CRIMINAL IDENTIFICATION SYSTEM USING FACE RECOGNITION AND FINGERPRINT IDENTIFICATION

CRIMINAL IDENTIFICATION SYSTEM USING FACE RECOGNITION AND FINGERPRINT IDENTIFICATION

ABSTRACT

Generic face recognition systems identify a person by comparing a digital image of that person with images in an existing face database. These systems facilitate criminal identification but are constrained by the use of a single biometric identification technique. This paper presents a web-based multi modal biometrics system with a centralized database that uses face recognition and fingerprint identification to identify a criminal with higher precision. The system makes use of the Principal Component Analysis based face recognition technique using the distance between nodal points to match face images. Existing methodologies were researched on various image processing techniques and face recognition techniques to come up with a more efficient flexible system that can be used in a practical environment.

Keywords - Biometrics, Face Recognition, Fingerprint Identification, Principal Component Analysis (PCA)

CHAPTER ONE

INTRODUCTION

This study presents a system that combines face recognition and fingerprint identification, two of the most prominent biometric identification techniques, to identify criminals. In today’s world the rapidly increasing rate of criminal activities causes a major threat to the security of a country and its civilians. To identify criminals and record their details, law enforcement agencies should have the necessary technology when a suspect is apprehended. As human beings, we have the inborn ability to recognize and distinguish between faces. However, this kind of intelligence is not available yet with computers. In order for software to emulate this kind of behavior it needs training. Researchers and software developers have come up with algorithms and mathematical modules and these have been implemented in various applications [1][2].

A. Face recognition

Face recognition scheme of human identification is probably the most user friendly and non-intrusive authentication method available and is one of the most acceptable biometric techniques utilized in various real-world applications [1]. The developing of face recognition system is quite difficult because the human face is quite complex, multidimensional and corresponding on environment changes [2]. Part of our research is focused on developing the computational model of face recognition that is simple, fast and accurate when implemented in different environments. Some of the main principals used in the area of face recognition are;

Eigenfaces / Principal Component Analysis (PCA) - The Principal Component Analysis is one of the most successful techniques that have been used to recognize faces in images. However, high computational cost and dimensionality is a major problem of this technique [3].

2. Feature Based Face Recognition - Face recognition can be done using features as well, but the problem of feature based face recognition in the setting where only a single example of each face is available for training. Hence mixture-distance based approaches can be used to improve feature based face recognition [4][5]. 3. Neural Network Based Face Recognition - Another technique is using neural networks, where unique identity used as inputs to the neural network to measure similarity in classification and recognition. In general the Back propagation neural network is used for the classification and recognition purposes [2].The main limitation is that, all face images should be retrained, whenever a new face is added to the database. Hence it is not practical for a criminal identification system. To overcome the limitations of above methods and provide more practical and feasible system with higher accuracy, we have used a combination of PCA and feature based approaches in the face recognition module for our system, which includes the use of 6 distances between nodal points of the human face to obtain a feature vector to represent each image, instead of pixel values of the image.

B. Fingerprint identification

Fingerprint based personal identification is used by law enforcement agencies in identifying a person suspected of committing a crime and it is routinely used in forensic laboratories and identification units around the world as it has been accepted in the court of law for nearly a century [6]. This technique relies on matching certain patterns in the fingertip which consist a set of ridges and minutiae points that is unique to each person making it an extremely accurate method of identification. The fingerprint module in our system is implemented by developing a complete algorithm which will make use of these unique fingerprint patterns. This module is integrated with the enhanced facial recognition module to produce a more accurate and reliable criminal identification system.

TOPIC: CRIMINAL IDENTIFICATION SYSTEM USING FACE RECOGNITION AND FINGERPRINT IDENTIFICATION

Format: MS Word

Chapters: 1 - 5
Delivery: Email

Number of Pages: 65

Price: 3000 NGN

In Stock

Buy Complete Material Now