Research on Image Classification of Sports Training Video Based on Grey Relational Analysis and Support Vector Machine

Aiming at the problem of poor performance of change detection in traditional sports training video image classification methods, a sports training video image classification method based on gray correlation analysis and support vector machine is designed. The decision denoising algorithm is used to denoise the sports training video image, and the sports training video image is divided into a 3×3 filtering neighborhood window to denoise. Eight neighborhood search method and parallel image algorithm are combined to extract the features of the sports training video image, obtain the specific contour of the feature area of the sports training video image, implement contour tracking for the specific contour of the feature area of the sports training video image, and merge multiple record sequence tables that may be generated when tracking the contour. Based on gray correlation analysis, the correlation degree of sports training video image features is analyzed, and the sports training video image features are classified by support vector machine, so as to realize the classification of sports training video image. In order to prove that this method has better change detection performance, the traditional sports training video image classification method is compared with this method. The experimental results show that this method has better change detection performance than the traditional method.