Notwithstanding the discovery of vaccines for Covid-19, the virus'srapid spread continues due to the limited availability of vaccines, especially inpoor and emerging countries. Therefore, the key issues in the pre...Notwithstanding the discovery of vaccines for Covid-19, the virus'srapid spread continues due to the limited availability of vaccines, especially inpoor and emerging countries. Therefore, the key issues in the presentCOVID-19 pandemic are the early identification of COVID-19, the cautiousseparation of infected cases at the lowest cost and curing the disease in the earlystages. For that reason, the methodology adopted for this study is imaging tools,particularly computed tomography, which have been critical in diagnosing andtreating the disease. A new method for detecting Covid-19 in X-rays and CTimages has been presented based on the Scatter Wavelet Transform and DenseDeep Neural Network. The Scatter Wavelet Transform has been employed as afeature extractor, while the Dense Deep Neural Network is utilized as a binaryclassifier. An extensive experiment was carried out to evaluate the accuracy ofthe proposed method over three datasets: IEEE 80200, Kaggle, andCovid-19 X-ray image data Sets. The dataset used in the experimental part consists of 14142. The numbers of training and testing images are 8290 and 2810,respectively. The analysis of the result refers that the proposed methods achievedhigh accuracy of 98%. The proposed model results show an excellent outcomecompared to other methods in the same domain, such as (DeTraC) CNN, whichachieved only 93.1%, CNN, which achieved 94%, and stacked Multi-ResolutionCovXNet, which achieved 97.4%. The accuracy of CapsNet reached 97.24%.展开更多
Objective:PCG represents the acoustic replay of heart sounds from the cardiac structure.To detect and analyse the different conditions of the heart,heart sound signals are essential.CVD is detected by classifiers who ...Objective:PCG represents the acoustic replay of heart sounds from the cardiac structure.To detect and analyse the different conditions of the heart,heart sound signals are essential.CVD is detected by classifiers who superficially identify the cardiac features.Abnormal sounds in systole or diastole could indicate valve stenosis or regurgitation.The presence of S3 or S4 sounds could indicate heart failure or stiffening of the heart muscle.This paper proposes a CVD detection technique using improved WST and DA classifiers.Method:The PCG was obtained from the Physionet dataset.The raw signals were pre-processed using 2D DCT.The 2D DCT was applied to a matrix containing 3000 sounds with 10000 samples.The DCT matrix was then filtered using a 20Hz–190 Hz Type II Chebyshev filter to remove the high frequency noise above 190 Hz.After filtering,the denoised PCG matrix was obtained from the DCT matrix using inverse 2D DCT.The PCG matrix was feature extracted using WST.WST produces low-frequency components by using the LPFs to filter high-frequency components.These features were then used with the DA classifier to predict the CVD.The DA classifier uses discriminant analysis pattern classification.The DA classifier learns the training PCG pattern,from WST features,and then classifies test samples as normal or abnormal.Results:The proposed method removed noise up to 99%.The 2D DCT filter provided an average noise improvement of 37.34 dB.Further tuning in filter order or attenuation level resulted in distortion of the PCG,and noise improvement declined.The DCT filter removed up to 99%of noise as per the SNR estimation.The improved WST and the DA classifier resulted in an accuracy of 99.63%.Conclusion:Comparative analysis with DNN,advocates the superiority of the proposed method.DNN classifiers provide accurate CVD classification but require a more expensive and complex GPU.The DA classifier requires only a CPU.This work demonstrated that superior CVD classification was obtained using a combination of WST features and the DA classifier with 94%accuracy.展开更多
Face attribute classification(FAC)is a high-profile problem in biometric verification and face retrieval.Although recent research has been devoted to extracting more delicate image attribute features and exploiting th...Face attribute classification(FAC)is a high-profile problem in biometric verification and face retrieval.Although recent research has been devoted to extracting more delicate image attribute features and exploiting the inter-attribute correlations,significant challenges still remain.Wavelet scattering transform(WST)is a promising non-learned feature extractor.It has been shown to yield more discriminative representations and outperforms the learned representations in certain tasks.Applied to the image classification task,WST can enhance subtle image texture information and create local deformation stability.This paper designs a scattering-based hybrid block,to incorporate frequency-domain(WST)and image-domain features in a channel attention manner(Squeezeand-Excitation,SE),termed WS-SE block.Compared with CNN,WS-SE achieves a more efficient FAC performance and compensates for the model sensitivity of the small-scale affine transform.In addition,to further exploit the relationships among the attribute labels,we propose a learning strategy from a causal view.The cause attributes defined using the causalityrelated information can be utilized to infer the effect attributes with a high confidence level.Ablative analysis experiments demonstrate the effectiveness of our model,and our hybrid model obtains state-of-the-art results in two public datasets.展开更多
