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

Area of Science:

• Biometric systems
• Mobile security
• Image processing

Background:

• Mobile biometrics are trending, necessitating privacy-conscious solutions.
• Touchless systems address public demand for hygiene and privacy in biometric identification.
• Existing mobile biometric methods require robust algorithms for accurate, touchless recognition.

Purpose of the Study:

• To propose a novel touchless palm print recognition system utilizing smartphone imaging.
• To develop efficient algorithms for hand tracking, image enhancement, and fast computation for mobile biometrics.
• To enhance the accuracy and speed of mobile-based biometric identification.

Main Methods:

• Developed a touchless palm print recognition system using smartphone cameras.
• Implemented a sliding neighborhood operation with local histogram equalization and LHEAT for image enhancement.
• Introduced an improved fuzzy-based k nearest centroid neighbor (IFkNCN) classifier for accelerated recognition.

Main Results:

• Achieved a high recognition accuracy of 98.64% for the touchless palm print system.
• The LHEAT approach effectively enhanced low-quality palm print images.
• The IFkNCN classifier demonstrated faster computation by reducing training data and removing outliers.

Conclusions:

• The proposed touchless palm print recognition system is effective for mobile applications.
• The combination of LHEAT and IFkNCN offers a promising solution for secure and efficient mobile biometrics.
• This research advances the field of mobile biometrics with a focus on privacy and performance.