基于暗原色的航空遥感影像薄雾去除与改进  

An improved method of haze removal for aerial digital image based on dark channel prior

作　　者：赵桂华[1,2,3] 邹晓亮[1,3] 戴晨光[1] 方勇[2] 陈虹[2] 万欢[1] 

机构地区：[1]信息工程大学测绘学院,郑州450052 [2]西安测绘研究所,西安710054 [3]西安测绘信息技术总站,西安710054

出　　处：《测绘科学》2013年第1期29-32,共4页Science of Surveying and Mapping

基　　金：国家863项目(2008AA121203);江西东华理工大学数字国土重点实验室基金项目(DLLJ201108)

摘　　要：本文对航空遥感影像的薄雾进行了研究,根据雾图像成像模型和基于暗原色先验知识,结合图像对比度的自适应调整,提出了一种对航空遥感影像薄雾去除的改进算法。采用数字航空相机DSS439获取的不同地物特征类型的应急遥感影像进行去雾实验,并对实验结果进行分析、比较与评价,验证了该方法的正确性和有效性。The research was focused on haze of aerial digital image. An improved haze removal method was proposed based on haze imaging model, dark channel prior and adaptive contrast enhancement. Using different feature images acquired by Digital Sensor System 439 on an emergency response flight test, de-haze experiment was made. The detailed analysis and images quality assessment between the original and result images were accomplished. The experimental results showed that the method was feasible and available.

关 键 词：航空数字遥感影像 暗原色先验知识 雾图像成像模型 薄雾去除 自适应对比度增强 

分 类 号：P231[天文地球—摄影测量与遥感] TP75[天文地球—测绘科学与技术]