High-speed face recognition using self-adaptive radial basis function neural networks

Sing, Jamuna Kanta; Thakur, S.; Basu, Dipak Kumar; Nasipuri, Mita; Kundu, Mahantapas

doi:10.1007/s00521-009-0242-6

Cited by 15 publications

(8 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thereafter, a test image is classified by the multilayer RBFNN [16,17] using its WFG-2DFLD -based features. The input layer of the above RBFNN comprises of D units (neurons) for a D-dimensional feature (input) vector.…”

Section: Radial Basis Function Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Weighted Fuzzy Generalized 2DFLD: A Fuzzy-Based Feature Extraction Technique for Face Recognition

Dey¹,

Sing²,

Chowdhury³

2017

IJMLC

Self Cite

View full text Add to dashboard Cite

Abstract-This paper proposes a weighted scheme and fuzzy logic-based feature extraction technique, called weighted fuzzy generalized two-dimensional Fisher's linear discriminant (WFG-2DFLD) and its use for face recognition using radial basis function (RBF) neural network as a classifier. In particular, the WFG-2DFLD method is extended version of the generalized two-dimensional Fisher's linear discriminant (G-2DFLD) method. Like G-2DFLD, WFG-2DFLD also maximizes class separability along row and column directions simultaneously. Firstly, it calculates fuzzy membership matrix by fuzzy k-nearest neighbour (Fk-NN) algorithm for the training samples. Secondly, the fuzzy membership values are combined with the training samples to obtained global mean and class-wise mean training images. Thereafter, the global and class-wise mean images are used to generate fuzzy within-class and fuzzy between-class scatter matrices along the row and column directions. In order to make more accurate for classification, different weights are incorporated to scatter matrices. Finally, by solving the Eigen value problems of these scatter matrices; we find the optimal fuzzy projection vectors, which actually used to generate more discriminant features for face recognition. The WFG-2DFLD method has been evaluated on the YALE, AT&T (formally known as ORL), UMIST and FERET face databases using RBF neural network. Simulation results demonstrate that the proposed WFG-2DFLD method can obtain higher recognition rates than some state-of-the-art face recognition methods.Index Terms-WFG-2DFLD, fuzzy projection vector, Fk-NN, RBFNN based classifier, matrix-based feature extraction. I. INTRODUCTIONIn the last few decades the pattern recognition, biometrics and feature extraction became very popular research area among the researchers [1]- [4]. In this context, the face recognition problem can be viewed as a classification problem where one or more test images are compared with the face images stored in the databases. Although the face recognition techniques from still images having controlled background have matured substantially, the face recognition task is still very challenging under uncontrolled environment and different issues related to human face i.e. facial expression (unhappiness, happiness) and facial pose, occlusion (wearing beards, mustaches and glasses), illumination variance, etc. Broadly, the face recognition Manuscript received August 21, 2017; revised November 10, 2017. Aniruddha Dey and Jamuna Kanta Sing are with the Department of Computer Science Engineering, Jadavpur University, Kolkata, jksing@ieee.org).Shiladitya Chowdhury is with the Department of Computer Application, Techno India, Kolkata, India (e-mail: dityashila@yahoo.com).methods can be classified into three main categories: (a) holistic approach, (b) feature based approach and (c) hybrid approach [1]- [9].Holistic approach: This approach uses all the information available in the face images as a whole. The subspace-based methods, like principal component analysis (PCA), l...

show abstract

Section: Radial Basis Function Neural Networkmentioning

confidence: 99%

“…. , ) between the hidden layer and output layer are estimated using the LMS algorithm [16,17]. The class of a test feature vector (and thus test image) is estimated as the index of the output layer unit, which produces maximum value and is defined as follows:…”

mentioning

confidence: 99%

Weighted Fuzzy Generalized 2DFLD: A Fuzzy-Based Feature Extraction Technique for Face Recognition

Dey¹,

Sing²,

Chowdhury³

2017

IJMLC

Self Cite

View full text Add to dashboard Cite

show abstract

“…Till now, most of the efforts have been made to address the first issue by proposing various feature extraction methods [3]- [7]. Very few efforts have been made to address both the above issues [8]- [10]. The present study addresses both the above issues to improve the performance of a face recognition system.…”

Section: Introductionmentioning

confidence: 98%

Face Recognition Using Posterior Distance Model Based Radial Basis Function Neural Networks

Thakur

Sing

Basu

et al. 2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The success rate of a face recognition system heavily depends on two issues, mainly, i) feature extraction method and ii) choosing/designing of a classifier to classify a new face image based on the extracted features. In this paper, we have addressed both the above issues by proposing a new feature extraction technique and a posterior distance model based radial basis function neural networks (RBFNN). First, the dimension of the face images is reduced by a new direct kernel principal component analysis (DKPCA) method. Then, the resulting face vectors are further reduced by the Fisher's discriminant analysis (FDA) technique to acquire lower dimensional discriminant features. During classification, when the RBFNN is not so confident to classify a test image, we have introduced a statistical method called the posterior distance model (PDM) to resolve the conflict. The PDM is an approach, which takes a decision by integrating the outputs of the RBFNN and a distance measure. We call the new classifier the posterior distance model based radial basis function neural networks (PDM-RBFNN). The proposed method has been evaluated on the AT&T database. The simulation results in terms of recognition rates are found to better than some of the existing related approaches.

show abstract

“…The center updating in [9] is based on an incremental scheme. In [10], an incremental technique is used for the updating of connecting weights in the output layer. These fast algorithms are implemented by software.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, only moderate acceleration can be achieved. Moreover, for the incremental algorithms [9,10], inappropriate selection of learning rate may severely degrade the training performance.…”

Section: Introductionmentioning

confidence: 99%

Efficient VLSI Architecture for Training Radial Basis Function Networks

Fan

Hwang

2013

Sensors

View full text Add to dashboard Cite

This paper presents a novel VLSI architecture for the training of radial basis function (RBF) networks. The architecture contains the circuits for fuzzy C-means (FCM) and the recursive Least Mean Square (LMS) operations. The FCM circuit is designed for the training of centers in the hidden layer of the RBF network. The recursive LMS circuit is adopted for the training of connecting weights in the output layer. The architecture is implemented by the field programmable gate array (FPGA). It is used as a hardware accelerator in a system on programmable chip (SOPC) for real-time training and classification. Experimental results reveal that the proposed RBF architecture is an effective alternative for applications where fast and efficient RBF training is desired.

show abstract

High-speed face recognition using self-adaptive radial basis function neural networks

Cited by 15 publications

References 26 publications

Weighted Fuzzy Generalized 2DFLD: A Fuzzy-Based Feature Extraction Technique for Face Recognition

Weighted Fuzzy Generalized 2DFLD: A Fuzzy-Based Feature Extraction Technique for Face Recognition

Face Recognition Using Posterior Distance Model Based Radial Basis Function Neural Networks

Efficient VLSI Architecture for Training Radial Basis Function Networks

Contact Info

Product

Resources

About