Limits...
Intensity Variation Normalization for Finger Vein Recognition Using Guided Filter Based Singe Scale Retinex.

Xie SJ, Lu Y, Yoon S, Yang J, Park DS - Sensors (Basel) (2015)

Bottom Line: However, the capacities and percentages of finger tissues (e.g., bone, muscle, ligament, water, fat, etc.) vary person by person.This usually causes poor quality of finger vein images, therefore degrading the performance of finger vein recognition systems (FVRSs).The experimental results on two public datasets demonstrate the effectiveness of the proposed method in enhancing the image quality and finger vein recognition accuracy.

View Article: PubMed Central - PubMed

Affiliation: Institute of Remote Sensing and Earth Science, College of Science, Hangzhou Normal University, Hangzhou 311121, China. shanj_x@hotmail.com.

ABSTRACT
Finger vein recognition has been considered one of the most promising biometrics for personal authentication. However, the capacities and percentages of finger tissues (e.g., bone, muscle, ligament, water, fat, etc.) vary person by person. This usually causes poor quality of finger vein images, therefore degrading the performance of finger vein recognition systems (FVRSs). In this paper, the intrinsic factors of finger tissue causing poor quality of finger vein images are analyzed, and an intensity variation (IV) normalization method using guided filter based single scale retinex (GFSSR) is proposed for finger vein image enhancement. The experimental results on two public datasets demonstrate the effectiveness of the proposed method in enhancing the image quality and finger vein recognition accuracy.

Show MeSH
Comparison of finger vein image enhancement using different methods on MMCBNU_6000: (a) original ROI images, and enhanced images using (b) IN method; (c) SR method; (d) SSR method; (e) HE method; and (f) the proposed GFSSR method.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541924&req=5

sensors-15-17089-f010: Comparison of finger vein image enhancement using different methods on MMCBNU_6000: (a) original ROI images, and enhanced images using (b) IN method; (c) SR method; (d) SSR method; (e) HE method; and (f) the proposed GFSSR method.

Mentions: Some ROIs of the samples from MMCBNU_6000 and their enhancements using different methods are displayed in Figure 10. It can be seen that HE can enhance the brightness level. However, as shown in Figure 10e, HE caused level saturation effects in some small regions. This leads to the presence of some dark regions, much darker than those in the input images. The same circumstance appears in the enhanced images using SSR, as shown in Figure 10d. Although the enhanced images using SSR have brighter distribution than the input images, the contrast between venous and non-venous does not show much of an increase. It can be seen in Figure 10b that the images enhanced using IN [12] has good image contrast and clear edges between the venous and non-venous regions. Unfortunately, some of the vein patterns are lost in the local regions. Moreover, the thickness of the veins in Figure 10b increases compared to those in the input images. For images enhanced using SR [13], low image contrast regions in ROI images is still poor in the corresponding regions in Figure 10c. The proposed GFSSR concentrates on investigating the effects of finger tissue on finger vein imaging. Borrowing the adjustable edge-preserving ability of guided filter, the images enhanced using the proposed GFSSR has better image contrast not only in the global image, but also in local regions. In addition, as shown in Figure 10f, the enhanced images have much clearer edges, especially in the vague local regions. Furthermore, the thickness of the veins in Figure 10f remains the same as with those in the input images.


Intensity Variation Normalization for Finger Vein Recognition Using Guided Filter Based Singe Scale Retinex.

Xie SJ, Lu Y, Yoon S, Yang J, Park DS - Sensors (Basel) (2015)

Comparison of finger vein image enhancement using different methods on MMCBNU_6000: (a) original ROI images, and enhanced images using (b) IN method; (c) SR method; (d) SSR method; (e) HE method; and (f) the proposed GFSSR method.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541924&req=5

sensors-15-17089-f010: Comparison of finger vein image enhancement using different methods on MMCBNU_6000: (a) original ROI images, and enhanced images using (b) IN method; (c) SR method; (d) SSR method; (e) HE method; and (f) the proposed GFSSR method.
Mentions: Some ROIs of the samples from MMCBNU_6000 and their enhancements using different methods are displayed in Figure 10. It can be seen that HE can enhance the brightness level. However, as shown in Figure 10e, HE caused level saturation effects in some small regions. This leads to the presence of some dark regions, much darker than those in the input images. The same circumstance appears in the enhanced images using SSR, as shown in Figure 10d. Although the enhanced images using SSR have brighter distribution than the input images, the contrast between venous and non-venous does not show much of an increase. It can be seen in Figure 10b that the images enhanced using IN [12] has good image contrast and clear edges between the venous and non-venous regions. Unfortunately, some of the vein patterns are lost in the local regions. Moreover, the thickness of the veins in Figure 10b increases compared to those in the input images. For images enhanced using SR [13], low image contrast regions in ROI images is still poor in the corresponding regions in Figure 10c. The proposed GFSSR concentrates on investigating the effects of finger tissue on finger vein imaging. Borrowing the adjustable edge-preserving ability of guided filter, the images enhanced using the proposed GFSSR has better image contrast not only in the global image, but also in local regions. In addition, as shown in Figure 10f, the enhanced images have much clearer edges, especially in the vague local regions. Furthermore, the thickness of the veins in Figure 10f remains the same as with those in the input images.

Bottom Line: However, the capacities and percentages of finger tissues (e.g., bone, muscle, ligament, water, fat, etc.) vary person by person.This usually causes poor quality of finger vein images, therefore degrading the performance of finger vein recognition systems (FVRSs).The experimental results on two public datasets demonstrate the effectiveness of the proposed method in enhancing the image quality and finger vein recognition accuracy.

View Article: PubMed Central - PubMed

Affiliation: Institute of Remote Sensing and Earth Science, College of Science, Hangzhou Normal University, Hangzhou 311121, China. shanj_x@hotmail.com.

ABSTRACT
Finger vein recognition has been considered one of the most promising biometrics for personal authentication. However, the capacities and percentages of finger tissues (e.g., bone, muscle, ligament, water, fat, etc.) vary person by person. This usually causes poor quality of finger vein images, therefore degrading the performance of finger vein recognition systems (FVRSs). In this paper, the intrinsic factors of finger tissue causing poor quality of finger vein images are analyzed, and an intensity variation (IV) normalization method using guided filter based single scale retinex (GFSSR) is proposed for finger vein image enhancement. The experimental results on two public datasets demonstrate the effectiveness of the proposed method in enhancing the image quality and finger vein recognition accuracy.

Show MeSH