摘要： 为提高图像边缘检测的精度与抗噪性能，提出一种基于蚁群优化的图像边缘检测算法。将图像像素梯度值和像素圆形邻域统计均值的相对差共同作为蚁群的启发信息，引导蚁群搜索图像边缘。实验结果表明，该算法能最大限度地保留边缘细节，并能抑制噪声和纹理，具有较好的鲁棒性。

关键词: 边缘检测, 蚁群优化, 特征提取, 梯度, 统计均值

Abstract: In order to improve the image edge detection accuracy and noise performance, this paper proposes an image edge detection algorithm based on ant colony optimization. The value of pixel gradient and the relative difference of statistical means of the pixel’s circle neighborhood are combined to be the heuristic information which can guide ant’s searching. Experimental results show that the edge detected by the proposed algorithm is robust to noise and texture, and contains most of the edge details.

Key words: edge detection, ant colony optimization, feature extraction, gradient, statistical mean

中图分类号: 

• TP391