An efficient face recognition system with face image representation using averaged wavelet packet coefficients, compact and meaningful feature vectors dimensional reduction and recognition using radial basis function ...An efficient face recognition system with face image representation using averaged wavelet packet coefficients, compact and meaningful feature vectors dimensional reduction and recognition using radial basis function (RBF) neural network is presented. The face images are decomposed by 2-level two-dimensional (2-D) wavelet packet transformation. The wavelet packet coefficients obtained from the wavelet packet transformation are averaged using two different proposed methods. In the first method, wavelet packet coefficients of individual samples of a class are averaged then decomposed. The wavelet packet coefficients of all the samples of a class are averaged in the second method. The averaged wavelet packet coefficients are recognized by a RBF network. The proposed work tested on three face databases such as Olivetti-Oracle Research Lab (ORL), Japanese Female Facial Expression (JAFFE) and Essexface database. The proposed methods result in dimensionality reduction, low computational complexity and provide better recognition rates. The computational complexity is low as the dimensionality of the input pattern is reduced.展开更多
Network embedding(NE)tries to learn the potential properties of complex networks represented in a low-dimensional feature space.However,the existing deep learningbased NE methods are time-consuming as they need to tra...Network embedding(NE)tries to learn the potential properties of complex networks represented in a low-dimensional feature space.However,the existing deep learningbased NE methods are time-consuming as they need to train a dense architecture for deep neural networks with extensive unknown weight parameters.A sparse deep autoencoder(called SPDNE)for dynamic NE is proposed,aiming to learn the network structures while preserving the node evolution with a low computational complexity.SPDNE tries to use an optimal sparse architecture to replace the fully connected architecture in the deep autoencoder while maintaining the performance of these models in the dynamic NE.Then,an adaptive simulated algorithm to find the optimal sparse architecture for the deep autoencoder is proposed.The performance of SPDNE over three dynamical NE models(i.e.sparse architecture-based deep autoencoder method,DynGEM,and ElvDNE)is evaluated on three well-known benchmark networks and five real-world networks.The experimental results demonstrate that SPDNE can reduce about 70%of weight parameters of the architecture for the deep autoencoder during the training process while preserving the performance of these dynamical NE models.The results also show that SPDNE achieves the highest accuracy on 72 out of 96 edge prediction and network reconstruction tasks compared with the state-of-the-art dynamical NE algorithms.展开更多
The most common reason for blindness among human beings is Glaucoma.The increase of fluid pressure damages the optic nerve which gradually leads to irreversible loss of vision.A technique for automated screening of Gl...The most common reason for blindness among human beings is Glaucoma.The increase of fluid pressure damages the optic nerve which gradually leads to irreversible loss of vision.A technique for automated screening of Glaucoma from the fundal retinal images is presented in this paper.This paper intends to explore the significance of both the approximate and detail coefficients through wavelet packet decomposition(WPD).Decomposition is done with "db3" wavelet function and the images are decomposed up to level-3producing 84 sub-bands.Two features,the energy and the entropy are calculated for each sub-band producing two feature matrices(158 images × 84 features).The above step is purely a statistical measure based on WPD.To enhance the diagnostic accuracy,the second phase considers the structural(biological) region of interest(ROI) in the image and then extracts the same features.It is worthy to note that direct biological features are not extracted to eliminate the drawbacks of segmentation whereas the biologically significant region is taken as biological-ROI.Interestingly,the detailed coefficient sub-bands(prominent edges) show more significance in the biological-ROI phase.Apart from enhancing the diagnostic accuracy by feature reduction,the paper intends to mark the significance indices,uniqueness and discrimination capability of the significant features(sub-bands) in both the phases.Then,the crisp inputs are fed to the classifier ANN.Finally,from the significant features of the biological-ROI feature matrices,the accuracy is raised to 85%which is notable than the accuracy of 79%achieved without considering the ROI.展开更多
文摘An efficient face recognition system with face image representation using averaged wavelet packet coefficients, compact and meaningful feature vectors dimensional reduction and recognition using radial basis function (RBF) neural network is presented. The face images are decomposed by 2-level two-dimensional (2-D) wavelet packet transformation. The wavelet packet coefficients obtained from the wavelet packet transformation are averaged using two different proposed methods. In the first method, wavelet packet coefficients of individual samples of a class are averaged then decomposed. The wavelet packet coefficients of all the samples of a class are averaged in the second method. The averaged wavelet packet coefficients are recognized by a RBF network. The proposed work tested on three face databases such as Olivetti-Oracle Research Lab (ORL), Japanese Female Facial Expression (JAFFE) and Essexface database. The proposed methods result in dimensionality reduction, low computational complexity and provide better recognition rates. The computational complexity is low as the dimensionality of the input pattern is reduced.
基金National Natural Science Foundation of China,Grant/Award Numbers:62173236,61876110,61806130,61976142,82304204.
文摘Network embedding(NE)tries to learn the potential properties of complex networks represented in a low-dimensional feature space.However,the existing deep learningbased NE methods are time-consuming as they need to train a dense architecture for deep neural networks with extensive unknown weight parameters.A sparse deep autoencoder(called SPDNE)for dynamic NE is proposed,aiming to learn the network structures while preserving the node evolution with a low computational complexity.SPDNE tries to use an optimal sparse architecture to replace the fully connected architecture in the deep autoencoder while maintaining the performance of these models in the dynamic NE.Then,an adaptive simulated algorithm to find the optimal sparse architecture for the deep autoencoder is proposed.The performance of SPDNE over three dynamical NE models(i.e.sparse architecture-based deep autoencoder method,DynGEM,and ElvDNE)is evaluated on three well-known benchmark networks and five real-world networks.The experimental results demonstrate that SPDNE can reduce about 70%of weight parameters of the architecture for the deep autoencoder during the training process while preserving the performance of these dynamical NE models.The results also show that SPDNE achieves the highest accuracy on 72 out of 96 edge prediction and network reconstruction tasks compared with the state-of-the-art dynamical NE algorithms.
文摘The most common reason for blindness among human beings is Glaucoma.The increase of fluid pressure damages the optic nerve which gradually leads to irreversible loss of vision.A technique for automated screening of Glaucoma from the fundal retinal images is presented in this paper.This paper intends to explore the significance of both the approximate and detail coefficients through wavelet packet decomposition(WPD).Decomposition is done with "db3" wavelet function and the images are decomposed up to level-3producing 84 sub-bands.Two features,the energy and the entropy are calculated for each sub-band producing two feature matrices(158 images × 84 features).The above step is purely a statistical measure based on WPD.To enhance the diagnostic accuracy,the second phase considers the structural(biological) region of interest(ROI) in the image and then extracts the same features.It is worthy to note that direct biological features are not extracted to eliminate the drawbacks of segmentation whereas the biologically significant region is taken as biological-ROI.Interestingly,the detailed coefficient sub-bands(prominent edges) show more significance in the biological-ROI phase.Apart from enhancing the diagnostic accuracy by feature reduction,the paper intends to mark the significance indices,uniqueness and discrimination capability of the significant features(sub-bands) in both the phases.Then,the crisp inputs are fed to the classifier ANN.Finally,from the significant features of the biological-ROI feature matrices,the accuracy is raised to 85%which is notable than the accuracy of 79%achieved without considering the ROI.
