基于加权直方图均衡化彩色图像增强仿真  

Color Image Enhancement Simulation Based on Weighted Histogram Equalization

作　　者：王利娟 常霞[1,2] 任旺 WANG Li-juan;CHANG Xia;RENG Wang(Key Laboratory of Intelligent Information and Big Data Processing of Ningxia Province,North Minzu University,Yinchuan Ningxia 750021,China;School of Math and Information Science,North Minzu University,Yinchuan Ningxia 750021,China;School of Computer Science and Engineering,North Minzu University,Yinchuan Ningxia 750021,China)

机构地区：[1]北方民族大学宁夏智能信息与大数据处理重点实验室,宁夏银川750021 [2]北方民族大学数学与信息科学学院,宁夏银川750021 [3]北方民族大学算机科学与工程学院,宁夏银川750021

出　　处：《计算机仿真》2021年第12期126-131,共6页Computer Simulation

基　　金：自然科学基金项目(11961001);宁夏自然科学基金项目(2018AAC03126);宁夏高等学校一流学科建设(NXYLXK2017B09);北方民族大学重大专项项目(ZDZX201801);宁夏智能信息与大数据处理重点实验室开放基金(2019KLBD004);北方民族大学2020年研究生创新项目(YCX20092)。

摘　　要：针对传统直方图均衡化(HE)算法在处理低质量图像时出现的细节信息丢失、部分灰阶合并的不足,提出一种彩色图像在HSV空间的加权HE图像增强算法。通过增加权重参数实现自适应调节图像像素信息的目的,算法构造了一种新的自适应映射函数。图像灰度级经过两次不同的映射有效地克服灰阶合并和细节丢失现象。对于输入图像进行颜色幅度拉伸和饱和度信息拉伸至最大化,完成对图像颜色信息的恢复。仿真结果证明,算法弥补了其它算法处理结果中出现的对比度低下和颜色信息丢失等不足,算法有效地改善了图像的质量。Aiming at the loss of detail information and the lack of partial gray level merging in the traditional his-togram equalization(HE) algorithm when processing low-quality images, a weighted HE image enhancement algo-rithm for color images in HSV space is proposed. By increasing the weight parameters to achieve the purpose of adap-tively adjusting the image pixel information, the algorithm constructs a new adaptive mapping function. The gray levelof the image undergoes two different mappings to effectively overcome gray level merging and loss of detail. For the in-put image, the color amplitude and saturation information are stretched to maximize, and the color information of theimage is restored. Simulation experiment results prove that the algorithm compensates for the low contrast and colorinformation loss in the processing results of other algorithms, and the algorithm effectively improves the image quality.

关 键 词：颜色幅度拉伸 颜色空间 加权直方图 直方图重新映射 黄金分割 饱和度拉伸 

分 类 号：TP317.[自动化与计算机技术—计算机软件与理论]