结合模糊聚类与色彩空间的遥感图像阴影检测

Shadow Detection Based on FLICM Clustering Algorithm and Color Space for Remote Sensing Images

高分辨率遥感图像中的阴影会导致特征信息的丢失、物体的假色调和形状失真,严重影响图像的质量。针对现有阴影检测算法对遥感图像中非匀质阴影区域及亮阴影区域易漏检、检测精度低等问题,该文结合模糊聚类及HSV色彩特征提出了一种新的遥感图像阴影检测算法。首先,对图像进行预处理,剔除亮度、饱和度与阴影区相似的干扰地物;其次,利用基于局部信息改进的模糊聚类算法对图像进行结合邻域信息的聚类,克服噪声情况下的错误聚类;最后,结合图像在HSV色彩空间的H分量和RGB空间的灰度直方图进行多阈值提取阴影,得到阴影区域。实验证明,改进算法有机结合了阴影像素点的邻域信息及色彩不变特性,在亮阴影及非匀质阴影区域检测中明显提高了检测精度。

In the high-resolution remote sensing images,the existence of the shadow will lead to the loss of feature information and shape distortion of the object with false color,which seriously influences the quality of the images.Due to the undetected error and low detection accuracy of inhomogeneous and bright shadow area in remote sensing images with existing shadow detection algorithms,a new high-resolution image shadow detection method is proposed in this paper,which combines the fuzzy clustering with HSV color features.Firstly,the distractions with similar intensity and saturation to shadow areas are eliminated in pre-processing.Then,an improved fuzzy local information clustering algorithm is used for image clustering combining with neighborhood information,which overcomes the error in noise situation.Finally,the shadow region is detected by multi-threshold extraction with the H component in HSV color space and the gray level histogram of RGB space.The experimental results demonstrate that the proposed method,which combines the neighborhood information of the shadow pixel and the color invariance well,improves the performance of bright and inhomogeneous shadow detection.