Premium
Rotation‐invariant image matching by geometrical parameter of equidistant edge pixel group
Author(s) -
Hayashi Hiroki,
Saitoh Fumihiko
Publication year - 2007
Publication title -
electrical engineering in japan
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.136
H-Index - 28
eISSN - 1520-6416
pISSN - 0424-7760
DOI - 10.1002/eej.20520
Subject(s) - equidistant , pixel , template matching , enhanced data rates for gsm evolution , computer vision , artificial intelligence , rotation (mathematics) , matching (statistics) , image (mathematics) , invariant (physics) , mathematics , position (finance) , computer science , geometry , statistics , finance , economics , mathematical physics
Template matching is a method of searching for a target image area that is similar to a template image. However, the target image area generally must have the right position relative to the template image. Therefore, the computational cost is high when the location and the inclination of the target image area are unknown. This paper proposes a method of searching for a target image area with arbitrary location and an arbitrary inclination by means of geometrical parameters obtained by an equidistant edge group. In this method, the difference of directions is calculated from the edge direction of a referenced edge pixel and from the directions from the referenced edge pixel to other selected edge pixels. After that, matching is performed, using the difference in directions and the matching rates are accumulated. Then the location and inclination of the target image area are sought by scanning the accumulation space. Experimental results show that the proposed method has higher reliability and lower computational cost for image matching. © 2007 Wiley Periodicals, Inc. Electr Eng Jpn, 161(1): 34–42, 2007; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20520