The iris recognition using gray level co-occurrence matrix for gabor wavelet transform and principal component analysis through path analysis test case
Iris biometric is considered as one of the most efficient and trusted biometric methods for authenticating users. This thesis proposed an effective iris recognition system based on two different hybrid algorithms. Firstly, both Gray Level Co-occurrence Matrix (GLCM) and (2D) Gabor filters were hybri...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/6152/ http://eprints.uthm.edu.my/6152/1/SENAN_ADEL_MAWLOOD.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Iris biometric is considered as one of the most efficient and trusted biometric methods for authenticating users. This thesis proposed an effective iris recognition system based on two different hybrid algorithms. Firstly, both Gray Level Co-occurrence Matrix (GLCM) and (2D) Gabor filters were hybridized to extract iris features. Secondly, the hybrid of Principal Component Analysis (PCA) and GLCM were used to obtain a secure iris recognition system. The features extracted by the hybrids methods were normalized and Euclidean distance was used to classify the features. Furthermore, the proposed designed system was tested in many cases and their performances were measured and compared in both algorithms. The processes of Path Analysis Test Case of the two algorithms were performed through the use of three classes of the image to evaluate the performance of the designed system and their reliability was measured. The results show that the overall system accuracy obtained by the Gabor and GLCM was 99.17% success rate with 0.82% false accepted rate and 0.83% false rejected rate while, the overall success rate obtained by the PCA algorithm was 99.15% with 0.82% false accepted rate and 0.85% false rejected rate. The reliability of Gabor filter was given as 95% and for PCA, 94%. The Gabor method overcame the PCA in all the tested cases. |
|---|