改良型MACH滤波器算法的形变目标识别(英文)

Distorted target recognition based on improved MACH algorithm

联合变换相关器可用于精确识别与定位目标,其瓶颈在于如何在复杂背景中识别形变目标,这也限制了目标模式识别技术的发展。为解决这一问题,提出了一种改良型的最大平均高度(MACH)滤波器算法。该算法对于旋转形变及尺寸形变目标均有很强的识别能力。根据大量的实验结果分析,对合成滤波器的控制参数进行了优化,使得该滤波器具有较高的形变识别容差,并能有效抑制复杂噪声。通过将频域中改良的MACH滤波器返回至空域,可获得含有目标多种不同状态的MACH参考模板。利用该模板,能有效增强相关峰亮度,扩大复杂背景中形变目标的识别范围。作为实例,对复杂背景中的战舰目标进行了计算机模拟实验与光学实验。实验结果证明了该算法的可行性和实用性。

Joint transform correlator （JTC） can make targets recognized and located accurately, but the bottleneck technique of JTC is how to recognize the distorted targets in cluttered scene. This has restricted the development of the pattern recognition for target image. In order to solve the problem, improved Maximum Average Correlation Height （MACH） filter algorithm was presented. The MACH algorithm has powerful capability of recognition for distorted targets（rotation and scale etc.）. According to the analysis result of amounts of experiments, the control parameters of the synthesized filter were optimized, which makes the filter have higher distortion tolerance and can suppress cluttered noise effectively. When improved MACH filter algorithm in frequency domain was projected to space domain, the MACH reference template image can be obtained which includes various forms of distorted target image. MACH reference template can sharpen the correlation peaks and expand recognizing scope for distorted targets in cluttered scene. As practical examples, computer simulation experiments and optical experiments for warship in cluttered scene were carried out. The experimental results prove the feasibility and actual effect of the algorithm.