On the Research of Automatic Counting Score Method and its Application in Shooting Training Based on CIELAB Color Model and Radial Basis Function

摘要

Aiming at the key technology of automatic target-reporting system, which are paper image analysis and bullet hole recognition, a fast hole-location and score-counting method is proposed in this paper. It is consisted of two key algorithms: the two-step fast & fine segmentation algorithm based on CIELAB color model and the self-adapt scoring algorithm based on radial basis function. Firstly, the fast Gauss filter is used to smooth the target image and avoid the influence of white loop on threshold settling. Secondly, based on the CIELAB color model, the A-channel layer of the green target paper is extracted for threshold segmentation. When the complete and effective target area is separated out, the 10-ring region and bullet hole location are identified through the calculation of image region roundness. Finally, by establishing the two-dimensional Euclidean space radial basis function, a self-adaptive fast scoring algorithm is realized which does not depend on the size of target paper and improves the generality of automatic target reporting system. In the last part of the paper, a series of experiments were carried out. All the results show that the proposed method is effective and reliable, with good robustness and adaptability. It is suggested that the application of our method will benefit on the improvement of target-reporting system in shooting training.