Edge detection via a fast and adaptive bidimensional empirical mode decomposition

Abstract

This paper presents a new approach of edge detection utilizing bidimensional empirical mode decomposition (BEMD) technique. For this purpose a recently developed fast and adaptive BEMD (FABEMD) is employed to decompose the given image into several bidimensional intrinsic mode functions (BIMFs). In FABEMD, order statistics filters (OSFs) are employed to get the upper and lower envelopes in the decomposition process, instead of the surface interpolation, which enables fast decomposition and well characterized BIMFs. Since the first BIMF provides the highest local spatial variations and/or scales of the image, this BIMF is then processed for obtaining the edge. Binarization and morphological operations are applied as post processing operations to the first BIMF to achieve the desired edge map. The proposed method is compared with two other standard techniques namely, Canny and Sobel edge operators. Simulation results with real images demonstrate the efficacy of the proposed algorithm for edge detection.