一种用于图像匹配的演化算法研究

Research of evolutionary algorithm used for image match

图像匹配在目标检测、目标识别与跟踪等应用中是一项极其重要的技术。图像匹配的方法很多,现有的方法均采用遍历式搜索策略,因而计算量的降低有限,如果不采用新的搜索策略,则难以在计算量的降低上取得实质性的突破,这是现有图像相关匹配算法的共同缺陷。将演化算法引入匹配技术中,利用相关匹配计算简单、抗噪声性能好的特点,提出了一种基于演化算法的图像相关匹配方法。实验结果表明该方法具有运算速度快、正确匹配率高、抗噪声性能好等特点。

Image match is an extremely important technology in the applications of target detecting,target identification and tracking.There are many image matching methods,now the available method adopts the ergodic dyadic search strategy without exception,the lessening calculating quantity＇s as a result is limited,if not adopt the new search strategies,the break-through being difficult to get a substantiality on the lessening calculating quantity＇s,this is the common faults of the image match algorithms.This paper uses the evolutionary algorithm into the match technology,uses the characteristics of the relevance match such as calculation simplicity,the anti-noise function well,and proposes an image match method base on the evolutionary algorithm.The experiment results indicate that the method has the characteristics such as the high speed calculation,correct matching rate height and anti-noise function regards.