基于BEMD和自适应阈值的多尺度边缘检测

Multi-scale Edge Detection Based on EMD and Adaptive Thresholds

利用了一种称为经验模式分解(Empirical Mode Decomposition,EMD)的技术,在多尺度下对图像进行分解,得到了多尺度信息并结合自适应阈值提取不同尺度下的边缘。将此技术应用于图像过滤和边缘检测中,发展了一种基于二维经验模式分解(BEMD)的算法,在多尺度下或空间频率中提取信号特征。这些特征,称为内蕴模函数,可通过一个过滤过程得到。在二维过滤过程中,利用形态学算子来检测区域极大值并通过径向基函数(RBFs)进行曲面插值运算,然后利用边缘信息的多尺度特征合成多尺度边缘,得到了单像素宽边缘。通过计算机仿真实验,结果表明这种方法不仅能准确地检测出图像边缘,而且还有效地抑制了噪声,这种算法的性能优于传统的边缘检测算法。

This paper makes good use of the characteristic of a technique called the empirical mode decomposition （EMD） to decompose the image at multi-scale,the obtained multi-scale information was combined with adaptive thresholds to extract image edge at different scales.The main contribution of our approach is to apply the EMD to image filtering and edge detection.We developed an algorithm based on bidimensional empirical mode decomposition （BEMD） to extract features at multiple scales or spatial frequencies.These features,called intrinsic mode functions,are extracted by a sifting process.The bidimensional sifting process is realized using morphological operators to detect regional maxima and thanks to radial basis functions （RBFs） for surface interpolation.Then using the multi-scale characteristic of the edge information and synthesizing multi-scale edge,the image edge with only a pixel width was obtained.The method was confirmed by computer simulation and the experiments results demonstrate that this method can not only detect image edge precisely,but also effectively restrain noise,and this algorithm performs better than the conventional image edge detection algorithms.