本文基于最大平均相关高度算法设计了一种归一化模板，解决了联合变换相关器的畸变目标识别问题。该方法通过对最大平均相关高度频域滤波器的控制参数优化，映射到空域获得初始模板，归一化后得到归一化模板。相比于初始模板，归一化模板大大提高了模板与待测目标的灰度相似性。仿真实验表明，该模板能有效增强待探测目标的相关峰亮度，在联合变换相关器上可实现对畸变目标的识别，其比例畸变容差为53%，旋转畸变容差为20°。通过实验，给出了畸变飞机的相关识别结果。%Normalized template based on Maximum Average Correlation Height (MACH) algorithm is designed,in this paper,to solve the problem of inaccurate recognition for distorted target(rotation or scale ) with Joint Transform Corre-lator (JTC).First of all,the performance control parameters of MACH filter were optimized in frequency domain;second-ly,the optimized MACH frequency domain filter was projected to object space to obtain a initial template;and then,nor-malized template was achieved via normalization for the initial template. The gray similarity of normalized template with the detected target is increased. The optical correlation simulations show that the brightness of the correlation peaks can be enhanced,and the recognition for distorted target on JTC is achieved. The scale distortion tolerance is 53%,and the rotation distortion tolerance is 20°. As practical examples,the optical simulations of a plane were presented.
展开▼
