Fingerprint Classification Method Based on Least Square Support Vector Machine and Detailed Image

A new online fingerprint identification method is proposed based on detailed image feature and Least-Square Support Vector Machines which has the fast calculate and better generalization capability. This algorithm uses binary tree theory to decompose the problem into three 2-class classification problem, utilizing improved indexing table thinning feature extraction algorithm, then using the support vector machine to optimize the three hyper-planes. Experimental results show that this algorithm improves the efficiency of fingerprint classification.