A novel method of straight-lines detection based on Hypercomplex Steerable Filters

• Published in: Proceedings of the 33rd Chinese Control Conference pp. 7361 - 7364
• Main Authors: Li, Ce, Pan, Xing, Xiao, Limei, Song, Wenyang, Pan, Zhengrong
• Format: Conference Proceeding
• Language: English
• Published: TCCT, CAA 01.07.2014
• Subjects: Color; Directional Filters; Filtering algorithms; Hypercomplex Steerable Filters; Image color analysis; Image edge detection; Information filtering; Interpolation; Quadrature Filters; Ridge Energy; Straight-lines Detection
• DOI: 10.1109/ChiCC.2014.6896221

Straight-lines detection is an important task in image process and image analysis. For the traditional algorithm, nature images sometimes have the problem with detection inaccurate and without the use of full color information. In order to solve those problems, we proposed a novel image straight-lines detection algorithm based on Hypercomplex Steerable Filters. Firstly, the Hypercomplex steerable directional filters and Hypercomplex steerable quadrature filters are constructed. Secondly, the color images of natural scenes are detected by them. Finally, straight-lines detection will be obtained by ridge energy of color images. The proposed straight-lines algorithm also can be used in electric power transmission line detection. The experimental results demonstrate the proposed straight-lines algorithm can achieve a better effect compared with the traditional methods.