联合K-means和形态学算子的图像边缘检测

Image Edge Detection of Joint K-means and Morphological Operator

针对传统边缘检测算子边缘定位精度的不足,提出了一种联合K-means和形态学算子的图像边缘检测方法。首先通过K-means算法对图像进行分割,再对分割后的图像求反,然后利用形态学作腐蚀操作,挖去图像内部像素点,最后利用分割图像减去形态学腐蚀后的图像获得图像边缘。实验结果表明,该方法能有效获得精确的图像边缘,比其他几种传统方法具有更优的检测效果。

Aiming at the shortcoming of edge localization accuracy of traditional edge detection operator,an edge detection method based on K-means and morphological operator is proposed. Firstly,the image is segmented by K-means algorithm,and then the segmented image is inverted. Then,the internal pixel of the image is dug by using morphological corrosion. Finally,the edge image is obtained by subtracting the segmentation image and the image of morphological corrosion. The experimental results show that the proposed method can obtain the accurate edge of the image effectively,and has better detection effect than other tradi. tional methods.