遥感图像质量雾霾影响的无参考评估（英文）  

No-reference haze effects assessment for remote sensing images

作　　者：谢凤英[1] 潘小溪 姜志国[1] 尹继豪[1] 

机构地区：[1]北京航空航天大学图像处理中心,北京100191

出　　处：《中国体视学与图像分析》2015年第1期22-27,共6页Chinese Journal of Stereology and Image Analysis

基　　金：国家自然科学基金(61471016;61371134);中国科学院天文联合基金(U1331108)

摘　　要：遥感技术可以提供大量有效信息,在监测和测绘领域占有十分重要的地位。然而,遥感图像通常会受到雾霾影响而降低质量。雾霾浓度评价算法可以评价场景中的雾霾程度,以便对雾霾影响严重的图像进行滤除,并对雾霾较轻的图像根据雾霾指标的大小进行自适应的去雾,从而保证遥感图像解译的可靠性。本文针对这个问题提出了一个雾霾影响遥感图像质量的评价指标。首先,根据雾霾区域的亮度值大于非雾霾区域的亮度值这一特性定义了雾霾图。然后,根据韦伯准则计算雾霾图的对比度,以此作为评价雾霾的指标。实验结果表明,该指标的评价结果与主观评价结果具有很好的一致性,可以代替主观评价对受雾霾影响的遥感图像进行质量筛选。Remote sensing technology is crucial in monitoring and surveying since it can provide plenty of effective information. However, remote sensing images are usually degraded by haze. In order to filter images distorted by haze, a mechanism that can be used for evaluating the haze effects is imperative. In this paper, we propose a novel metric to assess the effect of haze on remote sensing images. Firstly, a definition of haze map is given according to the observation that the intensity in haze regions is higher than that in haze-free regions. Then, the contrast of haze map based on Weber＇ s law is calculated and the metric for the effects of haze is obtained. The experimental results show that the proposed assessment metric achieves good consistency when compared with the results obtained by subjective evaluation.

关 键 词：遥感图像 图像质量评价 雾霾影响 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]