Detecting Object by Affine Transform Using Line

Abstract:

Hough transform is an effective way in object recognition and applied to many industrial processes. Based on the principle of Hough transform, a new algorithm which can detect objects through an affine transform was proposed in this paper. First, application of Hough transform to extract straight lines in a model image and a scene image, got these coordinates of the lines, sorted according to the direction angle. Because of affine transform and the periodic direction angle, the direction order of the lines on scene image were different from those on the model image, these lines on scene image were expanse a cycle. Finally affine transform parameters were applied to objects detection. The results showed the effectiveness of the algorithm.

• Periodical: Advanced Materials Research (Volumes 490-495)
• Pages: 1306-1310
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.490-495.1306
• Online since: March 2012
• Authors: Jing Li, Huai Yu Liu, Liu Rong Hong
• Keywords: Affine Transform, Hough Transform (HT), Object Detection

References

[1] Chau C. P, Siu W C, Adaptive dual-point Hough t ransform for object recognition, Computer Visionand Image Understanding , vol 96, no 1, pp.1216-1225, (2004).

DOI: 10.1016/j.cviu.2004.04.005

[2] Hiroyuki Nishi, Yoshimasa Kimura, and Takeshi Ishibashi, Word spotting framework using Hough transform of cepstrum distance matrix images, Proc. YKJCA2009, p.204, (2009).

[3] Satadal Saha, Subhadip Basu, et al, A Hough Transform based Technique for Text Segmentation, Journal of computing. vol 10, no 2, pp.134-141, (2010).

[4] Chen Linpeng, Zhang Guoliang, A New Algorithm for Line Detection Based on the Randomized Hough Transform, The 8th International Conference on Electronic Measurement and Instruments, pp.622-625, (2007).

DOI: 10.1109/icemi.2007.4350759

[5] Ballard D H, Generalizing the Hough transform to detect arbit rary shapes, Pattern Recognition , vol 13, no 2, pp.1112-122, 1981.

DOI: 10.1016/0031-3203(81)90009-1

[6] Mingwei Wang, Xiaolei Wang, Recognition of multibreak patterns by 8-neighborhood-based General Hough Transform, International Journal for Light and Electron Optics, vol 56, no 6, pp.3234-3239, (2007).

[7] Liu HongS, Chen Jian, A Improved General Hough Transform, , Journal of university of sciece and technology of China, vol 39, no 11, pp.1202-1206, (2009).

[8] Y. Furukawa and Y．Shinagawa, Accurate and robust line segment extraction by analyzing distribution around peaks in Hough space, Computer Vision and Image Understanding, vol 92, no 1, pp.1-25, (2003).

DOI: 10.1016/j.cviu.2003.07.002