Recently, biometric template protection has received ever increasing research interests. However, most of the proposed schemes are not yet sufficiently mature for large scale deployment, they do not meet the requirements of security, and high-recognition performance. For example, matching methods based on Euclidean measurement ignore the difference of different component easily. Mean quantization method may not be provide higher security. For the longer iris, fuzzy vault scheme will take less security when used directly. In this paper, a novel authentication schemes for biometric template protection is proposed. First, we introduce a sequence matching threshold method to enhance the recognition performance. Then, we propose a binary transform method using gray code for quantization to suit the sequence matching threshold. Finally, we propose Multi-Division Template Protection (MDTP) framework based on fuzzy vault with higher performance at different equation recovery situation. The experimental results show the superior performance of the proposed schemes than other methods.