Co-Occurrence Morphological Edge Detection

Morphological edge detection plays a significant role in image processing and computer vision. An edge is the boundary between an object and the background, and indicates the boundary between overlapping objects also. However, edge detection is challenged and difficult to find the discontinuities in surface orientation, changes in material properties and variations in scene illuminations. In this paper, an improved co-occurrence morphology edge detection algorithm is proposed. We exploit the co-occurrence filter (a combination of bilateral filter and gray-level co-occurrence matrix) to handle with noisy images, and two novel structural elements are used for edge detection. The proposed edge detection algorithm can reasonably consider noise reduction and preserve the detailed edge information. The experimental results demonstrate the superior performance with some existing methods and exhibit better anti-noise ability.