Application Research On Image Processing Technology In Shooting Training System

The efficiency of shoot training is urgent to been resolved in troops and shooting sports. The computer-aided shoot training system plays a very important role in saving ammunition, analyzing the process of shooting, improving training quality and efficiency, coaching training and so on. For this reason, many domestic and foreign organizations have done much in research and development of computer-aided shoot training system. Some practical computer-aided shooting training systems have been reported. Because of adopting infrared technology, the most of such systems are expensive and bring some negative impacts to trainers.This thesis aims to utilize digital image processing technology to resolve the problems such as image pre-processing, border recognizing, interfering object excluding, target center computing and so on in computer-aided shooting training systems, and to design and develop such a system based on visual C + + platform. The solution based on digital image processing technology not only make the final aided shooting training system in the support price much lower than the systems based on infrared technology, but it also can avoid the negative impacts to athletes brought by the laser and the weight of auxiliary equipment on guns.For accurately and real-time identifying the target, several typical recognition algorithms are firstly analyzed, and an improved template matching algorithm is then presented. This method can not only rule out interfering objects, but also give the accurate real-time identification to the target at different locations.For accurately and real-time calculating the target's center, the limitations of some existing algorithms are analyzed, and a new algorithm based on symmetry point of curve part is presented for target center calculation. Experiments show that this algorithm is accurate and efficient and meets the requirements of the aided shooting training system.