摘要
为了克服传统数字图像增强(LIP)模型图像增强的不足,提出了一种新的算法。该算法引入了模糊熵,充分利用了邻域视觉信息和人类视觉特征的模糊性。首先构造了新的非线性变换,增加了算法的普遍性;其次分析了传统图像增强算法中灰度值过校正的原因,并提出了解决方案;最后将图像的直方图合理的划分为亮区域,暗区域和中间区域提取3个区域的统计特征,并根据图像灰度值的统计特征自适应地选择参数α。仿真结果表明,该图像增强算法对图像的噪声有很好的抑制效果,增强对比度和视觉效果,锐化边缘并调整动态范围。
In order to overcome the shortcomings of traditional LIP model image enhancement,this paper proposes a new algorithm.Fuzzy entropy is introduced into the algorithm,which makes full use of the fuzziness of neighborhood visual information and human visual features.In this paper,firstly,a new nonlinear transformation is constructed to increase the universality of the algorithm;secondly,the reason of over correction of gray value in traditional image enhancement algorithm is analyzed,and the solution is proposed;finally,the histogram of the image is reasonably divided into bright region,dark region and intermediate region,and the statistical characteristics of three regions are extracted,and the adaptive selection is made according to the statistical characteristics of image gray value parameterα.Simulation results show that our image enhancement algorithm can effectively suppress image noise,enhance contrast and visual effect,sharpen edges and adjust dynamic range.
作者
王延年
钟正
Wang Yannian;Zhong Zheng(School of Electronic Information,Xi’an Engineering University,Xi’an 710048,China)
出处
《国外电子测量技术》
北大核心
2021年第3期10-15,共6页
Foreign Electronic Measurement Technology
基金
西安市科技局科技计划项目(201805030YD8CG14(1))
陕西省科技厅工业领域一般项目(2019GY-109)资助。
关键词
图像增强
模糊熵
模糊划分
人类视觉特征
image enhancement
fuzzy entropy
fuzzy partition
human visual characteristics
