Bimodal biometric authentication based on index level fusion for combined iris and fingerprint biometrics.
Saved in:
| Title: | Bimodal biometric authentication based on index level fusion for combined iris and fingerprint biometrics. |
|---|---|
| Authors: | Binu, D.1 (AUTHOR) binujethro@gmail.com |
| Source: | Biomedical Signal Processing & Control. May2026, Vol. 117, pN.PAG-N.PAG. 1p. |
| Subject Terms: | Iris recognition, Human fingerprints, Feature extraction, Mathematical optimization, Biometric identification |
| Abstract: | Biometric image retrieval is commonly applied to security in real-world systems in order to avoid fraudulent entry. Current unimodal biometric recognition techniques are generally of no use for correct identity verification. This paper proposes a new index-level fusion paradigm for bimodal biometric verification using iris and fingerprint features combined with Improved Locality Sensitive Hashing (ILSH) for fast and accurate retrieval while ensuring robustness as well as low computational cost. The proposed system mainly consists of a preprocessing, fingerprint feature extraction module, iris feature extraction module, fusion module, and matching module. The preprocessing module consists of fingerprint acquisition, enhancement, and resizing of fingerprint images, followed by discrete wavelet transformation (DWT) performed on the feature extraction module. In the iris module, first, the iris region is segmented from the eye image, and then it is normalized. Then, feature-level fusion is used to combine the extracted biometric features earlier in the matching process. Therefore, the extracted features are normalized based on the min–max normalization method. The normalized feature trajectories are then integrated to acquire the fused feature trajectory and are compared with the query image features using the ILSH indexing technique. A Python simulation tool is used for the simulation process. In the experimental part, the accuracy of 92.2 %, precision of 92 %, F-measure of 92.8 %, computation time of 83 ms, and CIP of 86.4 % are obtained by the proposed framework. [ABSTRACT FROM AUTHOR] |
| Database: | Supplemental Index |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstract: | Biometric image retrieval is commonly applied to security in real-world systems in order to avoid fraudulent entry. Current unimodal biometric recognition techniques are generally of no use for correct identity verification. This paper proposes a new index-level fusion paradigm for bimodal biometric verification using iris and fingerprint features combined with Improved Locality Sensitive Hashing (ILSH) for fast and accurate retrieval while ensuring robustness as well as low computational cost. The proposed system mainly consists of a preprocessing, fingerprint feature extraction module, iris feature extraction module, fusion module, and matching module. The preprocessing module consists of fingerprint acquisition, enhancement, and resizing of fingerprint images, followed by discrete wavelet transformation (DWT) performed on the feature extraction module. In the iris module, first, the iris region is segmented from the eye image, and then it is normalized. Then, feature-level fusion is used to combine the extracted biometric features earlier in the matching process. Therefore, the extracted features are normalized based on the min–max normalization method. The normalized feature trajectories are then integrated to acquire the fused feature trajectory and are compared with the query image features using the ILSH indexing technique. A Python simulation tool is used for the simulation process. In the experimental part, the accuracy of 92.2 %, precision of 92 %, F-measure of 92.8 %, computation time of 83 ms, and CIP of 86.4 % are obtained by the proposed framework. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 17468094 |
| DOI: | 10.1016/j.bspc.2026.109556 |