OpenCV–Based Image Matching Algorithm And It's Application To Target Location

Imageis an important source for mankind to obtain the information. With the rapid development of science and technology, acquisition and processing of images is no longer restricted to just humanity itself. This thesis detail an important issue of image processing: image matching. In this thesis, target location algorithm is researched based on many image location algorithms, improved chamfer matching algorithm and SURF-based image matching algorithm are proposed.Improved chamfer matching algorithm is often used in simple situations, it can locate target accurately.This algorithm is modified on the chamfer matching algorithm, it changes the calculation of shape distance, standard deviationfactor and directional differences factor are added in new shape distance. SURF-based image matching algorithmis often be used in complex situations, it can locate the center of target accurately. This algorithm can find out matching feature points set between template image and source image, but there are invalid feature points in this set. In this thesis, an effective feature points identify strategyand matching result searching strategy are presented, and this form aSURF-based image matching algorithm combined with SURF.In this thesis, plenty of matching experiments about improved chamfer matching algorithm and SURF-based image matching algorithm have been taken, and it is proved in the experiments that improved chamfer matching algorithm is more accurate, and it also proved that SURF-based image matching algorithm can be used in complex situations.