A Secure Biometric Key Generation Mechanism via Deep Learning and Its Application

Abstract

Biometric keys are widely used in the digital identity system due to the inherent uniqueness of biometrics. However, existing biometric key generation methods may expose biometric data, which will cause users’ biometric traits to be permanently unavailable in the secure authentication system. To enhance its security and privacy, we propose a secure biometric key generation method based on deep learning in this paper. Firstly, to prevent the information leakage of biometric data, we utilize random binary codes to represent biometric data and adopt a deep learning model to establish the relationship between biometric data and random binary code for each user. Secondly, to protect the privacy and guarantee the revocability of the biometric key, we add a random permutation operation to shuffle the elements of binary code and update a new biometric key. Thirdly, to further enhance the reliability and security of the biometric key, we construct a fuzzy commitment module to generate the helper data without revealing any biometric information during enrollment. Three benchmark datasets including ORL, Extended YaleB, and CMU-PIE are used for evaluation. The experiment results show our scheme achieves a genuine accept rate (GAR) higher than the state-of-the-art methods at a 1% false accept rate (FAR), and meanwhile satisfies the properties of revocability and randomness of biometric keys. The security analyses show that our model can effectively resist information leakage, cross-matching, and other attacks. Moreover, the proposed model is applied to a data encryption scenario in our local computer, which takes less than 0.5 s to complete the whole encryption and decryption at different key lengths.

Fingerprint bio‐key generation based on a deep neural network

“…It then directly converts it to binary, and subsequently performs RS encoding to test the extraction effect. In the paper, 23 the idea of end‐to‐end direct training generation of biological key is presented and the SENet‐DNN method is proposed. In the experiment, the SENet‐DNN method adopts the five‐layer SENet+3 FC proposed by Wang et al 23 as the biometric key generation model, and adopts the training method of binary classification.…”

“…In the paper, 23 the idea of end‐to‐end direct training generation of biological key is presented and the SENet‐DNN method is proposed. In the experiment, the SENet‐DNN method adopts the five‐layer SENet+3 FC proposed by Wang et al 23 as the biometric key generation model, and adopts the training method of binary classification. Under normal circumstances, identity confirmation is carried out by binary classification of positive and negative samples, and binary classification is easier to converge.…”

“…Each of the 3 FC layers is equipped with a dropout layer with a parameter of 0.35. The optimizer was SGD, the learning rate was 0.001, batch_size was 16, the loss function was the sum of the three loss functions proposed in the paper, 23 and the error correction number was set as 512 bits and error correction 16 bits. The experimental results are shown in Tables 2 and 3.…”

“…In general, the success rate and strength of the biological key generation are still not sufficiently high. Wang et al 23 proposed a deep neural network bio‐key generation model based on the SENet structure, which mainly generates bio‐keys for human faces. However, due to the complexity and instability of biometrics, and the low interpretability and controllability of the end‐to‐end generation model, the test effect of this model in the actual biometric database is not ideal.…”

“…Tables 2 and 3 and Figures [9][10][11][12] show the performance comparison between MCP-FPmodel algorithm and SENet-DNN algorithm. 23 Each algorithm was tested for 20 rounds to eliminate contingency. The accuracy rate and Loss changes were compared.…”