Performance Prediction Methodology for Biometric Systems Using a Large Deviations Approach

Schmid, Natalia A.; O'Sullivan, Joseph A.

doi:10.1109/tsp.2004.833863

Cited by 39 publications

(40 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their work did not support numerical results to illustrate these performance indices or demonstrate other applications using these performance indices. Some insights from [6] support the approach in [7] where authors have shown novel techniques to utilize the information in iris features for enhancing authentication. Figure 1 can be used to summarize the approach employed by Schimd et al [6] to compute the recognition capacity of a biometric system.…”

Section: (A) Prior Work On Capacity Of Biometric Systemmentioning

confidence: 78%

“…Some insights from [6] support the approach in [7] where authors have shown novel techniques to utilize the information in iris features for enhancing authentication. Figure 1 can be used to summarize the approach employed by Schimd et al [6] to compute the recognition capacity of a biometric system. They first formulate maximization of mutual information between a query template from any user i given by i X and user templates j Y in the database.…”

Section: (A) Prior Work On Capacity Of Biometric Systemmentioning

confidence: 78%

“…Schmid et al [6] outline a seminal approach in formulating performance indices like recognition capacity and reliability function of biometric system, refer Figure 1. Their work did not support numerical results to illustrate these performance indices or demonstrate other applications using these performance indices.…”

Section: (A) Prior Work On Capacity Of Biometric Systemmentioning

confidence: 99%

See 2 more Smart Citations

On estimating performance indices for biometric identification

Bhatnagar

Kumar

2009

Pattern Recognition

View full text Add to dashboard Cite

show abstract

Section: (A) Prior Work On Capacity Of Biometric Systemmentioning

confidence: 78%

Section: (A) Prior Work On Capacity Of Biometric Systemmentioning

confidence: 78%

Section: (A) Prior Work On Capacity Of Biometric Systemmentioning

confidence: 99%

See 1 more Smart Citation

On estimating performance indices for biometric identification

Bhatnagar

Kumar

2009

Pattern Recognition

View full text Add to dashboard Cite

show abstract

“…These can be considered as High, Medium and Low in [25]. Any physiological or behavioral attribute of human beings can provide as a biometric characteristic on which it satisfies the said requirements [26]. Table 1 compares the biometric features based on different aspects.…”

Section: Allied Workmentioning

confidence: 99%

Establishing User Authentication using Face Geometry

Banerjee¹,

Chatterjee²,

Bhattacharyya³

et al. 2014

IJCA

View full text Add to dashboard Cite

The use of digital media or information is very precarious for increasing of hacker in the nation. So the information security is one of the commanding articles to shield it. The biometric information security is one information security mechanism, which is powerful than conventional cryptography system. The biometric system plays a vital role in person recognition. The main reason of biometry is so popular in security, because there is no risk if something might be lost or stolen in case of traditional IDs and passwords. After several comparisons among possible features of a human face geometry processing approaches, an authorized person recognition system have been designed and developed. Freely accessible sample faces of different persons are used in this novel biometric authentication system. Furthermore, the functionality to extract features of face has been implemented to compare the new samples with user templates. The implementation has been evaluated by the FAR (False Acceptance Rate) and FRR (False Rejection Rate) of the system in order to reduce FAR. Examining the Distance between some objects of faces and angles between objects points are used in this system and this is the novelty.

show abstract

“…Much work has been done already on performance modeling and prediction of biometric systems such as fingerprint recognition [11], iris recognition [10], [9], and face recognition [8], [7], [3]. Specifically, in the work by Tabassi et al [11], the quality of a fingerprint image is defined as the normalized distance between matching and nonmatching similarity scores.…”

Section: Related Workmentioning

confidence: 99%

Modeling and Predicting Face Recognition System Performance Based on Analysis of Similarity Scores

Wang

Wayman

2007

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Abstract-This paper presents methods of modeling and predicting face recognition (FR) system performance based on analysis of similarity scores. We define the performance of an FR system as its recognition accuracy, and consider the intrinsic and extrinsic factors affecting its performance. The intrinsic factors of an FR system include the gallery images, the FR algorithm, and the tuning parameters. The extrinsic factors include mainly query image conditions. For performance modeling, we propose the concept of "perfect recognition," based on which a performance metric is extracted from perfect recognition similarity scores (PRSS) to relate the performance of an FR system to its intrinsic factors. The PRSS performance metric allows tuning FR algorithm parameters offline for near optimal performance. In addition, the performance metric extracted from query images is used to adjust face alignment parameters online for improved performance. For online prediction of the performance of an FR system on query images, features are extracted from the actual recognition similarity scores and their corresponding PRSS. Using such features, we can predict online if an individual query image can be correctly matched by the FR system, based on which we can reduce the incorrect match rates. Experimental results demonstrate that the performance of an FR system can be significantly improved using the presented methods.

show abstract

Performance Prediction Methodology for Biometric Systems Using a Large Deviations Approach

Cited by 39 publications

References 25 publications

On estimating performance indices for biometric identification

On estimating performance indices for biometric identification

Establishing User Authentication using Face Geometry

Modeling and Predicting Face Recognition System Performance Based on Analysis of Similarity Scores

Contact Info

Product

Resources

About