Lidar-based sporting object identification system
AssigneeSportsMEDIA Technology Corporation
InventorsGerard J. Hall, Matthew Bertram, Hans Weber
Life of this patent
Estimating the term…
Abstract
A system and method for identifying and locating sporting objects on a sporting terrain, for example, golf balls on a fairway or green using at least one LiDAR-based sensor. Visual cameras identify candidate locations for a golf ball based on identification of colored areas of images corresponding to the golf ball color. Subsequently, one or more LiDAR-based sensors confirm which candidate locations corresponding to a golf ball based on local changes in reflectivity of objects in the candidate areas.
Claims20
A light detection and ranging (LiDAR)-based system for identifying and locating a sporting object comprising:
- a processor, at least one LiDAR sensor, and at least one visual camera sensor;
- wherein the processor is in communication with the at least one LiDAR sensor and the at least one visual camera sensor, wherein the processor includes a computer vision analysis module;
- wherein the at least one visual camera sensor is configured to capture a snapshot of an area to generate visual camera data;
- wherein the computer vision analysis module is configured to utilize a machine learning (ML) algorithm to determine at least one region of interest (ROI) based on the visual camera data, wherein the at least one ROI includes at least one candidate area;
- wherein the at least one LiDAR sensor is configured to perform a LiDAR scan of the at least one ROI to determine a location of the sporting object and generate LiDAR data;
- and
- wherein the processor is operable perform a reflectivity analysis to determine whether the at least one candidate area includes the sporting object based on the LiDAR data and a reflectivity threshold.
A light detection and ranging (LiDAR)-based system for identifying and locating a sporting object comprising:
- a processor, at least one LiDAR sensor, and at least one visual camera sensor;
- and
- at least one battery configured to provide power to the at least one visual camera sensor and the at least one LiDAR sensor;
- wherein the processor is in communication with the at least one LiDAR sensor and the at least one visual camera sensor, wherein the processor includes a computer vision analysis module and a circularity filter, and wherein the processor is operable to perform a reflectivity analysis;
- wherein the at least one LiDAR sensor is configured to perform a LiDAR scan of an area to determine at least one region of interest (ROI) based on the reflectivity analysis and the circularity filter;
- wherein the reflectivity analysis and the circularity filter are operable to determine if the at least one ROI comprises at least one candidate area based on the LiDAR scan, a reflectivity threshold, a circularity threshold, and a circularity algorithm;
- wherein the at least one visual camera sensor is configured to perform a snapshot of the ROI to generate visual camera data;
- and
- wherein the computer vision analysis module is configured to utilize a machine learning (ML) algorithm to confirm or deny if the at least one candidate area includes the sporting object based on the visual camera data.
A light detection and ranging (LiDAR)-based method for identifying and locating a sporting object comprising:
- scanning an area, using at least one LiDAR sensor, to determine at least one region of interest (ROI) and generate LiDAR data;
- determining, using a reflectivity analysis and a circularity filter, at least one candidate area in the at least one ROI based on the LiDAR data, a reflectivity threshold, a circularity threshold, and a circularity algorithm;
- performing a snapshot of the at least one ROI, using at least one visual camera sensor, to generate visual camera data;
- and
- confirming or denying, using a computer vision analysis module, if the at least one candidate area includes the sporting object based on the visual camera data.
Dependencies are inferred from the claim text.