Title :
Edge detection for image processing using second directional derivative
Author_Institution :
Dept. of Mech. Eng., Kwang-Wu Inst. of Technol., China
Abstract :
One function of data preprocessing is to convert a visual pattern into an electrical pattern or to convert a set of discrete data into a mathematical pattern so that those data are more suitable for computer analysis. With regard to edge detection, we define an edge to occur in a pixel if and only if there is some point in the pixel´s area having a negatively sloped zero crossing of the second directional derivative which is taken in the direction of a nonzero gradient at the pixel should be marked as a step edge pixel, its underlying gray tone intensity surface (proper thresholding) must be estimated on the basis of the pixels in its neighborhood. Therefore, a functional form consisting of a linear combination of the tensor products of discrete directional derivatives are easily computed from this kind of function. Various results of computer analysis by using different window size and intensity function are included in this report. The best combination of the parameters for second directional derivative in edge finding will also be presented in this study
Keywords :
edge detection; computer analysis; data preprocessing; edge detection; edge finding; electrical pattern; gray tone intensity surface; image processing; negatively sloped zero crossing; proper thresholding; second directional derivative; step edge pixel; tensor products; visual pattern; Arithmetic; Equations; Image edge detection; Image processing; Mathematics; Polynomials;
Conference_Titel :
Industrial Automation and Control: Emerging Technologies, 1995., International IEEE/IAS Conference on
Conference_Location :
Taipei
Print_ISBN :
0-7803-2645-8
DOI :
10.1109/IACET.1995.527639