基于ENVI/IDL的遥感影像阴影区域内水陆分割线的提取方法

An approach for split line extraction of water and land from shadow area of remote sensing images based on ENVI/IDL

针对遥感影像阴影区域内水陆分割线的提取问题,提出一种结合C3分量和影像波段属性的提取方法。在ENVI/IDL软件平台下,将多光谱影像和全色波段影像融合,生成高分辨率多光谱影像,由RGB彩色空间构建C1C2C3彩色不变特征中的C3分量,以检测影像中的阴影部分,结合水体对近红外波段的强吸收特性,采用C3分量与NIR波段的比值图像,突出阴影区域水体与陆地的边界。利用Otsu阈值分割、形态学滤波算法和边缘检测,提取阴影区域的水陆分割线,并与直接用近红外波段分割方法比较。实验结果表明,本方法对阴影区域内的水陆分割线提取具有较高的精度和效率。

For extracting split line of water and land in the shadow area,an approach had been proposed which combined the C3 component and band properties of images.The algorithm in this paper implemented on ENVI/IDL software platform;firstly,By fusing multispectral images and panchromatic image to generate high resolution multi spectral image.For distinguishing shaded portion of the images,the C3 component of C1C2C3 color invariants had been established by RGB color space.The boundary of water and land in shadow area had been highlighted through ratio of C3 and NIR which based on strong absorption properties of water in near-infrared band.Finally,the split line extracted via Otsu threshold segmentation,mathematical morphology algorithm and edge detection.And compare with directly segmentation by NIR.The experimental results indicate that the proposed approach of extract split line of water and land performs higher precision and efficiency.