VHDL implementation of pattern based template matching in satellite images
Milchah V Thomas, C. Kanagasabapthi, Siva Sankar Yellampalli · 2017 International Conference On Smart Technologies For Smart Nation (SmartTechCon) · 2017
Image Template Matching is one of the basic problems in image processing and has applications in the field of pattern recognition and computer vision. It is a useful operation for filtering, edge detection, target detection and image registration. Template Matching is the process of determining the presence and the location of a template or pattern of an object inside the source image by using any similarity measure techniques. Conventional correlation based methods are computation expensive but have simple equation that helps in easier hardware implementation for real-time application. In this paper, a more accurate Template Matching algorithm for larger images based on correlation coefficient (normalized form of correlation) is presented with implementation in MatLab tool. Taken the advantage of template size wise processing and making use of output of generalized target detection as a mask, this algorithm can achieve high computation efficiency and good matching results even if, multiple templates may be or may not be present in the source image are involved in matching process. A VHDL implementation of this algorithm is also performed in Altera Quartus II tool.