Abstract: The security of templates is the critical part of biometric system and one of the most crucial issues in any proposed system. In fact, features such as voice, face, fingerprints and many others can be covertly acquired or stolen by an attacker and misused. Therefore, storing biometric templates a secure way is crucial. This study proposes a novel approach that combines an improved encryption method with a new concealment technique to establish a secure data template storing system. The Hill Cipher algorithm has been improved to be more secure by using a large and random key with large data block and also extending it to include the special characters. In the other hand a new concealment method proposed by combine two of the popular methods DWT and DCT for embedding and extraction the secret data in order to compensate the drawbacks of both of them and to make the hidden information much more secure against the attacks, wavelet transform which use dyadic filters to decompose cover image into 4-levels (HH, HL, LH and LL) and discrete cosine Transforms to convert a signal of the selected coefficients (HH, HL and LH) into elementary frequency components then according to the percentages entering by a user the encrypted data will distribute. The efficiency of encryption and concealment have been checked out with a number of widely used metrics such as Peak Signal to Noise Ratio (PSNR) and Normalized Correlation Coefficient (NCC).