Title :
Research on feature points extraction and matching technology of pipe-wall images
Author :
Li, Songsong ; Li, Xiang ; Chen, Xiaoming ; Deng, Changhui
Author_Institution :
Dalian Ocean Univ., Dalian, China
Abstract :
Pipeline defect recognition is necessary to achieve the online inspection of pipeline. In this paper, the method of pipeline visual inspection is proposed. According to the principle of pipe imaging, Harris corner extraction operator is used to present the NCC algorithm of corner detection and automatic point matching. The feature point identification and matching algorithm of pipeline defect is studied, and the corresponding analytical result indicate the extraction and matching algorithm of pipeline visual inspection is effective.
Keywords :
edge detection; feature extraction; image matching; inspection; pipelines; structural engineering computing; Harris corner extraction; NCC algorithm; feature points extraction; matching technology; pipe imaging; pipe wall images; pipeline defect recognition; pipeline visual inspection; Algorithm design and analysis; Error analysis; Feature extraction; Image edge detection; Inspection; Pipelines; Robots;
Conference_Titel :
Intelligent Control and Information Processing (ICICIP), 2010 International Conference on
Conference_Location :
Dalian
Print_ISBN :
978-1-4244-7047-1
DOI :
10.1109/ICICIP.2010.5564303
