Article ID Journal Published Year Pages File Type
10360248 Image and Vision Computing 2005 7 Pages PDF
Abstract
In this paper, a new edge detection approach combining Zernike moments operator with Sobel operator is proposed. This detection consists of two steps: Firstly, all probable edge points are detected using Sobel operator; Secondly, Zernike moments operator is used to relocate the edge precisely from the points detected by Sobel operator. In the second step, two masks (one real and one complex) are deduced based on Zernike moments theory, and a new criterion for edge point, i.e. magnitude of the complex mask, has been established. Testing of this new detection approach demonstrates the advantage of high precision in subpixel as Zernike moments operator and short run time 79% more efficient than Zernike moments operator.
Related Topics
Physical Sciences and Engineering Computer Science Computer Vision and Pattern Recognition
Authors
, , , ,