# US 10,872,267 B2 · Method for identification of characteristic points of a calibration pattern within a set of candidate points in an image of the calibration pattern > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,872,267 B2 - Document ID: US10872267B2 - Status: Granted - Kind: B2 - Application number: 15/779,904 - Filed: 2016-11-17 - Published: 2018-12-13 - Granted: 2020-12-22 - Earliest claimed: 2015-11-30 - Assignees: Aptiv Technologies AG - Applicants: Aptiv Technologies Limited - Inventors: Rafal Dlugosz; Waldemar Dworakowski; Zofia Dlugosz; Krzysztof Gongolewski - Web page: https://patentgrep.com/patent/US10872267B2 ## CPC - G06T 7/80: Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration - G06V 10/757: {Matching configurations of points or features} - G06T 2207/30208: Marker matrix - G06T 2207/30244: Camera pose - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method for identification of characteristic points of a calibration pattern within an image includes: (i) overlaying a template arrangement of template points over the candidate point such that a principal point of the template points coincides with the candidate point, wherein the template arrangement corresponds to the calibration pattern or to a sub-pattern of the calibration pattern, (ii) for each template point of the template arrangement, except the principal point, identifying from the set of candidate points the candidate point closest to the template point, and (iii) determining a degree of deviation by summing the distances between each template point of the template arrangement, except the principal point, and the candidate point closest to this template point; and identifying as characteristic points of the calibration pattern all those candidate points of the set of candidate points with a degree of deviation below a deviation threshold. ## Claims ### Claim 1 A method for identification of characteristic points of a calibration pattern within an image of the calibration pattern, the method comprising the steps of for each candidate point of a set of candidate points derived from the image of the calibration pattern: overlaying a template arrangement of template points over the candidate point such that a principal point of the template points coincides with the candidate point, wherein the template arrangement corresponds to the calibration pattern or to a sub-pattern of the calibration pattern, for each template point of the overlaid template arrangement, except the principal point, identifying from the set of candidate points the candidate point closest to the template point, and determining a degree of deviation by summing the distances between each template point of the template arrangement, except the principal point, and the candidate point closest to this template point; and identifying as characteristic points of the calibration pattern all those candidate points of the set of candidate points with a degree of deviation below a deviation threshold. ### Claim 2 (depends on 1) The method in accordance with claim 1, wherein for each candidate point of the set of candidate points, a region of interest is defined around the candidate point having a predefined area and shape so as to encompass the template arrangement when the template arrangement is overlaid to the candidate point, and wherein for each template point of the template arrangement, except the principal point, the candidate point closest to the template point is identified from those candidate points lying in the region of interest. ### Claim 3 (depends on 1) The method in accordance with claim 1, wherein the calibration pattern is an arrangement of sub-patterns comprising at least a first sub-pattern and a second sub-pattern, wherein for each candidate point of the set of candidate points, a first degree of deviation is determined by summing the distances between each template point of a first template arrangement, except a principal point coinciding with the candidate point, and the candidate point closest to this template point, and a second degree of deviation is determined by summing the distances between each template point of a second template arrangement, except a principal point coinciding with the candidate point, and the candidate point closest to this template point, wherein the first and second template arrangements correspond to the first and second sub-patterns, respectively, and wherein the candidate point is identified as a characteristic point of a first type, if its first degree of deviation is less than its second degree of deviation and is below a first deviation threshold, and is identified as a characteristic point of a second type, if its second degree of deviation is less than its first degree of deviation and is below a second deviation threshold. ### Claim 4 (depends on 3) The method in accordance with claim 3, wherein the candidate point is only identified as a characteristic point of a respective type, if not both of its first degree of deviation and its second degree of deviation are equal to zero. ### Claim 5 (depends on 1) The method in accordance with claim 1, wherein the distance between a template point of the template arrangement and a candidate point identified as candidate point closest to this template point is the square root of the sum of the squares of the coordinate differences between the template point and the candidate point. ### Claim 6 (depends on 1) The method in accordance with claim 1, wherein the distance between a template point of the template arrangement and a candidate point identified as candidate point closest to this template point is the sum of the squares of the coordinate differences between the template point and the candidate point. ### Claim 7 (depends on 1) The method in accordance with claim 1, wherein the distance between a template point of the template arrangement and a candidate point identified as candidate point closest to this template point is the sum of the absolute coordinate differences between the template point and the candidate point. ### Claim 8 (depends on 1) The method in accordance with claim 1, wherein after identifying the characteristic points the characteristic points are output for further evaluation. ### Claim 9 (depends on 8) The method in accordance with claim 8, wherein each characteristic point is output together with the candidate points identified as closest to a respective template point of the template arrangement overlaid over the characteristic point. ### Claim 10 (depends on 1) A non-transitory computer-readable storage-medium comprising one or more programs for execution by one or more processors of a first device, the one or more programs including instructions which, when executed by the one or more processors, cause the first device to perform the method of claim 1. ### Claim 11 A method for identification of characteristic points of a calibration pattern within an image of the calibration pattern, comprising: for each candidate point of a set of candidate points derived from the image of the calibration pattern: overlaying a template arrangement of template points over the candidate point such that a principal point of the template points coincides with the candidate point, wherein the template arrangement corresponds to the calibration pattern or to a sub-pattern of the calibration pattern; and determining a degree of deviation by summing the distances between each template point of the template arrangement, except the principal point, and the candidate point closest to this template point. ### Claim 12 (depends on 11) The method in accordance with claim 11, wherein for each template point of the overlaid template arrangement, except the principal point, identifying from the set of candidate points the candidate point closest to the template point. ### Claim 13 (depends on 11) The method in accordance with claim 11, wherein identifying as characteristic points of the calibration pattern all those candidate points of the set of candidate points with a degree of deviation below a deviation threshold. ### Claim 14 (depends on 11) The method in accordance with claim 11, wherein for each candidate point of the set of candidate points, a region of interest is defined around the candidate point having a predefined area and shape so as to encompass the template arrangement when the template arrangement is overlaid to the candidate point. ### Claim 15 (depends on 14) The method in accordance with claim 14, wherein for each template point of the template arrangement, except the principal point, the candidate point closest to the template point is identified from those candidate points lying in the region of interest. ### Claim 16 (depends on 11) The method in accordance with claim 11, wherein the calibration pattern is an arrangement of sub-patterns comprising at least a first sub-pattern and a second sub-pattern, wherein for each candidate point of the set of candidate points, a first degree of deviation is determined by summing the distances between each template point of a first template arrangement, except a principal point coinciding with the candidate point, and the candidate point closest to this template point. ### Claim 17 (depends on 16) The method in accordance with claim 16, wherein a second degree of deviation is determined by summing the distances between each template point of a second template arrangement, except a principal point coinciding with the candidate point, and the candidate point closest to this template point. ### Claim 18 (depends on 17) The method in accordance with claim 17, wherein the first and second template arrangements correspond to the first and second sub-patterns, respectively, and wherein the candidate point is identified as a characteristic point of a first type, if its first degree of deviation is less than its second degree of deviation and is below a first deviation threshold, and is identified as a characteristic point of a second type, if its second degree of deviation is less than its first degree of deviation and is below a second deviation threshold.