Human vision depends heavily on retinal tissue.The loss of eyesight may result from infections of the retinal tissues that are treated slowly or do not work at all.Additionally,the diagnosis is susceptible to inaccura...Human vision depends heavily on retinal tissue.The loss of eyesight may result from infections of the retinal tissues that are treated slowly or do not work at all.Additionally,the diagnosis is susceptible to inaccuracies when a large dataset is involved.Therefore,a fully automated transfer learning approach for diagnosing diabetic retinopathy(DR)is suggested to minimize human intervention while maintaining high classification accuracy.To address this issue,we proposed a transfer learning-based trilateral attention network(TaNet)for the classification.To boost the visual quality of the DR pictures,a contrast constrained adaptive histogram equalization approach is applied.The pre-processed pictures are then segmented using a bilateral segmentation network(BiSeNet).The BiSeNet segmented the optic disc and blood vessels individually.After the completion of segmentation,the features are extracted.Feature extraction is based on the wavelet scattering transformation approach.The results of many trials were evaluated against the Messidor-2,EYEPACS,and APTOS 2019 datasets.The proposed model was created using a refined pre-trained technique and transfer learning methodology.Finally,the suggested framework was tested using efficiency assessment methods,and the classification rate was recorded as having above 98%sensitivity,specificity,precision,and accuracy.The proposed approach yields greater performance and shows enhancement towards the existing approach.展开更多
Side-channel analysis(SCA)has become an increasing important method to assess the physical security of cryptographic systems.In the process of SCA,the number of attack data directly determines the performance of SCA.W...Side-channel analysis(SCA)has become an increasing important method to assess the physical security of cryptographic systems.In the process of SCA,the number of attack data directly determines the performance of SCA.With sufficient attack data,the adversary can achieve a successful SCA.However,in reality,the cryptographic device may be protected with some countermeasures to limit the number of encryptions using the same key.In this case,the adversary cannot use casual numbers of data to perform SCA.The performance of SCA will be severely dropped if the attack traces are insufficient.In this paper,we introduce wavelet scatter transform(WST)and short-time fourier transform(STFT)to non-profiled side-channel analysis domains,to improve the performance of side-channel attacks in the context of insufficient data.We design a practical framework to provide suitable parameters for WST/STFT-based SCA.Using the proposed method,the WST/STFT-based SCA method can significantly enhance the performance and robustness of non-profiled SCA.The practical attacks against four public datasets show that the proposed method is able to achieve more robust performance.Compared with the original correlation power analysis(CPA),the number of attack data can be reduced by 50–95%.展开更多
文摘Notwithstanding the discovery of vaccines for Covid-19, the virus'srapid spread continues due to the limited availability of vaccines, especially inpoor and emerging countries. Therefore, the key issues in the presentCOVID-19 pandemic are the early identification of COVID-19, the cautiousseparation of infected cases at the lowest cost and curing the disease in the earlystages. For that reason, the methodology adopted for this study is imaging tools,particularly computed tomography, which have been critical in diagnosing andtreating the disease. A new method for detecting Covid-19 in X-rays and CTimages has been presented based on the Scatter Wavelet Transform and DenseDeep Neural Network. The Scatter Wavelet Transform has been employed as afeature extractor, while the Dense Deep Neural Network is utilized as a binaryclassifier. An extensive experiment was carried out to evaluate the accuracy ofthe proposed method over three datasets: IEEE 80200, Kaggle, andCovid-19 X-ray image data Sets. The dataset used in the experimental part consists of 14142. The numbers of training and testing images are 8290 and 2810,respectively. The analysis of the result refers that the proposed methods achievedhigh accuracy of 98%. The proposed model results show an excellent outcomecompared to other methods in the same domain, such as (DeTraC) CNN, whichachieved only 93.1%, CNN, which achieved 94%, and stacked Multi-ResolutionCovXNet, which achieved 97.4%. The accuracy of CapsNet reached 97.24%.
文摘Objective:PCG represents the acoustic replay of heart sounds from the cardiac structure.To detect and analyse the different conditions of the heart,heart sound signals are essential.CVD is detected by classifiers who superficially identify the cardiac features.Abnormal sounds in systole or diastole could indicate valve stenosis or regurgitation.The presence of S3 or S4 sounds could indicate heart failure or stiffening of the heart muscle.This paper proposes a CVD detection technique using improved WST and DA classifiers.Method:The PCG was obtained from the Physionet dataset.The raw signals were pre-processed using 2D DCT.The 2D DCT was applied to a matrix containing 3000 sounds with 10000 samples.The DCT matrix was then filtered using a 20Hz–190 Hz Type II Chebyshev filter to remove the high frequency noise above 190 Hz.After filtering,the denoised PCG matrix was obtained from the DCT matrix using inverse 2D DCT.The PCG matrix was feature extracted using WST.WST produces low-frequency components by using the LPFs to filter high-frequency components.These features were then used with the DA classifier to predict the CVD.The DA classifier uses discriminant analysis pattern classification.The DA classifier learns the training PCG pattern,from WST features,and then classifies test samples as normal or abnormal.Results:The proposed method removed noise up to 99%.The 2D DCT filter provided an average noise improvement of 37.34 dB.Further tuning in filter order or attenuation level resulted in distortion of the PCG,and noise improvement declined.The DCT filter removed up to 99%of noise as per the SNR estimation.The improved WST and the DA classifier resulted in an accuracy of 99.63%.Conclusion:Comparative analysis with DNN,advocates the superiority of the proposed method.DNN classifiers provide accurate CVD classification but require a more expensive and complex GPU.The DA classifier requires only a CPU.This work demonstrated that superior CVD classification was obtained using a combination of WST features and the DA classifier with 94%accuracy.
