摘要
在拼接的遥感影像中,往往会存在色调、亮度分布不均匀现象,严重影响影像的视觉效果,不利于判读分析。为校正遥感影像不均匀性,引入基于均值和方差的Retinex算法。实验通过单尺度Retinex算法、多尺度Retinex算法、色彩恢复的多尺度Retinex增强算法和基于均值和方差的Retinex算法分别对遥感影像中色调、亮度分布不均匀进行处理,并做对比分析。实验结果表明,基于均值和方差的Retinex算法与其他3种算法比较,在匀光、匀色以及保持影像清晰度方面效果明显,并且解决了影像色彩失真问题,达到较好的视觉效果。
Uneven distribution of hue and brightness, which affects the visual effects of images seriously, is very common in the stitched remote sensing images and is detrimental to interpretation and analysis. In order to correct the non-uniformity of remote sensing images, an Retinex algorithm based on mean and variance is introduced. Through ingle-scale Retinex algorithm, Multi-scale Retinex algorithm, Multi-scale Retinex algorithm with color restoration and Retinex algorithm based on mean and variance, the experiment deals with the uneven distribution of hue and brightness in remote sensing images and makes a comparative analysis. The experimental results show that the Retinex algorithm based on mean and variance is more effective than other three algorithms in dodging uniform color and preserving image clarity, and it solves the problem of image color distortion and achieves good visual effect.
作者
刘海龙
张安兵
王贺封
赵玉玲
田志秀
赵兵杰
LIU Hailong;ZHANG Anbing;WANG Hefeng;ZHAO Yuling;TIAN Zhixiu;ZHAO Bingjie(School of Earth Science and Engineering,Hebei University of Engineering, Handan 056038,China;School of Mining and Geomatics Engineering,Hebei University of Engineering, Handan 056038,China;College of Geoscience and Surveying Engineering,China University of Mining and Technology (Beijing) , Beijing 100083, China)
出处
《测绘地理信息》
2019年第4期107-110,共4页
Journal of Geomatics
基金
河北省研究生创新项目(16129026)
关键词
遥感影像
色调
亮度
校正
均匀
视觉效果
remote sensing image
hue
intensity values
correcting
uniformity
visual effects
