Abstract: A hand vein pattern recognition algorithm based on moment invariants and Support Vector Machine（SVM） is presented in the paper.After the vein pattern was segmented from the vein image by an improved NiBlack method，authors obtained the vein pattern skeleton by an improved conditional thinning algorithm.On the basis，the algorithm extracted seven modified moment invariants of the vein pattern skeleton as the input vector of SVM for the recognition of vein pattern.The experiments on a database including 500 samples has achieved a recognition rate of 95.5%，and it indicates the algorithm is effective.