Design and Implementation of Intelligent Analysis Technology in Sports Video Target and Trajectory Tracking Algorithm

Abstract

The object tracking is an important task to future generations to find accurate information from frame (sports or any cctv). The available models like RFO, DT, Xboosting, Gradient boosting, and CNN are outdated for sports video tracking using video. The target and trajectory finding are very important for accurate tracking. Therefore, for crossover, the above limitation’s intelligence analysis is performed with ANN. The video tracking can be very simple and efficient with proposed model. As a result, this ANN offers a sports camera target detection system based on an upgraded grey neural network. This research uses the automatic threshold technique to divide the target object of sports camera and derive the designated target of sports video based on the pixel grey difference of the target image in sports video. The moving objects are tracked by the object tracking algorithm, and features are extracted to obtain the results. The performance measures like accuracy of 95.23%, sensitivity of 94.45%, and recall of 93.34% had been attained.

Visual Image Recognition of Basketball Turning and Dribbling Based on Feature Extraction

“…In recent years, the number of sports videos has exploded thanks to the development of Internet technology and multimedia technology [1][2][3][4]. To acquire valuable information about players, actions, events from sports videos, it is necessary to analyze sports videos and summarize the tactics [5][6][7][8][9]. Compared with that of other sports, the video analysis of basketball is rather complicated, because basketball requires the cooperation between multiple players.…”