基于归一化标量权重映射与融合金字塔的图像对比度增强算法研究  

Research on Image Contrast Enhancement Algorithm Based on Normalized Scalar Weight Map Coupled Fusion Pyramid

作　　者：马永强 李瑞俊 李俊芳 董芳 Ma Yongqiang, Li Ruijun, Li Junfang, Dong Fang(Department of Computer, Jining Normal University, Ulanqab Inner Mongolia 012000, China)

机构地区：[1]集宁师范学院计算机学院,内蒙古乌兰察布012000

出　　处：《信息与电脑》2018年第20期42-43,共2页Information & Computer

基　　金：内蒙古自治区高等学校科学研究项目"基于归一化标量权重映射与融合金字塔的图像对比度增强算法研究"(项目编号:NJZY16318)

摘　　要：为了弥补目前全局增强与局部增强存在的不足,并有效解决已有对比度增强算法在高信号区域和图像纹理区域经常出现的颜色变化、增强过度等问题,提出定义对比度和亮度度量模型,构造标量权重映射并进行归一化处理,制定融合规则对高斯金字塔和拉普拉斯金字塔进行融合,利用融合金字塔进行图像重构。在MATLAB系统中进行实验,与其他算法比较,该算法的对比度图片增强质量最好。In order to solve the deficiency of global content enhancement and local details enhancement, as well as to solve these defects such as color change and over enhancement in the high signal activity area or image texture,this paper defined the contrast and brightness metric model to construct the scalar weight map and then normalize it,then defined fusion rules of the Gaussioan pyramid decomposition and Laplacian pyramid decomposition,used the fusion pyramid to finish the image reconstruction.Experiments results based on MATLAB show that the contrast enhancement quality of this algorithm is best.

关 键 词：归一化 标量权重映射 融合金字塔 对比度增强 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]