An Image Matching Method of New Illumination-Robust and Anti-Blur Combined Moment Invariants
Liu Hua · Chuangan jishu xuebao · 2013
Novel combined moment invariants are put forward to deal with the problem of the poor adaptability to the illumination variation and the blur noise in the cases of the conventional image feature matching. The new combined moment invariants are re-built on the basis of Hu moments and are applied in image matching. The experimental results have demonstrated that,in virtue of its illumination robustness and blur invariance,the proposed approach can effectively overcome the decline of the correct matching rate influenced by both the illumination change and the fuzzy noise. Ensuring the high precision,the matching efficiency is greatly improved. Utilizing the new combined moment invariants as image descriptors to complete image matching is feasible and validate. It offers a nice reference value.