CAJ | 학술논문

直方图规定化是图像增强领域一个常用的算法，文中提出一种通过高斯函数加权的直方图规定化的图像去雾算法。首先通过分析晴天与雾霾天气下图像的直方图的特点，提出一种通过对高斯函数中方差的改变和高斯函数的加权的方式，解决了原有的单纯高斯函数直方图规定化图像偏暗的问题。通过实验图像的对比可以看出，文中提出的算法可有效去除雾霾天气的影响，其处理效果明显优于直方图规定化算法，而且计算量小、处理速度快、不需要人工干预。
직방도규정화시도상증강영역일개상용적산법，문중제출일충통과고사함수가권적직방도규정화적도상거무산법。수선통과분석청천여무매천기하도상적직방도적특점，제출일충통과대고사함수중방차적개변화고사함수적가권적방식，해결료원유적단순고사함수직방도규정화도상편암적문제。통과실험도상적대비가이간출，문중제출적산법가유효거제무매천기적영향，기처리효과명현우우직방도규정화산법，이차계산량소、처리속도쾌、불수요인공간예。
Histogram specification is a commonly used algorithm in image enhancement field. Propose an image haze removal algorithm of histogram specification based on the weighted Gaussian probability density function ( Gaussian PDF) in this paper. Firstly,analyzing the characteristics of image histogram that captured in sunny,fogging and haze weather. Then,solve the weak intensity problem of image specification of the single Gaussian function through changing the variance and weighted Gaussian PDF. The experimental results show the algorithm is able to remove the fog effectively,which is superior to the some existing algorithms of histogram specification about effi-ciency. It also has many advantages such as low computation,fast processing speed,no manual intervention.