基于模拟退火算法的红外图像自适应对比度增强

基于模拟退火算法击和非完全Beta函数，提出了一种自适应红外图像对比度增强方法。首先基于输入红外图像的灰度直方圈给出一种有效的判据。对输入红外图像先利用所提出的判据判断图像的对比度类型．然后利用这个判据来确定灰度变换参数的搜索空间，进一步指导模拟退火算法的搜索方向和初值的选取，利用模拟退火算法在上述确定的灰度变换参数空间中搜索最佳的灰度变换参数，从而获得一条最佳的灰度变换曲线，实现对图像进行全局增强处理。实验结果表明，该算法在有效地提高红外图像整体对比度的同时，很好地保留了图像中的细节部分信息。算法在视觉质量上优于传统的直方图均衡法、反锐化掩膜法。

Adaptive Contrast Enhancement of Infrared Image Based on Simulated Annealing Algorithm

A new kind of infrared image adaptive contrast enhancement approach based on simulated annealing algorithm and incomplete Beta function is given. Firstly, an efficient criterion is given based on gray level histogram of input image. Input image contrast type is determined by the criterion mentioned above, and then the criterion is used to determine the searching space of gray transformation parameters and guide the search orientation and the selection of initial values of simulated annealing algorithm. The optimal gray transformation parameters are obtained in the parameter space with the simulated annealing algorithm. Thus an optimal gray transformation curve is obtained and global contrast enhancement is done to the input image based on the optimal gray transformation curve. The performance of the new approach is analyzed. Lots of experiments have been done in order to validate the efficiency of the new approach. Based on the experimental result, the new approach can enhance the global contrast of the infrared image effectively while it also keeps the details of the target in the infrared image well. In visual quality, the new approach is better than traditional histogram equalization method and un sharp mask method.