一种新的红外图像增强算法

针对传统红外图像增强容易丢失细节这一缺陷,采用基于分频的同态滤波与奇异值分解相结合的方法,对红外图像进行增强。在同态滤波的基础上,用高通滤波将其在频率域分成高频和低频,分别对高频和低频进行处理,高频通过线性增强,低频先经过奇异值分解,而后加入噪声处理,改变其奇异值,得到新的奇异值矩阵,最后对低频进行灰度调整再加上高频处理红的图像相加得到最后的增强后的图像。通过仿真,对比传统图像增强,该方法处理在峰值信噪比和均方值两方面均有改善。

A New Algorithm for Infrared Image Enhancement

For the existing problems of losing image details in the process of traditional image enhancement algorithms, this paper used a new method on the basis of homomorphic filtering combined with the frequency division and singular value decomposition to enhance the infrared image . Based on the homomorphic filter, With a high-pass filter in the frequency domain is divided into high frequency and low frequency, respectively to deal with high frequency and low frequency, through linear increase in high frequency, low frequency through the singular value decomposition, and then add noise processing, change its singular value, get new singular value matrix, and finally to gray-scale adjustment combined with low-frequency and high-frequency processing red images together to get the last of the enhanced images. Through the simulation and compared with the traditional image enhancement, this method in both peak signal-to-noise ratio and the mean square value are improved.