Goal detection method and system combining target detection and dynamic difference

A technology of target detection and detection method, applied in the field of computer vision, which can solve the problems of arranging score sheets or scorers, impractical, etc.

Pending Publication Date: 2021-01-05
中科人工智能创新技术研究院(青岛)有限公司
View PDF2 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it is impractical to have score sheets or scorekeepers for all basketball games, especially for amateur games

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Goal detection method and system combining target detection and dynamic difference
  • Goal detection method and system combining target detection and dynamic difference
  • Goal detection method and system combining target detection and dynamic difference

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027]The purpose of the present disclosure is to address the problems encountered in the automatic scoring of goal detection in the basketball scene. On the one hand, it introduces a target detection algorithm to quickly determine the basket area, and on the other hand, it uses computer graphics technology to perform visual data in the target scene. Process and analyze to obtain the movement of objects in the target scene. First, a fast target detection algorithm is needed to quickly locate the target area during use; then an algorithm that uses a dynamic difference method to obtain a moving target is designed to determine whether there is a goal action in the basket area; finally this disclosure encapsulates the algorithm into goal detection System for actual testing deployment.

[0028]To achieve the above objective, a goal detection method based on a target detection framework (taking YoLo-V3 as an example) is provided. Considering the requirements of real-time and practicality, a ...

Embodiment 2

[0066]The present disclosure also provides a goal detection system combining target detection and dynamic difference, including:

[0067]The data acquisition module is configured to collect scene data and video to be tested, and mark the goal area in the scene data;

[0068]The goal area coordinate acquisition module is configured to: establish a goal detection model, train the goal detection model through the marked scene data, and obtain the trained goal detection model; input the video to be tested into the goal detection model to obtain The coordinates of the goal area of ​​the video to be tested;

[0069]The goal detection module is configured to: preprocess the video to be tested to obtain an image to be tested, and process the image to be tested through a dynamic difference algorithm to obtain a moving target image; determine whether the moving target image has a moving target in the goal area coordinates, if If yes, output goal data, if not, output no goal data.

[0070]Further, the spe...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a goal detection method and system combining target detection and dynamic difference, and the method comprises the steps: collecting scene data and a to-be-detected video, and marking a goal region in the scene data; establishing a goal detection model, and training the goal detection model through the marked scene data to obtain a trained goal detection model; inputting a to-be-detected video into the goal detection model to obtain goal area coordinates of the to-be-detected video; preprocessing the to-be-detected video to obtain a to-be-detected image, and processing the to-be-detected image through a dynamic difference algorithm to obtain a moving target image; and judging whether a moving target exists in the moving target image in the coordinates of the goal area, if so, outputting goal data, and if not, outputting no goal data.

Description

Technical field[0001]The present disclosure relates to computer vision, and in particular to a goal detection method and system combining target detection and dynamic difference.Background technique[0002]Target detection methods have been widely used in application scenarios such as intelligent video surveillance and autonomous driving, and they also have potential applications in sports. When a basketball game is held, a score table or scorekeeper is required to record the score. However, it is impractical to arrange score sheets or scorers for all basketball games, especially for amateur games. Therefore, automatic score detection and counting of basketball courts is feasible. More importantly, now is the era of social networks. Players want to share their highlights on the basketball court. How to automatically detect basketball scores, which will help the app to cut, upload, store and share Wonderful video is a technical problem that needs to be solved urgently.Summary of the in...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/40G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V20/42G06V10/30G06V10/44G06V2201/07G06N3/045G06F18/23213G06F18/214
Inventor 张兆翔宋纯锋岳绍龙王海滨单彩峰
Owner 中科人工智能创新技术研究院(青岛)有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap