Current Issue Cover
基于物理模型的雾天图像复原新方法

陈功,王唐,周荷琴(中国科学技术大学自动化系,合肥 230027)

摘 要
由于雾天场景的能见度很低,因此为了保证视频监控系统的正常工作,需要对视频图像进行去雾处理。根据光学原理,雾天条件下场景的能见度下降是与场景深度呈指数关系的,鉴于现有的基于物理模型方法的去雾效果不够理想,因此提出了一种新的基于物理模型的图像复原方法,该方法首先对雾天场景的光学成像建模;然后借助于一张晴天和一张雾天场景的参考图像,计算出场景各点的深度比关系,最后利用深度关系复原雾天图像或雾天视频。实验结果证明,该方法是有效的和实用的。
关键词
A Novel Physics based Method for Restoration of Foggy Day Images

CHEN Gong,WANG Tang,ZHOU Heqin,CHEN Gong,WANG Tang,ZHOU Heqin,CHEN Gong,WANG Tang,ZHOU Heqin()

Abstract
Scene visibility is very low in foggy days We need to defog the surveillance video to make sure the intelligent visual surveillance system work normally According to optics theory, the reduction of scene visibility is exponential to the scene depth The defog effect of existing modelbased methods are not good enough In this paper, a new physicsbased method of image restoration is presented This method models scene points in foggy day firstly Then the scene depth is calculated for two images which are captured in clear day and foggy day respectively Image or video taken in foggy day is restored using depth information finally The experiments demonstrate the effectiveness and practicality of this method
Keywords
QQ在线


订阅号|日报