2011 | OriginalPaper | Chapter
An Accelerated Approach of Template Matching for Rotation, Scale and Illumination Invariance
Authors : Tanmoy Mondal, Gajendra Kumar Mourya
Published in: Control, Computation and Information Systems
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Template matching is the process for determining the presence and location of a certain reference in the scene sub image. The conventional approach like spatial correlation, cross correlation is computationally expensive and error prone when the object in the image is rotated or translated. The conventional approach gives erroneous result in the presence of variation in illumination of the scene sub image. In this paper, an algorithm for a rotation, scale and illumination invariant template matching approach is proposed, which is based on the combination of multi-resolution wavelet technique, projection method and Zernike moments. It is ideally suited for highlighting local feature points in the decomposed sub images, and the result is computationally saving in the matching process. As demonstrated in the result, the wavelet decomposition along with Zernike moments makes the template matching scheme feasible and efficient for detecting objects in arbitrary orientation and rotations.