A comparative analysis of different neural networks for face recognition using principal component analysis, wavelets and efficient variable learning rate

Bhati, Raman; Jain, Sarika; Maltare, Nilesh; Mishra, Durgesh Kumar

doi:10.1109/iccct.2010.5640486

Cited by 9 publications

(4 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Atanassov et al [3] made a generalized net for parallel optimization of Multi-Layer Feedforward Neural Network on assigned training pairs with variable learning rate backpropagation algorithm. Bhati et al [5] showed that single layer neural network can give 100% recognition accuracy if correct learning rate is assigned to it. Li et al [23] showed that the improved VLBP can quickly and accurately predict the concentration of target elements in Energy Dispersive X-Ray Fluorescence (EDXRF).…”

Section: Variable Learning Rate Backpropagation (Gd)mentioning

confidence: 99%

Effect of Hidden Neuron Size on Different Training Algorithm in Neural Network

Kumar¹,

Sodhi²

2022

Comm. Math. Appl.

View full text Add to dashboard Cite

We use different types of training algorithms in the neural network. But, we cannot say which kind of training algorithm is fast for a given problem. So, in this survey paper, we are trying to find which types of training algorithm are better for categorization problems. For this purpose, we used ten types of training algorithms in the pattern network in MATLAB. We used Levenberg-Marquardt (LM), Bayesian regularization backpropagation (BR), BFGS Quasi-Newton (BFG), Resilient Backpropagation (RP), Scaled Conjugate gradient backpropagation (SCG), Conjugate Gradient with Powell/Beale Restarts (CGB), Fletcher-Powell Conjugate Gradient (CGF), Polak-Ribiere Conjugate Gradient (GDM), One Step Secant (OSS), and Variable Learning Rate Backpropagation (GD) algorithm. In this survey paper, we also check, affects of these algorithms on neural network when we applied different types of hidden neuron size. During this survey we found some new facts. We found that RP, SCG, CGB, CGF and OSS are fastest algorithms. BFG takes more time with respect to hidden neuron size. GDM and GD take more epochs. BR algorithm is not acceptable for image categorization.

show abstract

Section: Variable Learning Rate Backpropagation (Gd)mentioning

confidence: 99%

Effect of Hidden Neuron Size on Different Training Algorithm in Neural Network

Kumar¹,

Sodhi²

2022

Comm. Math. Appl.

View full text Add to dashboard Cite

show abstract

“…Raman Bhati et al [6] represented the feature vectors with the help of Eigenfaces. They proposed a technique for the selection of the learning rate in a single layer feed forward neural network and BPNN also.…”

Section: Human Facesmentioning

confidence: 99%

“…As this is orthogonal transform so the energy of the image is matched up with the energy of the given coefficients. In the below given figure the values VF 5 , VF 6 , VF 7 finding the resultant scale matrices. 5.…”

Section: Multilevel Haar Wavelet Transformmentioning

confidence: 99%

See 1 more Smart Citation

Modular Approach for Face Recognition System using Multilevel Haar Wavelet Transform, Improved PCA and Enhanced Back Propagation Neural Network

Agarwal¹,

Prakash²

2013

IJCA

View full text Add to dashboard Cite

With rapidly developing technology, it is important to invent an efficient and effective security system that will help to detect unauthorized access on any system. Hence it is essential necessary to implement a highly secure, economic and reliable face recognition system to enable the protection of computer systems from unauthorized access. So, in this paper a fast efficient approach is proposed for the recognition of human faces. Thus, firstly Haar wavelet transform is implemented for multilevel decomposition of face image into several subband images. The decomposed image subbands are then used as input by Improved Principal Component Analysis (IPCA) approach for extracting features with the help of eigenvalues and Eigen vectors. Then classification of features using Back Propagation Neural Network (BPNN) is done and finally image is being recognized comparing it to the testing images existing in the images database. More efficient BPNN is used to improve the recognition rate and to overcome the problems associated with variations in illumination and poses. Thus feed forward neural network shows the effectiveness of the proposed algorithm.

show abstract

Biologically Inspired Algorithm for Increasing the Number of Artificial Neurons

Kolobashkina

2019

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

A comparative analysis of different neural networks for face recognition using principal component analysis, wavelets and efficient variable learning rate

Cited by 9 publications

References 6 publications

Effect of Hidden Neuron Size on Different Training Algorithm in Neural Network

Effect of Hidden Neuron Size on Different Training Algorithm in Neural Network

Modular Approach for Face Recognition System using Multilevel Haar Wavelet Transform, Improved PCA and Enhanced Back Propagation Neural Network

Biologically Inspired Algorithm for Increasing the Number of Artificial Neurons

Contact Info

Product

Resources

About