结合直方图均衡和模糊集理论的红外图像增强  

Enhancement of Infrared Image Combined with Histogram Equalization and Fuzzy Set Theory

作　　者：云海姣 吴志勇[1] 王冠军[1,2] 刘雪超[3] 梁敏华[1] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所光电测控部,长春130033 [2]中国科学院大学,北京100049 [3]西安建筑科技大学机电工程学院,西安710055

出　　处：《计算机辅助设计与图形学学报》2015年第8期1498-1505,共8页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61405191);吉林省自然科学基金(201115124)

摘　　要：针对靶场光电测量设备采集到的红外图像存在对比度低和边缘模糊的问题,提出结合直方图均衡化和模糊集理论图像的增强方法。首先采用对数隶属度函数将图像转换到模糊域,再对图像进行多次非线性变换,增强边缘信息;其次提出改进的自适应平台直方图均衡化算法,提高红外图像对比度;最后将2幅图像按加权系数相乘后叠加融合出最终图像。实验结果证明,该方法克服了传统增强算法易出现的过增强和局部信息被弱化等问题,增强后的图像细节丰富,具有更好的视觉效果,能够满足靶场的实际要求,具有广泛的工程应用价值.Infrared images collected by optical measure equipment are low contrast and with fuzzy image edge. In order to solve these problems, a novel method combined with platform histogram equalization and fuzzy theory is proposed. Firstly, it uses logarithm membership function to translate gray image into the fuzzy domain, then image＇s edges are carried out non-linear transformations in the fuzzy domain to enhance image＇s edges information. Next, the method combined with improving and adaptive platform histogram equalization is presented to improve the image contrast effectively. Finally, the fusion of two images multiplied by weighting coefficients is applied. Experimental results show that this algorithm has an excellent enhancement effect, which overcomes some shortcomings that traditional algorithms showed, such as the phenomenon of the excessive enhancement and weakening of the local gray information. Meanwhile there are more abundant details and more pleasing visual content in the enhanced images, which enables to satisfy the demands of shooting range and is of great and wide value for engineering application.

关 键 词：图像增强 红外图像 自适应平台直方图均衡 模糊集理论 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]