Sports Video Moving Target Detection and Tracking Based on Improved Gaussian Mixture Model
Download as PDF
DOI: 10.25236/ISMHI.2019.038
Corresponding Author
Xuming Xiao
Abstract
Aiming At the Shortcomings of Current Sports Video Moving Target Detection, a Hybrid Model for Sports Video Moving Target Detection and Tracking is Proposed. by Analyzing the Shortcomings of Gaussian Mixture Model, the Original “Background Re Composition, Model Update, Background Update, Target Detection” Color Image Gray-Scale Image is Transformed into the Dynamic Background Field with Small Pixel Similarity. in Order to Improve the Efficiency and Accuracy of the Hybrid Model for Moving Video Targets, Gray Histogram Based Target Tracking is Added. Experimental Results Show That This Method Has Strong Interference Processing Ability for Sports Video, Large Detection Range, Good Detection and Tracking Effect.
Keywords
Gaussian Mixture Model; Sports Video; Moving Target; Target Detection; Target Tracking