Research on Anti-Camouflaged Target System Based on Spectral Detection and Image Recognition
WANG Bo1, GAO Yu-bin2, LU Xu-tao3
1. School of Computer Science and Control Engineering, North University of China, Taiyuan 030051, China 2. School of Science, North University of China, Taiyuan 030051, China 3. School of Information and Communication Engineering, North University of China, Taiyuan 030051, China
Abstract:To be able to quickly and efficiently identify Enemy camouflaged maneuvering targets in the wild environment, target recognition system was designed based on spectral detection technology and video target recognition method. System was composed of the visible light image acquisition module and static interferometer module. The system used image recognition technology to obtain two dimensional video images of measurement region, and through spectrum detection technology to identify targets. Ultimately, measured target was rebuilt on the corresponding position in the image, so the visual target recognition was realized. After the theoretical derivation, identifiable target function formula of the system was obtained, and based on the functional relationship to complete the quantitative experiments for target recognition. In the experiments, maneuvering target in the battlefield environment was simulated by a car. At different distances, the background was respectively selected to detect a flat wasteland, bushes and abandoned buildings. Obvious target, coated camouflage target and covered disguises target was respectively spectrum detection. Experimental results show that spectrum detection technology can overcome the shortcomings of unrecognized the camouflaged target by traditional image target recognition method. Testing background had some influence on spectrum detection results, and the continuity of the background was conducive to target recognition. Covered disguises target was the hardest to identify in various camouflage mode. As the distance between the target and the system increases, signal to noise ratio of the system was reduced. In summary, the system can achieve effective recognition of camouflaged targets to meet the design requirements.
王 博1,高玉斌2,鲁旭涛3 . 基于光谱探测与图像识别的反伪装目标系统研究 [J]. 光谱学与光谱分析, 2015, 35(05): 1440-1444.
WANG Bo1, GAO Yu-bin2, LU Xu-tao3 . Research on Anti-Camouflaged Target System Based on Spectral Detection and Image Recognition . SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(05): 1440-1444.