针对小波能量比在平滑的红外图像中提取过渡区存在误判的问题,提出用基于小波变换过渡区的算法。该算法通过小波变换将红外图像的低频和高频进行分离,对高频分量进行分析,构造方差,对方差进行阈值处理,重构过渡区,从而克服了小波能量比将红外背景误判为过渡区的不足。实验结果表明:本文算法提取出来的红外图像的过渡区具有很好的准确性。
In this paper, an algorithm is put forward based on wavelet transform, specific to the misjudge from extracting transition region in smooth infrared image by Wavelet energy ratio parameters. This algorithm separates the low and high frequency of infrared image first, analyzes the high frequency components, constructs variance to be judged by the threshold, then reconstructs transition region. It can overcome the trouble of transition region misjudged after these processes, compared to the algorithm with wavelet energy ratio parameter. The experimental results show that the improved algorithm can accurately extract the transition region of infrared image.