Translator Disclaimer
9 January 1984 High Speed Implementation Of Linear Feature Extraction Algorithms
Author Affiliations +
This paper describes a novel digital implementation of the Hough Transform for high speed extraction of linear features from images of complex scenes. In addition to detection, the Hough Transform estimates both location and orientation of a linear feature. The implementation is based on using the image intensity histogram generator that is available as a high speed hardware feature on some commercial digital image display and processors. It is shown that linear feature extraction is reduced to detection of peaks in a number of onedimensional histogram arrays. The number of histogram generation operations is equal to the number of angles into which the 0°-18° range is quantized. Using this method, the processing time for extraction of a linear feature using 16 angles is on the order of 0.5 second. Unlike regular implementations of the Hough Transform, the amount of computation is completely independent of the number of detected pixels (pixels with value 1) in the binary input image to the Rough Transform.
© (1984) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Hassan Mostafavi and Mukesh Shah "High Speed Implementation Of Linear Feature Extraction Algorithms", Proc. SPIE 0432, Applications of Digital Image Processing VI, (9 January 1984);

Back to Top