A Synthetic Fusion Rule Based on FLDA and PCA for Iris Recognition Using 1D Log-Gabor Filter

Table 2

Comparison of results.

Methods	Database	Matching process	Feature extraction	FAR; (%)	FRR; (%)	EER; (%)	Recognition rate (%)
Attarchi et al. [17]	CASIA Database	-	1-D Log-Gabor and 2D-PCA	-	-	0.68	99.32
Wen-Shiung et al. [18]	UBIRIS	Euclidean distance	2D-LDA and 2D-PCA	0.00	0.216	0.74	99.20
Masek [27]	Dataset	Hamming distance with XOR	2D Gabor	0.005	0.238	0.35	99.65
Avila [29]	Database	Hamming distance	Zero-Crossing	0.03	2.08	0.21	99.79
Li Ma et al. [30]	Database	Expanded binary feature vector and exclusive OR operations	Class of 1D Wavelets, i.e.,1-D Intensity Signals	0.02	1.98	0.29	99.71
Tisse [31]	Database	Hamming distance	2D Gabor	1.84	8.78	0.41	99.59
Rai et al. [32]	CASIA Database	SVM and hamming distance with XOR	1-D Log-Gabor wavelets	0.07	0.33	-	-
Soliman et al. [33]	CASIA-V3 CASIA-V1	Hamming distance with XOR	1-D Log-Gabor wavelets	-	-	-	98.80
Dehkordi et al. [34]	CASIA-V3	Adaptive hamming distance with XOR	2-D Log-Gabor wavelets	-	0.06	-	99.96
YONG et al. [35]	IITD Database	Hamming distance	2D Gabor Filter/1D Log-Gabor and LDA	-	-	-	98.92
Proposed	CASIA-V1	Hamming distance with XOR	FLDA/PCA and 1D Log-Gabor filter	0.16	0.00	0.01	99.99