一种基于高斯函数的直方图规定化算法  

A Histogram Specification Algorithm Based on Gaussian PDF

作　　者：肖斌[1] 王晅[1] 毕秀丽[1] 王振邦[1] 

机构地区：[1]陕西师范大学物理学与信息技术学院,陕西西安710062

出　　处：《铁道学报》2006年第4期119-122,共4页Journal of the China Railway Society

摘　　要：提出一种基于高斯函数的直方图规定化算法,本算法首先根据原图像的信息求出灰度平均值和灰度平均对比度,并利用这两个参数估计输出图像的灰度平均值和灰度平均对比度,然后利用输出图像的灰度平均值和灰度平均对比度构成高斯函数来规定输出图像的直方图,从而控制输出图像的灰度平均值和灰度平均对比度。此算法处理效果明显优于直方图均衡化算法,而且计算量小、处理速度快、不需要人工干预,可用于计算资源有限、实时性要求较高的场合。而且通过改变单一参量就可以控制输出图像的对比度,在工程应用中为用户提供了调控与选择手段。This paper proposes an algorithm of histogram specification based on the Gaussian probability density function （Gaussian PDF）. The proposed algorithm calculates the values of the average gray level and contrast of the input image, and uses these two parameters to estimate the average gray level and contrast of the output image, and then specifies the histogram of the output image with the Gaussian probability density function, the expectation and variance of which are respectively the average gray level and contrast. The performance of the algorithm is evaluated and experimental results show the superiority of the proposed algorithm compared with some existing algorithms of histogram equalization. It also has many advantages over other image enhancement methods inrespect of low computational complexity, high efficiency, no manual intervention and easy control of the contrast of the output image by alternating only one parameter, and it can be used in real time operations and systems with limited computational resources.

关 键 词：图像增强 高斯函数 空间域法 变换域法 模糊处理法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TP317.4[自动化与计算机技术—计算机科学与技术]