Limits...
A Robust and Fast Computation Touchless Palm Print Recognition System Using LHEAT and the IFkNCN Classifier.

Jaafar H, Ibrahim S, Ramli DA - Comput Intell Neurosci (2015)

Bottom Line: To accelerate the recognition process, a new classifier, improved fuzzy-based k nearest centroid neighbor (IFkNCN), was implemented.By removing outliers and reducing the amount of training data, this classifier exhibited faster computation.Our experimental results demonstrate that a touchless palm print system using LHEAT and IFkNCN achieves a promising recognition rate of 98.64%.

View Article: PubMed Central - PubMed

Affiliation: Intelligent Biometric Group, School of Electrical and Electronic Engineering, Universiti Sains Malaysia Engineering Campus, 14300 Nibong Tebal, Penang, Malaysia.

ABSTRACT
Mobile implementation is a current trend in biometric design. This paper proposes a new approach to palm print recognition, in which smart phones are used to capture palm print images at a distance. A touchless system was developed because of public demand for privacy and sanitation. Robust hand tracking, image enhancement, and fast computation processing algorithms are required for effective touchless and mobile-based recognition. In this project, hand tracking and the region of interest (ROI) extraction method were discussed. A sliding neighborhood operation with local histogram equalization, followed by a local adaptive thresholding or LHEAT approach, was proposed in the image enhancement stage to manage low-quality palm print images. To accelerate the recognition process, a new classifier, improved fuzzy-based k nearest centroid neighbor (IFkNCN), was implemented. By removing outliers and reducing the amount of training data, this classifier exhibited faster computation. Our experimental results demonstrate that a touchless palm print system using LHEAT and IFkNCN achieves a promising recognition rate of 98.64%.

No MeSH data available.


Input membership function.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4465684&req=5

fig13: Input membership function.

Mentions: The distances were sorted in ascending order to determine the minimum and maximum distance. The threshold was set such that the training samples fell within a selected threshold distance and were considered inliers. Otherwise, they were considered to be outliers. To determine the threshold, triangle inequality was applied. The triangle inequality method requires that the distance between two objects (reference point and training samples; reference point and query point) cannot be less than the difference between the distances to any other object (query point and the training samples) [35]. More specifically, the distance between the query point and training samples satisfies the triangle inequality condition as follows:(17)dy,xj≤dxj,z+dy,z,where d(y, z) is the distance from the query point to reference sample. In this study, the maximum distance obtained from (16) was assumed to be d(y, z). For faster computation, the distance between training sample and reference sample d(xj, z) was discarded. To eliminate the computation of d(xj, z), (17) was rewritten as follows:(18)2dy,xj≤dxj,z+dy,z.Because d(y, xj) ≤ d(xj, z), the value of d(xj, z) is not necessary, and (18) can be rearranged as follows:(19)dy,xj≤12dy,z.The choice of threshold values is important because a large threshold value requires more computation. A small threshold makes the triangle inequality computation useless. To tackle the problem, the candidate outlier detection can be expressed by the fuzzy IF-THEN rules. Each input set was modeled by two functions, as depicted in Figure 13.


A Robust and Fast Computation Touchless Palm Print Recognition System Using LHEAT and the IFkNCN Classifier.

Jaafar H, Ibrahim S, Ramli DA - Comput Intell Neurosci (2015)

Input membership function.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4465684&req=5

fig13: Input membership function.
Mentions: The distances were sorted in ascending order to determine the minimum and maximum distance. The threshold was set such that the training samples fell within a selected threshold distance and were considered inliers. Otherwise, they were considered to be outliers. To determine the threshold, triangle inequality was applied. The triangle inequality method requires that the distance between two objects (reference point and training samples; reference point and query point) cannot be less than the difference between the distances to any other object (query point and the training samples) [35]. More specifically, the distance between the query point and training samples satisfies the triangle inequality condition as follows:(17)dy,xj≤dxj,z+dy,z,where d(y, z) is the distance from the query point to reference sample. In this study, the maximum distance obtained from (16) was assumed to be d(y, z). For faster computation, the distance between training sample and reference sample d(xj, z) was discarded. To eliminate the computation of d(xj, z), (17) was rewritten as follows:(18)2dy,xj≤dxj,z+dy,z.Because d(y, xj) ≤ d(xj, z), the value of d(xj, z) is not necessary, and (18) can be rearranged as follows:(19)dy,xj≤12dy,z.The choice of threshold values is important because a large threshold value requires more computation. A small threshold makes the triangle inequality computation useless. To tackle the problem, the candidate outlier detection can be expressed by the fuzzy IF-THEN rules. Each input set was modeled by two functions, as depicted in Figure 13.

Bottom Line: To accelerate the recognition process, a new classifier, improved fuzzy-based k nearest centroid neighbor (IFkNCN), was implemented.By removing outliers and reducing the amount of training data, this classifier exhibited faster computation.Our experimental results demonstrate that a touchless palm print system using LHEAT and IFkNCN achieves a promising recognition rate of 98.64%.

View Article: PubMed Central - PubMed

Affiliation: Intelligent Biometric Group, School of Electrical and Electronic Engineering, Universiti Sains Malaysia Engineering Campus, 14300 Nibong Tebal, Penang, Malaysia.

ABSTRACT
Mobile implementation is a current trend in biometric design. This paper proposes a new approach to palm print recognition, in which smart phones are used to capture palm print images at a distance. A touchless system was developed because of public demand for privacy and sanitation. Robust hand tracking, image enhancement, and fast computation processing algorithms are required for effective touchless and mobile-based recognition. In this project, hand tracking and the region of interest (ROI) extraction method were discussed. A sliding neighborhood operation with local histogram equalization, followed by a local adaptive thresholding or LHEAT approach, was proposed in the image enhancement stage to manage low-quality palm print images. To accelerate the recognition process, a new classifier, improved fuzzy-based k nearest centroid neighbor (IFkNCN), was implemented. By removing outliers and reducing the amount of training data, this classifier exhibited faster computation. Our experimental results demonstrate that a touchless palm print system using LHEAT and IFkNCN achieves a promising recognition rate of 98.64%.

No MeSH data available.