Goal recognition method using intelligent analysis of basketball images under 6G Internet of Things technology. (14th July 2022)

• Record Type: Journal Article
• Authors: Qin, Guangyu; Yu, Yang
• Is Part Of: International journal of grid and utility computing. Volume 13:Number 2/3(2022)
• Page Start: 138
• Page End: 145
• Publication Date: 2022-07-14
• Subjects: object detection -- data mining -- Internet of Things -- Hough circle transform; Electronic data processing -- Distributed processing -- Periodicals; Electronic commerce -- Management -- Computer programs -- Periodicals; 004.605
• Languages: English
• ISSNs: 1741-847X

Abstract

Goal recognition is an important technology for basketball special automatic tests. And accurate and timely judgement of basketball goals is very helpful during the basketball sport. Therefore, an intelligent basketball goal recognition and analysis system based on 6G Internet of Things is proposed. First, data mining technology is used to explore the factors affecting the goal rate, and the basketball images are pre-processed. Second, the background-difference and the three-frame difference method are fused to detect the basketball target during the image pre-processing. Finally, the basketball features are extracted by image calibration technology, and they are compared with the parameters in the system to judge whether the goal can be scored. The experimental results show that the frame rate of the proposed algorithm is 58 frames per second, and meets the real-time processing requirement of 25 frames. Therefore, false detection rate of the algorithm is a low no matter it's a layup from the left or the right, and the algorithm can be used in basketball games or examinations.