耦合H-minima与数学形态学的分水岭遥感图像分割方法

An Improved Watershed Method for Remote Sensing Image Segmentation Coupling H-minima with Mathematical Morphology

传统的分水岭方法分割遥感图像易受原始图像本身局部纹理、噪声等因素的影响而产生过分割的现象。为解决这个问题,提出了一种改进的分水岭遥感图像分割方法。该方法引入自适应滤波算法和数学形态学标记方法,构造一个自适应滤波器来削弱噪声、局部纹理等干扰。首先利用自适应滤波器对图像噪声剔除,然后采用数学形态梯度算子对图像进行梯度运算与开闭重构和进行H-minima变换得到极小值标记图像,最后再利用分水岭算法分割图像。实验结果表明,与传统的分水岭算法和经典的分割方法相比,本研究改进的方法可减少过分割区域,能获得连续封闭的边缘,显著提升了边缘定位精度。

The conventional watershed algorithm for image segmentation is sensitive to noise and texture,which often generate over-segmented results.To solve this problem,an improved watershed segmentation for remote sensing image is proposed.An adaptive filtering algorithm and a morphological marker method are introduced to weaken the influence of noise and texture.Firstly,the remote sensing image is enhanced by adaptive filtering.Then,the gradient of the image obtained from the adaptive filtering is calculated by a morphological gradient operator and reconstructed by opening and closing,and the minimum labeled image is obtained by H-minima transformation.Finally,the watershed algorithm is used to segment the remote sensing image.The experimental results show that the proposed method can reduce the over-segmented area,obtain continuous closed edges,and significantly improve the accuracy of edge location compared with the conventional watershed algorithm.