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
Examples of parameters (a,b) of the guided filter according to different parameters (ε,r).
© Copyright Policy
Related In: Results  -  Collection

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

sensors-15-17089-f004: Examples of parameters (a,b) of the guided filter according to different parameters (ε,r).

Mentions: A result of guided filter at a point , is obtained by a weighted sum of an intensity value at in a given guidance image, s, and an average of a patch centered at in a given input image, . a and b are used as its weights, which are determined by whether a patch centered at a point in the guidance image has relatively high variance, as defined in Equations (3) and (4). If a patch has relatively high variance, a becomes relatively large and contributes to more than . Otherwise, contributes to more than . As shown in Figure 4, r is a radius of a patch and becomes larger. A region with relatively high variance also becomes larger and their variance values become smaller, so the contribution of to at a position becomes smaller than in a case of using a smaller r. While its smoothing effect becomes stronger, its edge-preserving effect becomes weaker, but its tendency still remains. Therefore, the guided filter can work pursuing for a given purpose, when its parameter values and a guidance image are chosen properly.


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)

Examples of parameters (a,b) of the guided filter according to different parameters (ε,r).
© Copyright Policy
Related In: Results  -  Collection

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

sensors-15-17089-f004: Examples of parameters (a,b) of the guided filter according to different parameters (ε,r).
Mentions: A result of guided filter at a point , is obtained by a weighted sum of an intensity value at in a given guidance image, s, and an average of a patch centered at in a given input image, . a and b are used as its weights, which are determined by whether a patch centered at a point in the guidance image has relatively high variance, as defined in Equations (3) and (4). If a patch has relatively high variance, a becomes relatively large and contributes to more than . Otherwise, contributes to more than . As shown in Figure 4, r is a radius of a patch and becomes larger. A region with relatively high variance also becomes larger and their variance values become smaller, so the contribution of to at a position becomes smaller than in a case of using a smaller r. While its smoothing effect becomes stronger, its edge-preserving effect becomes weaker, but its tendency still remains. Therefore, the guided filter can work pursuing for a given purpose, when its parameter values and a guidance image are chosen properly.

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