# US 2018/0322641 A1 · Moving body tracking method, moving body tracking device, and program > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0322641 A1 - Document ID: US20180322641A1 - Status: Published application - Kind: A1 - Application number: 15/774,438 - Filed: 2016-10-14 - Published: 2018-11-08 - Earliest claimed: 2015-11-13 - Assignees: Panasonic Intellectual Property Management Co., Ltd. - Applicants: Panasonic Intellectual Property Management Co., Ltd. - Inventors: Junko UEDA - Web page: https://patentgrep.com/patent/US20180322641A1 ## CPC - G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments - A63B 24/0021: {Tracking a path or terminating locations} - G06T 7/194: involving foreground-background segmentation - G06T 7/248: {involving reference images or patches} - H04N 7/18: Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast - A63B 2024/0031: {at the starting point} - G06T 2207/10016: Video; Image sequence - G06T 2207/30224: Ball; Puck - G06T 2207/30241: Trajectory - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - A63B: APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT (apparatus for passive exercising, massage A61H) - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract In a case in which the size of a moving body changes by a position of the moving body, in order to accurately track the moving body without an effort of the user regardless of a change in the size, the present disclosure estimates a size of the moving body in each of the images of the frames, updates template data based on the estimated size of the moving body, and compares a candidate region that is a candidate of a region where the moving body is present in each of the images of the frames with the template data, and determines whether or not the moving body is positioned in the candidate region based on the comparison result. ## Claims ### Claim 1 A moving body tracking method of tracking a position of a moving body in each of images of a plurality of frames configuring a sports image obtained by capturing a game of a sports, by a processor, the method comprising: estimating a size of the moving body in each of the images of the frames; updating template feature amount to a feature amount generated from a template image corresponding to the estimated size of the moving body, based on the estimated size of the moving body; and comparing a feature amount of a candidate region that is a candidate of a region where the moving body is present in each of the images of the frames with the template feature amount corresponding to the estimated size of the moving body, and determining whether or not the moving body is positioned in the candidate region based on the comparison result. ### Claim 2 (depends on 1) The moving body tracking method of claim 1, wherein the comparison between the feature amount of the candidate region and the template feature amount is performed by calculating a distance between a color histogram of the candidate region and a color histogram of the template feature amount. ### Claim 3 (depends on 1) The moving body tracking method of claim 1, wherein the processor further updates a parameter including, at least, a predetermined threshold value for detecting the moving body, or a predetermined threshold value for tracking the moving body, based on the estimated size of the moving body. ### Claim 4 (depends on 1) The moving body tracking method of claim 1, further comprising: changing the template feature amount used for the comparison according to a stage to which the estimated size of the moving body belongs, among a plurality of stages. ### Claim 5 (depends on 4) The moving body tracking method of claim 4, further comprising: generating the template feature amount using an image in a vicinity of the moving body in a case in which the estimated size of the moving body changes to another stage of the plurality of stages. ### Claim 6 (depends on 1) The moving body tracking method of claim 1, further comprising: scanning a rectangle region in the image of the frame by sequentially comparing the rectangle region with the template data in a case in which a candidate region satisfying the condition is not present, and determining whether or not the moving body is present in the rectangle region based on the comparison result, and generating the template feature amount using an image in a vicinity of the rectangle region where it is determined that moving body is present. ### Claim 7 (depends on 1) The moving body tracking method of claim 1, wherein an initial value of the template feature amount is generated in advance before the game of the sports starts. ### Claim 8 (depends on 1) The moving body tracking method of claim 1, further comprising: generating a trajectory detection image based on the images of the plurality of frames, and detecting a trajectory of the moving body based on the trajectory detection image. ### Claim 9 (depends on 1) The moving body tracking method of claim 1, further comprising: displaying the template image on a display unit and performing a display for receiving an operation of deleting a predetermined template feature amount. ### Claim 10 A moving body tracking device for tracking a position of a moving body in each of images of a plurality of frames configuring a sports image obtained by capturing a game of a sports, wherein a size of the moving body in each of the images of the frames is estimated; template feature amount to a feature amount generated from a template image corresponding to the estimated size of the moving body, based on the estimated size of the moving body is updated; and a feature amount of a candidate region that is a candidate of a region where the moving body is present in each of the images of the frames is compared with the template feature amount corresponding to the estimated size of the moving body, and it is determined whether or not the moving body is positioned in the candidate region based on the comparison result. ### Claim 11 (depends on 10) The moving body tracking device of claim 10, wherein the comparison between the feature amount of the candidate region and the template feature amount is performed by calculating a distance between a color histogram of the candidate region and a color histogram of the template feature amount. ### Claim 12 (depends on 10) The moving body tracking device of claim 10, further updates a parameter including at least a predetermined threshold value for detecting the moving body, or a predetermined threshold value for tracking the moving body, based on the estimated size of the moving body. ### Claim 13 (depends on 10) The moving body tracking device of claim 10, further changes the template data used for the comparison according to a stage to which the estimated size of the moving body belongs, among a plurality of stages. ### Claim 14 (depends on 13) The moving body tracking device of claim 13, further generates the template feature amount using an image in a vicinity of the moving body in a case in which the estimated size of the moving body changes to another stage of the plurality of stages. ### Claim 15 (depends on 10) The moving body tracking device of claim 10, further scans a rectangle region in the image of the frame by sequentially comparing the rectangle region with the template data in a case in which a candidate region satisfying the condition is not present, determines whether or not the moving body is present in the rectangle region based on the comparison result, and generates the template feature amount using an image in a vicinity of the rectangle region where it is determined that moving body is present. ### Claim 16 (depends on 10) The moving body tracking device of claim 10, wherein an initial value of the template feature amount is generated in advance before the game of the sports starts. ### Claim 17 (depends on 10) The moving body tracking device of claim 10, further generates a trajectory detection image based on the images of the plurality of frames, and detects a trajectory of the moving body based on the trajectory detection image. ### Claim 18 (depends on 10) The moving body tracking device of claim 10, further displays the template image on a display unit and performs a display for receiving an operation of deleting a predetermined template feature amount. ### Claim 19 A storage medium that stores program which causes a processor to sequentially execute a process of tracking a position of a moving body in each of images of a plurality of frames configuring a sports image obtained by capturing a game of a sports, the process comprising: estimating a size of the moving body in each of the images of the frames; updating template feature amount to a feature amount generated from a template image corresponding to the estimated size of the moving body, based on the estimated size of the moving body; and comparing a feature amount of a candidate region that is a candidate of a region where the moving body is present in each of the images of the frames with the template feature amount corresponding to the estimated size of the moving body, and determining whether or not the moving body is positioned in the candidate region based on the comparison result.