Vehicle based Images Matching by Improved SURF Algorithm

Abstract:

Aimed at the matching problem for color images obtained by vehicle based mobile mapping system, an improved algorithm is proposed based on speeded up robust features(SURF) algorithm, which is robust in rotation and resizing. A dimension incremented color SURF algorithm based on Gaussian color model is proposed. RGB color model is transformed to Gaussian color model firstly, then the new SURF method is employed for matching experiment, which is improved by additional 48 dimentional colour features vectors. At last, the epiplor constraint is applied to the elimination of false matches. The experimental results show that the proposed algorithm performs favorably on matching points numbers, matching accuracy, and distribution uniformity of matching points, although a slightly more calculation is needed in comparison with the original SURF algorithm. It can meet the matching demand of mobile mapping system (MMS).

Key words:mobile mapping system (MMS); color speeded up robust features (SURF); Gaussian color model; stereo image matching; epipolar constraint