改进的图像相关匹配算法

A Fast Correlation Matching Algorithm

摘要:

将遗传算法用于低对比度小目标图像相关匹配跟踪,提出了改进的快速图像相关匹配算法。结合图像匹配本身的特点,采用新的编码方式克服了二进制编码的缺点。采用NPROD匹配准则在同样的匹配效果条件下,提高了匹配精度,该算法在计算速度上较原始算法提高了至少12倍以上,同时该算法还具有每帧图像匹配计算时间基本恒定的优点,便于在实际系统中采用。实验结果表明该方法可以有效地跟踪低对比度小目标。

Abstract:

A fast image correlation matching algorithm is put forward for low contrast TV image sequence of a ship's target using genetic algorithms. A new coding method is used to overcome the defects of binary coding method. According the NPROD matching criterion the precision is improved and the computation speed is above 12 times than the original algorithm. The computation time is nearly constant for each frame matching operation. Experimental results show that the method can track the low contrast and small target effectively.