Due to requirements and necessities in digital image research, image matching\nis considered as a key, essential and complicating point especially for\nmachine learning. According to its convenience and facility, the most applied\nalgorithm for image feature point extraction and matching is Speeded-Up\nRobust Feature (SURF). The enhancement for scale invariant feature transform\n(SIFT) algorithm promotes the effectiveness of the algorithm as well as\nfacilitates the possibility, while the application of the algorithm is being applied\nin a present time computer vision system. In this research work, the aim\nof SURF algorithm is to extract image features, and we have incorporated\nRANSAC algorithm to filter matching points. The images were juxtaposed\nand asserted experiments utilizing pertinent image improvement methods.\nThe idea based on merging improvement technology through SURF algorithm\nis put forward to get better quality of feature points matching the efficiency\nand appropriate image improvement methods are adopted for different\nfeature images which are compared and verified by experiments. Some\nresults have been explained there which are the effects of lighting on the underexposed\nand overexposed images.
Loading....