A novel line detection algorithm based on endpoints estimation

Line segment is the most important feature for shape analysis and object recognition. In this study, a novel line detection method starting from endpoints is proposed. Firstly, the junctions and breakpoints are detected from the edge map. Then the connected edge pixels between any two junctions or breakpoints will be extracted, and taken as a line segment or a curve by using small eigenvalue analysis. Next, the corners are detected from each curve. In this way, each curve is divided into several parts by the corners, and each part between two corners is identified by using the small eigenvalue analysis. We have tested our algorithm on a wide collection of images and compared it with a lot of proposed line detectors. The results show that our algorithm is efficient for extracting line segments from an image.

[1]  Hungwen Li,et al.  Fast Hough transform: A hierarchical approach , 1986, Comput. Vis. Graph. Image Process..

[2]  N. H. C. Yung,et al.  Curvature scale space corner detector with adaptive threshold and dynamic region of support , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[3]  Rafael Grompone von Gioi,et al.  LSD: A Fast Line Segment Detector with a False Detection Control , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Josef Kittler,et al.  The Adaptive Hough Transform , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  丁伟利 Ding Weili,et al.  Extracting Straight Lines from Building Image Based on Edge Orientation Image , 2010 .

[7]  P. Nagabhushan,et al.  A simple and robust line detection algorithm based on small eigenvalue analysis , 2004, Pattern Recognit. Lett..

[8]  Cuneyt Akinlar,et al.  EDLines: A real-time line segment detector with a false detection control , 2011, Pattern Recognit. Lett..

[9]  Erkki Oja,et al.  A new curve detection method: Randomized Hough transform (RHT) , 1990, Pattern Recognit. Lett..

[10]  Richard Egli,et al.  Old and new straight-line detectors: Description and comparison , 2008, Pattern Recognit..

[11]  Chang-Sung Jeong,et al.  A straight line detection using principal component analysis , 2006, Pattern Recognit. Lett..

[12]  J. B. Burns,et al.  Extracting straight lines , 1987 .