摘要：An optimal thenar palmprint classification model is proposed in this paper. Firstly, the thenar palmprint image is enhanced using a high-frequency emphasis filter and histogram equalization. Then, from the enhanced image thirteen textural features of gray level co-occurrence matrix (GLCM) are extracted as classification feature vectors. Finally, the SVM classifier is used for classification and the best classification model will be obtained through comparing the classification results of different kernel functions and feature vectors. The experimental results proved the feasibility and effectiveness of this model for thenar palmprint classification.
