High Precision Recognition Method of Small Area Fingerprints Based on Machine Vision

Abstract

The realm of fingerprint recognition often encounters challenges in the precise identification of feature points within small regions due to limited information availability. This can hinder recognition accuracy, particularly in forensic investigations. To address this, this paper introduces a novel machine vision technique tailored for small-area fingerprint recognition. The proposed method meticulously extracts intricate features and enhances image clarity, resulting in improved recognition precision. Furthermore, it explores promising applications in biometric security systems and access control.

Introduction

Fingerprints, with their unique and enduring characteristics, have long been employed for personal identification. Traditional fingerprint recognition methodologies rely on the identification of specific features, known as minutiae points, within fingerprint images. However, these methods often falter when confronted with small-area fingerprints, frequently encountered in restricted regions of larger fingerprint images or on objects with limited surface area. The scarcity of information in such fingerprints poses a challenge in accurately identifying feature points, leading to compromised recognition accuracy.

Proposed Method

To overcome the limitations of conventional techniques in small-area fingerprint recognition, this paper presents an innovative approach grounded in machine vision. At the heart of the proposed method lies a descriptor that meticulously analyzes estimated values of critical fingerprint parameters. Utilizing this descriptor, the method extracts detailed feature points and constructs a frequency field. This field serves as a pivotal tool in enhancing the clarity of the small-area fingerprint image, thereby improving its quality and facilitating the extraction of intricate features.

Experimental Results

To rigorously evaluate the efficacy of the proposed method, extensive experiments were conducted. The results resoundingly demonstrated the method’s prowess in accurately extracting detailed features from seemingly obscure fingerprint images. Remarkably, the average recognition time was significantly reduced compared to conventional approaches, boasting a reduction of over 50%. Additionally, the technique exhibited a more uniform distribution of feature points, excelling in the identification of ridge features on image edges.

Potential Applications

The proposed machine vision technology’s implications extend beyond forensic science. Its enhancements in efficiency and accuracy hold the potential to bolster the reliability of biometric authentication systems employed in security systems and access control. The method’s ability to extract detailed features from small-area fingerprints can elevate the performance of biometric systems, particularly in scenarios where only partial fingerprints are available.

Conclusion

The proposed high precision recognition method for small area fingerprints, leveraging the power of machine vision, represents a significant advancement in fingerprint recognition technology. It effectively addresses the limitations of conventional techniques in handling small-area fingerprints, culminating in improved recognition accuracy and reduced processing time. The method also ventures into potential applications in biometric security systems and access control, underscoring its versatility and practical significance.