基金supported by the National Key Research and Development Project of China(Grant No.2018AAA0100802)Opening Foundation of National Engineering Laboratory for Intelligent Video Analysis and Application,and Experimental Center of Artificial Intelligence of Beijing Normal University.
文摘Face attribute classification(FAC)is a high-profile problem in biometric verification and face retrieval.Although recent research has been devoted to extracting more delicate image attribute features and exploiting the inter-attribute correlations,significant challenges still remain.Wavelet scattering transform(WST)is a promising non-learned feature extractor.It has been shown to yield more discriminative representations and outperforms the learned representations in certain tasks.Applied to the image classification task,WST can enhance subtle image texture information and create local deformation stability.This paper designs a scattering-based hybrid block,to incorporate frequency-domain(WST)and image-domain features in a channel attention manner(Squeezeand-Excitation,SE),termed WS-SE block.Compared with CNN,WS-SE achieves a more efficient FAC performance and compensates for the model sensitivity of the small-scale affine transform.In addition,to further exploit the relationships among the attribute labels,we propose a learning strategy from a causal view.The cause attributes defined using the causalityrelated information can be utilized to infer the effect attributes with a high confidence level.Ablative analysis experiments demonstrate the effectiveness of our model,and our hybrid model obtains state-of-the-art results in two public datasets.
文摘Human vision depends heavily on retinal tissue.The loss of eyesight may result from infections of the retinal tissues that are treated slowly or do not work at all.Additionally,the diagnosis is susceptible to inaccuracies when a large dataset is involved.Therefore,a fully automated transfer learning approach for diagnosing diabetic retinopathy(DR)is suggested to minimize human intervention while maintaining high classification accuracy.To address this issue,we proposed a transfer learning-based trilateral attention network(TaNet)for the classification.To boost the visual quality of the DR pictures,a contrast constrained adaptive histogram equalization approach is applied.The pre-processed pictures are then segmented using a bilateral segmentation network(BiSeNet).The BiSeNet segmented the optic disc and blood vessels individually.After the completion of segmentation,the features are extracted.Feature extraction is based on the wavelet scattering transformation approach.The results of many trials were evaluated against the Messidor-2,EYEPACS,and APTOS 2019 datasets.The proposed model was created using a refined pre-trained technique and transfer learning methodology.Finally,the suggested framework was tested using efficiency assessment methods,and the classification rate was recorded as having above 98%sensitivity,specificity,precision,and accuracy.The proposed approach yields greater performance and shows enhancement towards the existing approach.
基金This work is supported in part by National Key R&D Program of China(No.2022YFB3103800)National Natural Science Foundation of China(No.U1936209,No.62002353,No.62202231 and No.62202230)+2 种基金China Postdoctoral Science Foundation(No.2021M701726)Jiangsu Funding Program for Excellent Postdoctoral Talent(No.2022ZB270)Yunnan Provincial Major Science and Technology Special Plan Projects(No.202103AA080015).
文摘Side-channel analysis(SCA)has become an increasing important method to assess the physical security of cryptographic systems.In the process of SCA,the number of attack data directly determines the performance of SCA.With sufficient attack data,the adversary can achieve a successful SCA.However,in reality,the cryptographic device may be protected with some countermeasures to limit the number of encryptions using the same key.In this case,the adversary cannot use casual numbers of data to perform SCA.The performance of SCA will be severely dropped if the attack traces are insufficient.In this paper,we introduce wavelet scatter transform(WST)and short-time fourier transform(STFT)to non-profiled side-channel analysis domains,to improve the performance of side-channel attacks in the context of insufficient data.We design a practical framework to provide suitable parameters for WST/STFT-based SCA.Using the proposed method,the WST/STFT-based SCA method can significantly enhance the performance and robustness of non-profiled SCA.The practical attacks against four public datasets show that the proposed method is able to achieve more robust performance.Compared with the original correlation power analysis(CPA),the number of attack data can be reduced by 50–95%.
