scispace - formally typeset
Search or ask a question
Author

Dr.K. Umamaheswari

Bio: Dr.K. Umamaheswari is an academic researcher. The author has contributed to research in topics: Fingerprint Verification Competition & Fingerprint recognition. The author has an hindex of 1, co-authored 1 publications receiving 9 citations.

Papers
More filters
Journal ArticleDOI
01 Dec 2011
TL;DR: It is shown that verification imposes additional requirements on multimodal fusion when compared to conventional verification systems, and it is argued that the usual performance metrics of false accept and false reject rates are insufficient yardsticks for continuous verification and proposed new metrics against which to benchmark the system.
Abstract: A multimodal biometrics face and fingerprint recognition system is a computer application for automatically identifying or verifying a person from face by using cameras and fingerprint by using sensors or fingerprint readers or fingerprint scanners. Proposed paper uses Face and Fingerprint recognition technique for verification in ATM systems. There are two types. The first one is verification. Compare the two faces and fingerprint images and decide whether the user (current user image) is an genuine user or imposter. These are decision level. Second one is identification this is where the system compares the given input image to all other images in the database and gives a ranked list of matches. Multimodal biometrics verification system that verifies the presence of a user is genuine or not. Two modalities are currently used—face and fingerprint—but our theory can be readily extended to include more modalities. We show that verification imposes additional requirements on multimodal fusion when compared to conventional verification systems. We also argue that the usual performance metrics of false accept and false reject rates are insufficient yardsticks for continuous verification and propose new metrics against which we benchmark our system.

11 citations


Cited by
More filters
01 Jan 2016
TL;DR: The project enhances the security authentication of customers using ATM by using a core controller using fingerprint recognition system of ATmega128 in-system programmable flash and a new method of authentication is presented.
Abstract: ATM was introduced to boost the cashless policy in Nigeria. Current trend of Cybercrime facilitate the need for an enhanced fingerprint application on ATM machine with GSM Feedback mechanism. The mechanism enable unassigned fingerprint authentication of customers with quick code and secret code. The project enhances the security authentication of customers using ATM. A core controller using fingerprint recognition system of ATmega128 in-system programmable flash is explored. An SM630 fingerprint module is used to capture fingerprints with DSP processor and optical sensor for verification, using AT command of GSM module for feedback text messaging (i.e. sending of Quick and Secret-Codes respectively). Upon system testing of capable reduction of ATM fraud using C program, the new method of authentication is presented.

12 citations

Journal ArticleDOI
TL;DR: The proposed work is to enhance the security in ATM system with multimodal biometrics along with email verification code which provides two level security to the system.
Abstract: Unimodal biometrics uses a single source of biometric system for personal identification. It has a variety of problems such as noise in the sense data, Intra-class variations, Inter-class similarities and spoof attacks. Multibiometrics is a combination of one or more biometrics. In multibiometrics the noise in any one of the biometrics will lead to high false reject rate (FRR) while identification. All these problems are addressed by Multimodal biometrics. Multimodal biometrics is the integration of two or more types of biometrics system (e.g. Fingerprint and Face, Face and Iris, Iris and Fingerprint). It provides a secondary means of identification in case sufficient data is not extracted from a given biometric sample. The main objective is to provide a higher level security to the distributed system and to protect the biometric template by making use of biometric cryptosystem. In the existing approaches multibiometrics which is a combination of one or more biometrics is used to provide security along with feature level fusion to combine the biometric template. The failure of one of the biometrics in multibiometrics system leads to the serious issue. The proposed work is to enhance the security in ATM system with multimodal biometrics along with email verification code which provides two level security to the system.

10 citations

Journal ArticleDOI
TL;DR: A framework comprising of feature recognition using LGL filter and similarity comparison using score level fusion is proposed and a series of stages enhances well the recognition performances using the proposed solution.
Abstract: Nowadays, Biometric system automatically identifies the unique feature of an individual for better evaluation and verification in recognition systems. Face and iris recognition in biometric identification systems is considered as most accurate procedure with higher recognition rate. CCTV surveillance plays a major role in human recognition and identification with the help of intelligent systems. The biometric system combined with CCTV output analyzes the data with/without human intervention. This paper presents an approach of human identification with recognition using facial and iris biometric from lower resolution images. Also, Lower resolution in image clarity is a major constraint in recognizing the individuals from distance with biometric values. To overcome problem in individual recognizing, the combination of Gabor and Legendre filter is combined. The use of hybrid log-Gabor–Legendre (LGL) filter improves the recognition pattern of face and iris in multi-spectral images. After log Gabor–Legendre filtration, two new techniques such as phase quadrant method for iris and LGBPHS method for face are used to improve recognition pattern. Hence, a framework comprising of feature recognition using LGL filter and similarity comparison using score level fusion is proposed. A series of stages enhances well the recognition performances using the proposed solution. Experiments established the validity against existing linear techniques for facial and iris image recognition pattern from CCTV cameras for automated human identification and verification.

9 citations

Proceedings ArticleDOI
01 Feb 2014
TL;DR: Experimental results shows that all the bank accounts can be accessed without using cards through this Iris recognition efficiently, which is provided for ATM banking system.
Abstract: Security and Authentication of individuals is necessary for our daily lives especially in ATMs.But the security provided with ATM systems has some backdoors. It has been improved by using biometric verification techniques like face recognition, fingerprints, voice and other traits, comparing these existing traits, there is still need for considerable computer vision. Iris recognition is a particular type of biometric system that can be used to reliably identify a person uniquely by analyzing the patterns found in the iris. In the proposed system a new approach- IBIO(Iris recognition based BIOmetric verification) is provided for ATM banking system. Initially Iris images are collected as datasets and maintained in agent memory. Then the Iris and pupil are detected from the image, removing noises. The features of the iris were encoded by convolving the normalized iris region with 2D-Gabor filter. The Hamming distance was chosen as a matching metric, which gave the measure of how many bits disagreed between the templates of the iris. Experimental results shows that all the bank accounts can be accessed without using cards through this Iris recognition efficiently.

5 citations

Journal ArticleDOI
TL;DR: This paper proposes a method in which the PIN is replaced by the biometrics of the individual to have a more secure transaction and results are promising when compared with other existing similar techniques.
Abstract: The authentication system used at the automated teller machines ATM is a unique personal identification number PIN. This PIN can be easily tapped and misused. In this paper we propose a method in which the PIN is replaced by the biometrics of the individual to have a more secure transaction. A complete hardware system is designed to capture the biometric traits such as face, fingerprint and palm vein. The captured images are pre-processed and then features are extracted which are then fused at feature level. Cryptography is applied on the fused feature vector. Matching is performed using Euclidean distance at the server end. Palm vein is particularly chosen as a biometric trait along with widely used face and fingerprint because it is unique and is impossible to forge the vein pattern of an individual. Curvelet and Wavelet Transforms are used for the feature extraction. Experimental results indicate a good level of security and recognition rate of 91% and 89% is achieved on our own generated database. The results are promising when compared with other existing similar techniques.

3 citations