Study Of Target-Scoring Algorithm Based On Image Recognition In Automatic Scoring System For Shooting Sports

The Automatic Scoring System for Shooting Sports is a system developed forshooting matches according to Competition Rules of Shooting Sports. It realizes theautomation of target replacing, scoring, and results management, but cannot be appliedto shoot-off because of its limited scoring precision. In this paper, we analyze theexisted result-determining method. Based on the related theory of image recognitionand processing, an improved target-scoring algorithm is proposed. In the new algorithm,the information of the original image of the target sheet is taken full advantage of. Aproper oriented-filtering mask is designed to enhance the loop lines in the image alongtheir orientation and further recognize them. Then, according to the geometricrelationship between bullethole and loop lines, the corresponding loop-value (shootingscore) is calculated. This improvement greatly decreases the influence of such factors asnoise disturbance and geometric distortion on automatic target-scoring, further increasesthe scoring accuracy of the Automatic Scoring System for Shooting Sports, and satisfiesthe requirements of various types of shooting matches including shoot-off. Thisalgorithm guarantees the reliable performance of the system in its practical application.