Research On Automatic Recognition And Location Technology Of Salvo Bomb Shell

The real-time automatic recognition and location technology of salvo bombshell is widely used in the artillery military training evaluation system. Traditional method of manual location has been unable to meet the military requirements of modern digital battlefield. Aiming at the practical application requirement of the realtime accurate recognition and location of salvo bomb shell, A solution of automatic recognition and accurate location of salvo bomb shell based on infrared imaging and binocular vision(the PZC system) is proposed in this paper, and the key technologies including the automatic detection of salvo bomb shell in monocular field of view and the automatic matching of homonymous points in binocular field of view have been studied deeply and systematically. Relevant results have been integrated in the PZC system. The feasibility of the solution is verified by multiple battlefield experiments. The main work of this paper is as follows.The related literature at home and abroad was studied. On the basis of the existing research and aiming at the characteristics of the burst point target and infrared image are researched, a system integrated solution about automatic detection and location of salvo bomb shell has been proposed in this paper. We set up left and right infrared observation stations to obtain infrared image sequences in each battlefield. The infrared images are getting pre-processed. Automatic detection and location of salvo bomb shell will be launched in infraredimage sequences after pre-processed.In this paper, an automatic exposure compensation control solution for infrared observation stations is proposed. An edge detection algorithm based on unsharp masking is proposed to enhance the detail of infrared image with a significant effect.Due to the inherent characteristics of bomb shell target, the traditional target detection algorithm is not ideal. Through the research of the VIBE background modeling algorithm, A target detection method based on the characteristics of VIBE and bomb shell is proposed in this paper. We also designed a post-processing method to eliminate the internal fragmentation of the target. According to the individual characteristics and group characteristics of bomb shell target, we proposed a method of correlation tracking of salvo bomb shells based on the target chain and the characteristics of space-time sequence features of the targets.This paper analyzes and studies the characteristics and difficulties of the binocular matching of salvo bomb shells. Based on the time-slice segmentation and bi-verificationof the constraint of shortest common perpendicular line, a binocular matching algorithm of salvo bomb shells has been proposed. This paper uses the common perpendicular midpoint method to calculate 3D coordinates of space physics of the matched bomb shells in previous step.In this paper, the key algorithm of the PZC system is realized based on Open CV2.4.9in VS2010 platform. Through the experimental analysis based onvideo images of battlefield live fire and out-field simulation experiment by using proposed algorithm, the effectiveness and practicability of the solution are verified by the comprehensive method.It is proved that the proposed algorithmhas a certain effect on automatic detection and matching of salvo bomb shells.