SOO-CHANG PEIHorng, Ji-HweiJi-HweiHorng2009-03-042018-07-062009-03-042018-07-06199501678655http://ntur.lib.ntu.edu.tw//handle/246246/142436https://www.scopus.com/inward/record.uri?eid=2-s2.0-0029326371&doi=10.1016%2f0167-8655%2895%2980007-G&partnerID=40&md5=a652f0da11272beb5c96209aa0996628A new method is proposed to detect circular arcs. The detection of a circular arc includes the determination of its center, radius, length, and midpoint. The Hough transform is applied to detect circular arcs by using center and radius as parameters. A peak value in the parameter space of the Hough transform indicates the existence of a circular arc. The center and radius of the circular arc are given by the coordinates of the peak. The arc points are separated from the remaining edge points of the image. Then, the near-peak Hough transform data of the arc points is analyzed to determine the arc length and midpoint. © 1995.application/pdf917027 bytesapplication/pdfen-USCircular arc; Covariance matrix; Hough transformAlgorithms; Edge detection; Mathematical transformations; Matrix algebra; Parameter estimation; Circular arc; Covariance matrix; Hough transform; Pattern recognitionjournal article10.1016/0167-8655(95)80007-G2-s2.0-0029326371http://ntur.lib.ntu.edu.tw/bitstream/246246/142436/1/24.pdf