Adaptive polar transform and fusion for human face image processing and evaluation

Table 2 Experimental results on IRIS face database

Sl. no.	Method		Sensitivity	Specificity	Accuracy
1	Unregistered + No Fusion + PCA + MLP	Visual images	90.42	81.60	88.73
1	Unregistered + No Fusion + PCA + MLP	Thermal images	87.04	84.62	85.38
2	Unregistered + Pixel Fusion + PCA + MLP		91.93	86.32	90.86
3	Unregistered + Wavelet Fusion + PCA + MLP		94.21	85.30	92.50
4	Log Polar Registered + No fusion + PCA + MLP	Visual images	90.18	80.80	88.39
4	Log Polar Registered + No fusion + PCA + MLP	Thermal images	89.24	78.70	87.22
5	Adaptive Log Polar Registered + No fusion + PCA + MLP	Visual images	91.84	84.70	90.47
5	Adaptive Log Polar Registered + No fusion + PCA + MLP	Thermal images	90.99	82.80	89.42
6	Log Polar Registered + Pixel Fusion + PCA + MLP		92.53	86.70	91.41
7	Adaptive Log Polar Registered + Pixel Fusion + PCA + MLP		94.77	88.10	93.50
8	Log Polar Registered + Wavelet Fusion + PCA + MLP		95.84	92.80	95.26
9	Adaptive Log Polar Registered + Wavelet Fusion + PCA + MLP		98.46	97.90	98.36