원문정보
초록
영어
To improve unconstrained iris recognition system performance in different environments, a performance improvement method of unconstrained iris recognition based on domain adaptation metric learning is proposed. A kernel matrix is calculated as the solution of domain adaptation metric learning. The known Hamming distance computing by intra-class and inter-class is used as the optimization learning constraints in the process of iris recognition. An optimal Mahalanobis matrix is computed for certain cross-environment system, then distance between two iris samples is redefined. The experimental results indicate that the proposed method can increase the accuracy of the unconstrained iris recognition in different circumstances, improving the classification ability of iris recognition system.
목차
1. Introduction
2. Unconstrained Iris Recognition and Similarity Measure
2.1. The Unconstrained Iris Recognition System
2.2 Similarity Measure for Iris Recognition
3. DAML Method and Novel Resolving Scheme
3.1. Framework of DAML Method
3.2 Constraints for Optimal Matrix Learning
3.3 Kernel Solution of Optimal Mahalanobis Distance Matrix
4. Optimization Algorithm Based on DAML Method
4.1 Optimal Target in Unconstrained Iris Recognition
4.2. Calculating Optimal Matrix for Iris Recognition
4.3. Iris Feature Matching
5. Experimental Details and Results
5.1. Unconstrained Iris Image Dataset
5.2. Segmentation and Feature Extraction
5.3. Experimental Settings
5.4. Experimental results
6. Conclusion
References