基于分数阶微分的图像边缘检测算法

Image Edge Detection Algorithm Based on Fractional Differential

针对传统边缘检测算法对于图像边缘提取存在边缘缺失、不连续等问题,为提高边缘的完整性与连续性,提出一种基于分数阶微分的边缘检测算法。由G-L定义构造分数阶微分掩模算子,使用不同阶次的算子对高、低频图像分别进行边缘提取,然后将两部分边缘进行融合,最终得到连续完整的图像边缘。实验结果表明,该算法不仅提高了边缘信息的完整性,还保留了更多的纹理细节及获得了较好的抗噪性能。

In order to improve the integrity and continuity of edge,an edge detection algorithm based on fractional order differentiation is proposed to solve the problem of edge missing and discontinuity in traditional edge detection algorithm.The fractional differential mask operator is constructed by the definition of G-L.The operators of different orders are used to extract the edges of high-frequency and low-frequency images respectively,and then the two parts of edges are fused to obtain continuous and complete image edges.Experimental results show that the algorithm not only improves the integrity of edge information,but also retains more texture details and achieves better anti noise performance.