Title :
Primitives segmentation and association for a line character
Author :
Chen, Yung-Sheng
Author_Institution :
Dept. of Electr. Eng., Yuan-Ze Inst. of Technol., Taoyuan, Taiwan
Abstract :
An effective approach to perform segmentation and association among line primitives and junctions for a line image is presented. The concerned problem is that the sensitivity and liability of lines appears in conventional thinning methods. This demerit become an impediment to extract effective line features in the neighborhood of a junction. The proposed approach can not only segment the line primitives and junctions but can also construct their relationships for further interpretation of the given line image
Keywords :
edge detection; feature extraction; image segmentation; optical character recognition; association; image interpretation; junctions; liability; line character; line feature extraction; line image; line primitives; segmentation; sensitivity; thinning methods; Character recognition; Filters; Image analysis; Image segmentation; Impedance; Labeling; Optical character recognition software; Pattern recognition; Shape; Skeleton;
Conference_Titel :
Document Analysis and Recognition, 1993., Proceedings of the Second International Conference on
Conference_Location :
Tsukuba Science City
Print_ISBN :
0-8186-4960-7
DOI :
10.1109/ICDAR.1993.395790
