基于 Shearlet 域方向模极大值和改进蜂群的图像边缘检测

摘要

In order to extract more accurate and clearer edges from an image ,a method of image edge detection based on direction modulus maxima and improved bee colony in Shearlet domain is proposed . Firstly , the im‐age is decomposed by nonsubsampled Shearlet transform . Then , for the low frequency component , the im‐proved bee colony algorithm is used to accurately detect the basic contour line of the image edge , w hile for the high frequency components , the direction modulus maxima algorithm is applied to detect the abundant details of the image edge . Finally , the edges of image w hich contain complete contour and abundant details are ob‐tained through fusion .Experimental results show that ,compared with the existing methods of edge detection such as Canny method , improved ant colony method , improved bee colony method , improved nonsubsampled Contourlet modulus maxima method , the image edges detected by the proposed method are located accurately and can be complete and clear , with abundant details . The proposed method can more effectively improve the performance of edge detection and requires less running time .%为从图像中提取出更为准确、清晰的边缘，本文提出一种基于Shearlet域方向模极大值和改进蜂群的边缘检测方法。对图像进行非下采样Shearlet分解；对于低频分量，利用改进的蜂群算法准确检测出边缘的基本轮廓线；而对于高频分量，采用方向模极大值算法检测出图像中丰富的边缘细节；融合后得到轮廓完整、细节丰富的图像边缘。实验结果表明：与Canny方法、改进的蚁群方法、改进的蜂群方法、改进的非下采样Contourlet模极大值方法相比，本文提出的方法检测出的图像边缘定位准确、完整清晰、细节丰富，边缘检测效果更好，且运行时间较少。