光学联合变换相关技术的分析-光学工程专业论文.docx

捅要传统的匹配滤波相关需要事先制作匹配滤波器，这种方法计算量十 捅要 传统的匹配滤波相关需要事先制作匹配滤波器，这种方法计算量十 分庞大、速度慢。最主要的是很难实现实时处理，判断识别精度也不高。 进而人们提出了光学联合变换相关的方法，通过光学相关的方法进行自 动识别克服匹配相关的缺点，即能实现实时，又能实现高精度。但经典 的联合变换相关存在大的相关旁瓣、很大的相关带宽和低的光效率等。 为了得到更好的相关结果，本文提出对联合功率谱进行非线性处理的方 法。为了解决图像平移、旋转、比例变化对相关的影响，可通过傅里叶 变换解决平移的问题；通过Polar变换解决旋转的问题：通过Ln变换 解决图像比例变换的影响。 本论丈主要对提出的联合变换相关的整体结构进行了描述：详细介 绍了联合傅里叶变换的相关理论；对经典联合变换相关进行了仿真实 验；给出了联合变换功率谱二值化的仿真结果；介绍了Polar和Ln更 换详细的算法；最后给出了基本的光学实验结果。 关键词：目标识葑≮合主≤；关 Pol。，巅：Ln矗．联函率谱 AbstractConventional Abstract Conventional matched filter-based correlator needs to make matched filter in advance，it has many disadvantages：large calculation，slow speed， not suitable for real—time processing，and recognition precision is low So optical joint transform correlation is proposed．Optical joint transform correlation can recognition automatically,conquer drawback of filter-based correlator，and it is suitable for real—time processing，can achieve well recognition．However,the correlation signal obtained by the classical joint transform correlator(JTC)is associated with large correlation sidelobes， large correlation bandwidth，and low optical efficiency．In this paper,in order to achieve better correlation result，the nonlinear processing of joint transform power spectrum(JTPS)is put forward．In order to resolve parallel shift，rotation and scale shift of images，parallel shift Can be resolved by Fourier transform；rotation Can be resolved by Polar transform；scale shift Can be resolved by Ln transform． In the paper,we describe whole flame of joint transform correlation； introduce interrelated theories of joint Fourier transform in detail；make imitation experiment of computer of classical transform correlation；present experiment results of binary JTPS；introduce arithmetic of Polar transform and Ln transform in detail；and achieve basic optical experiment results． Keywords：object recognition，joint transform correlation，Polar transform，Ln transform，joint transform power spectrum． 第一章绪论§1．1光学信息处理介绍