Welcome to Acta Armamentarii ! Today is

Acta Armamentarii ›› 2015, Vol. 36 ›› Issue (5): 854-860.doi: 10.3969/j.issn.1000-1093.2015.05.013

• Paper • Previous Articles     Next Articles

Infrared Target Recognition Algorithm Based on Contour Fragment Matching and Graph Searching

SU Juan1, YANG Luo2, ZHANG Yang-yang1   

  1. (1.907 Section, The Second Artillery Engineering University, Xi’an 710025, Shaanxi,China;2.Meteorological Office of 96427 Troops, PLA, Baoji 721006, Shaanxi, China)
  • Received:2014-04-08 Revised:2014-04-08 Online:2015-07-09
  • Contact: SU Juan E-mail:suj04@mails.tsinghua.edu.cn

Abstract: A target recognition method based on contour fragment matching and graph searching is proposed for complicated background interference and contour fragment missing in forward looking infrared (FLIR) target recognition. The contour features are extracted from the reference image and the real-time image, and decomposed into the fragments,and the orientation information of each fragment are utilized to realize the fragment matching and construct the matching graph. Under the constraint of the spatial relationship between the reference contour fragments, the graph searching strategy is used to traverse the matching graph to obtain candidate target contour fragment combinations. Finally, the global similarity measure is used to obtain the final recognition. Experimental results show that the proposed method has good adaptability and robustness to complicated background interference and contour fragment missing.

Key words: ordnance science and technology, target recognition, shape matching, fragment matching, graph searching, spatial relationship

CLC Number: