarchy (see FIG. 1). [0069] In the first ray-marching step, a resolution N times lower than the defined geometric resolution is started with. I.e., a view ray is followed for each 2.sup.Nth vertical and 2.sup.Nth horizontal pixel for one of the cameras in each pair. Marching is performed along the ray for a fixed length starting from the conservative estimate of the geometry constructed as described above. For each step along the ray, a rectangular filter is constructed in world space and projected on both cameras. Colors are sampled with mipmapping in the corresponding color textures, and the distance between both patches is computed using a mean-subtracted Sum of Absolute Difference (SAD) cost function).
Therefore, it would have been obvious to a person of ordinary skill in the art at the time that the invention was made to modify the apparatus of Baek with the teachings of Assarsson to perform stereo matching of the surrounding points for "real-time generation of 3D imaging that [is] fast and computational efficient while providing acceptable image quality" and "to be able to easily estimate and utilize normal information, stereo matching is done by raymarching view rays from one camera in each pair and storing the results as view-space geometry buffers" [Assarsson 0067].
Regarding claim 8, Baek, Li, and Assarsson teach the method of claim 7. Baek further teaches a non-transitory computer-readable medium storing one or more programs including instructions, which when executed by one or more processors of an information processing apparatus, cause the information processing apparatus to perform the information processing method ([149] The memory 220 may store various data for operation of the parking assisting device 200, such as a program for processing or controlling the processor 230. [150] Such memory 220 may include at least one of various hardware storage media such as ROM, RAM, EPROM, flash drive, hard drive, etc. to store the various data).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Baek in view of Li, Assarsson, and Yang (US20200142045A1).
Regarding claim 3, Baek, Li, and Assarsson teach the apparatus of claim 1. Yang, in the same field of endeavor of surrounding point estimation, teaches wherein, if an error of a number of the surrounding points having the values of distances from the imaging apparatus to positions indicated by the surrounding points, which are similar to the value of a distance from the imaging apparatus to a position indicated by the target point ([0065] d.sub.k represents the Euclidean distance between the to-be-measured point and the kth reference anchor node. The weight index β can be determined through experiments. Or, since its change causes change of average error, there is a functional relationship between β and the average error. Curve fitting is performed according to the weight index and the scatter plot of average positioning error, and the β value of the smallest value taken from the average positioning error is the optimal value. [0066] Calculation of α is the same as the calculation of β, and function relation equation of the α and the average positioning error is determined through experiments or curve fitting, the extremum of the function is solved to determine the optimal α value), or an error of a ratio of the number of the surrounding points having the values of the distances from the imaging apparatus to the positions indicated by the surrounding points, which are similar to the value of the distance from the imaging unit to the position indicated by the target point,
with respect to the number of the surrounding points exceeds a predetermined value, the setting the surrounding points increases the surrounding points that serve as the population for calculating the number or ratio of points with similar distances ([0078] FIG. 4 is a diagram that shows variations of KNN, GKNN (Gauss K-nearest neighbor) and EDW algorithm average error as the number of K nearest neighbor points increases. In the diagram, the GKNN algorithm is a KNN algorithm after Gaussian filtering. It can be seen that, initially, as K increases, the average error began to decrease).
Therefore, it would have been obvious to a person of ordinary skill in the art at the time that the invention was made to modify the apparatus of Baek with the teachings of Yang to use an error based on the surrounding points because "By selecting the appropriate K value, the above algorithms can all achieve the smallest error" [Yang 0078].
Claims 5-6 are rejected under 35 U.S.C. 103 as being unpatentable over Baek in view of Li, Assarsson, and Uchida (US20100226544A1).
Regarding claim 5, Baek, Li, and Assarsson teach the apparatus of claim 1. Uchida, in the same field of endeavor of feature point target setting, teaches wherein the setting the feature point preferentially sets the feature point having a shorter distance from the imaging apparatus, rather than at the feature point having a longer distance from the imaging apparatus, to a position indicated by the feature point as the target point ([0061] Based on the space distance which is the distance between the actual position of the feature point candidate and the distance measuring sensor 3 (i.e. the target vehicle) for example, the feature point selecting part 12 selects as a first feature point among plural feature point candidates the feature point candidate whose space distance is the smallest among the feature point candidates. [0045] The moving state estimating device 100 includes a control device 1 which is connected to an imaging sensor 2, a distance measuring sensor 3).
Therefore, it would have been obvious to a person of ordinary skill in the art at the time that the invention was made to modify the apparatus of Baek with the teachings of Uchida to select a target point based on a shorter distance because "By tracking a feature point in sequence according to an order of, for example, a feature point, a primary corresponding point, a secondary corresponding point, a tertiary corresponding point, etc., the moving state estimating device 100 can continuously estimate the moving state of the target vehicle beginning at the time when the moving state estimating device 100 obtains the image used for the selection of the feature point" [Uchida 0075].
Regarding claim 6, Baek, Li, and Assarsson teach the apparatus of claim 1. Uchida, in the same field of endeavor of feature point target setting, teaches wherein the setting the feature point preferentially sets the feature point having a shorter two-dimensional distance excluding depth information from a center of the first image as the target point than the feature point having a longer two-dimensional distance ([0061] Based on the space distance which is the distance between the actual position of the feature point candidate and the distance measuring sensor 3 (i.e. the target vehicle) for example, the feature point selecting part 12 selects as a first feature point among plural feature point candidates the feature point candidate whose space distance is the smallest among the feature point candidates. [0076] The moving state estimating part 14 is a part for estimating a moving state of a target vehicle. In an example, the moving state estimating part 14 generates homography by using the two-dimensional coordinate of the feature point selected by the feature point selecting part 12 and the two-dimensional coordinate of the corresponding point extracted by the corresponding point extracting part 13. [0045] The moving state estimating device 100 includes a control device 1 which is connected to an imaging sensor 2, a distance measuring sensor 3).
Therefore, it would have been obvious to a person of ordinary skill in the art at the time that the invention was made to modify the apparatus of Baek with the teachings of Uchida to select a target point based on a shorter distance because "By tracking a feature point in sequence according to an order of, for example, a feature point, a primary corresponding point, a secondary corresponding point, a tertiary corresponding point, etc., the moving state estimating device 100 can continuously estimate the moving state of the target vehicle beginning at the time when the moving state estimating device 100 obtains the image used for the selection of the feature point" [Uchida 0075].
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jacqueline R Zak whose telephone number is (571)272-4077. The examiner can normally be reached M-F 9-5.
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 at (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.
/JACQUELINE R ZAK/Examiner, Art Unit 2666
/EMILY C TERRELL/Supervisory Patent Examiner, Art Unit 2666
Prosecution Insights