高分辨率航空影像阴影自动检测和去除  

Automatic Shadow Detection and Removal in High Resolution Aerial Imagery

作　　者：涂继辉[1,2] 杜红[1] 眭海刚[2] 徐川[2] TU Jihui;DU Hong;SUI Haigang;XU Chuan(Electronics&Information School,Yangtze Universtiy,Jingzhou,Hubei 434023,China;State Key Laboratory of Remote Sensing and Information Engineering,Wuhan University,Wuhan 430079,China)

机构地区：[1]长江大学电信学院,湖北荆州434023 [2]武汉大学测绘遥感信息工程国家重点实验室,武汉430079

出　　处：《遥感信息》2020年第2期38-45,共8页Remote Sensing Information

基　　金：武汉大学测绘遥感信息工程国家重点实验室开放基金项目(16E01);国家自然科学基金项目(41771457);国家重点研发计划项目(2016YFB0502600)。

摘　　要：针对高分辨率航空影像的阴影检测与去除问题,通过对阴影的光谱特征的分析,提出了一种新的自动化阴影检测和去除的方法。该方法首先利用约束颜色不变量和水平集分割相结合的方法对阴影进行自动检测,获取完整的阴影区域和精确的阴影轮廓线;然后利用阴影轮廓点的邻域作为同质区域,进行补偿因子估算达到去除阴影。实验证明,该方法能够自动准确地对阴影进行检测和去除,不但提升了阴影区域的亮度,而且使得地物细节得到了很好的恢复。In order to solve the problem that shadow interferes the information extraction in high resolution aerial image,this paper put forwards a novel automatic shadow detection and removal method after analyzing the characteristics and the mathematical model of shadow.This method firstly automatically detects the shadow area based on color invariants method in the condition constraint and using the level set evolution method,which obtains the complete area and accurate boundary of the shadow.Then,the neighborhood window pixels of the shadow boundary are used as the homogeneous region to estimate the compensation factor,which removes the shadow.Experimental results show that this method can automatically detect the shadow area and remove the shadow,and it makes the brightness of shadow area heightened and the details of sheltered reappeared efficiently.

关 键 词：高分辨率航空影像 颜色不变量 水平集 阴影检测 阴影去除 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]