Researching | Finger Vein Recognition Based on Improved AlexNet

Laser & Optoelectronics Progress, Volume. 57, Issue 8, 081005(2020)

Finger Vein Recognition Based on Improved AlexNet

Zhiyong Tao¹, Yalei Hu^1,2、*, and Sen Lin¹

Author Affiliations

¹School of Electronic & Information Engineering, Liaoning Technical University, Huludao, Liaoning 125105, China;

²Fuxinlixing Technology Company Limited, Fuxin, Liaoning 123000, China

show less

Abstract Get PDF(in Chinese)

Figures&Tables (16)

References (29)

Paper Information

Figures & Tables(16)

Fig. 1. Image collection method. (a) Direct light collection; (b) light reflection collection

Download full size

Fig. 2. Finger vein pretreatment process

Download full size

Fig. 3. Gray scale linear interpolation schematic

Download full size

Fig. 4. Finger vein effect after CLAHE algorithm processing. (a) Original image; (b) image enhancement

Download full size

Fig. 5. AlexNet model structure

Download full size

Fig. 6. Schematic of SPP

Download full size

Fig. 7. Im-AlexNet model structure renderings

Download full size

Fig. 8. Curves of JY_DB. (a) Loss curves; (b) recognition accuracy curves

Download full size

Fig. 9. Curves of SD_DB. (a) Loss curves; (b) recognition accuracy curves

Download full size

Table 1. AlexNet model structure parameters
View table
Table 1. AlexNet model structure parameters
Layer Number of filters Size of kernel Stride
Conv1 96 11×11 4
Pool1 - 3×3 2
Conv2 256 5×5 1
Pool2 - 3×3 2
Conv3 384 3×3 1
Conv4 384 3×3 1
Conv5 256 3×3 1
Pool3 - 3×3 2
FC layer1 4096 1×1 1
FC layer2 4096 1×1 1
FC layer3 1000 1×1 1

Table 2. Im-AlexNet model structure parameters

Table 2. Im-AlexNet model structure parameters

Layer	Number of filters	Size of kernel	Stride
Conv1	96	5×5	4
Conv2	96	1×1	1
Pool1	-	3×3	2
Conv3	256	5×5	1
Conv4	256	1×1	1
Pool2	-	3×3	2
Conv5	384	3×3	1
Conv6	384	1×1	1
Conv7	384	3×3	1
Conv8	384	1×1	1
Conv9	256	3×3	1
Conv10	256	1×1	1
Pool3(SPP)	-	-	-
FC layer1	1024	1×1	1
FC layer2	Class	1×1	1

Table 3. Finger vein datasets division
View table
Table 3. Finger vein datasets division

Table 4. Experimental parameters
View table
Table 4. Experimental parameters
Parameter Value
Batch 128
Epoch 500
Numbers of samples per epoch 1280
LR of SGD 0.001
Momentum of SGD 0.9
Training number 64000
Dropout 0.25

Table 5. Comparison of training time before and after network improvement
View table
Table 5. Comparison of training time before and after network improvement
Network Database Time/min
AlexNet JY_DB 125
SD_DB 84
Im-AlexNet JY_DB 29
SD_DB 25

Table 6. Comparison of recognition accuracy before and after network improvement
View table
Table 6. Comparison of recognition accuracy before and after network improvement
Network Database Accuracy/%
AlexNet JY_DB 95.04
SD_DB 94.36
Im-AlexNet JY_DB 99.25
SD_DB 98.23

Table 7. Comparison of recognition accuracy of different image feature algorithms on SD_DB

Table 7. Comparison of recognition accuracy of different image feature algorithms on SD_DB

Method	Accuracy /%
SPF^[24]	87.00
SPCF^[25]	92.71
CLAHE+directional dilation (DD)^[26]	90.72
CNN (ULDFV-VGG16^[15])	92.60
CNN (ULDFV-Xception^[15])	93.50
CNN (ULDFV-ResNet^[15])	96.60
CNN (Inception-ResNet V2^[15])	96.70
Dula-sliding window+location+Pseudo-elliptical transformer+2D-PCA^[27]	97.02
Block-based average absolute deviation (AAD) features^[28]	97.76
CNN (Proposed CNN^[29])	97.48
CNN (AlexNet)	94.36
CNN (Im-AlexNet)	98.23

Tools

Get Citation

Copy Citation Text

Zhiyong Tao, Yalei Hu, Sen Lin. Finger Vein Recognition Based on Improved AlexNet[J]. Laser & Optoelectronics Progress, 2020, 57(8): 081005

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Image Processing

Received: Jul. 26, 2019

Accepted: Sep. 6, 2019

Published Online: Apr. 3, 2020

The Author Email: Yalei Hu (mrhu165981@163.com)

DOI:10.3788/LOP57.081005

Topics

laser devices and laser physics

Lasers and Laser Optics

laser manufacturing

Instrumentation, Measurement and Metrology