An Edge Detection Method Based on Directional Tracing
Edge detection is an essential preprocessing step in image analysis and recognition. In this paper, a novel edge detection method based on direction tracing is proposed. Based on our understanding of edges in a gray level image, an edge can be defined as a set of pixels where gray values vary sharply in some direction and slowly in its orthogonal direction. Each edge point has a direction, which is determined according to the absolute values of derivatives in four directions, horizontal, vertical, 45 degree and 135 degree directions. Two thresholds are chosen with statistical method to ensure the important features are grasped. Weaker edges are traced by sharp edges in terms of direction information. Nonmaximum suppression guarantees only one response to a single edge. Experimental results show the advantages of the proposed method compared with the conventional method such as Canny and Sobel edge detector.
KeywordsEdge Detection Edge Point Edge Pixel Lena Image Gray Level Image
Unable to display preview. Download preview PDF.
- 3.Gunn, S.R.: Edge detection error in the discrete Laplacian of a Gaussian. In: Proc. 1998 International Conference on Image Processing, vol. 2, pp. 515–519 (1998)Google Scholar
- 5.Jahne, B.: Digital Image Processing, 6th edn. Springer, Berlin (2005)Google Scholar
- 10.Torre, V., Poggio, T.: On edge detection. A. I. Memo768. MIT (1984)Google Scholar
- 11.Prewitt, J.M.S.: Object enhancement and extraction. In: Lipkin, B.S., Rosenfeld, A. (eds.) Picture Processing and Psychopictorics., Academic Press, New York (1970)Google Scholar
- 12.Roberts, L.G.: Machine perception of three-dimensional solids. In: Tippet, J.T. (ed.) Optical and Electro-Optical Information Processing., MIT Press, Cambridge (1965)Google Scholar
- 13.Sobel, I., Feldman, G.: A 3x3 Isotropic Gradient Operator for Image Processing, presented at a talk at the Stanford Artificial Project (1968),; unpublished but often cited, orig. In: Duda, R., Hart, P. Pattern Classification and Scene Analysis, pp. 271–272. John Wiley and Sons (1973)Google Scholar