查看更多>>摘要：Crime scene investigation(CSI)image is key evidence carrier during criminal investiga-tion,in which CSI image retrieval can assist the public police to obtain criminal clues.Moreover,with the rapid development of deep learning,data-driven paradigm has become the mainstream method of CSI image feature extraction and representation,and in this process,datasets provide effective support for CSI retrieval performance.However,there is a lack of systematic research on CSI image retrieval methods and datasets.Therefore,we present an overview of the existing works about one-class and multi-class CSI image retrieval based on deep learning.According to the research,based on their technical functionalities and implementation methods,CSI image retrieval is roughly classified into five categories:feature representation,metric learning,generative adversar-ial networks,autoencoder networks and attention networks.Furthermore,We analyzed the remain-ing challenges and discussed future work directions in this field.
查看更多>>摘要：Hyper-and multi-spectral image fusion is an important technology to produce hyper-spectral and hyper-resolution images,which always depends on the spectral response function and the point spread function.However,few works have been payed on the estimation of the two degra-dation functions.To learn the two functions from image pairs to be fused,we propose a Dirichlet network,where both functions are properly constrained.Specifically,the spatial response function is constrained with positivity,while the Dirichlet distribution along with a total variation is imposed on the point spread function.To the best of our knowledge,the neural network and the Dirichlet regularization are exclusively investigated,for the first time,to estimate the degradation functions.Both image degradation and fusion experiments demonstrate the effectiveness and superiority of the proposed Dirichlet network.
查看更多>>摘要：Accurate detection of exercise fatigue based on physiological signals is vital for reason-able physical activity.As a non-invasive technology,phonocardiogram(PCG)signals possess a robust capability to reflect cardiovascular information,and their data acquisition devices are quite convenient.In this study,a novel hybrid approach of fractional Fourier transform(FRFT)com-bined with linear and discrete wavelet transform(DWT)features extracted from PCG is proposed for PCG multi-class classification.The proposed system enhances the fatigue detection performance by combining optimized FRFT features with an effective aggregation of linear features and DWT features.The FRFT technique is employed to convert the 1-D PCG signal into 2-D image which is sent to a pre-trained convolutional neural network structure,called VGG-16.The features from the VGG-16 were concatenated with the linear and DWT features to form fused features.The fused features are sent to support vector machine(SVM)to distinguish six distinct fatigue levels.Experi-mental results demonstrate that the proposed fused features outperform other feature combinations significantly.
查看更多>>摘要：Hyperspectral images typically have high spectral resolution but low spatial resolution,which impacts the reliability and accuracy of subsequent applications,for example,remote sensing classification and mineral identification.But in traditional methods via deep convolution neural net-works,indiscriminately extracting and fusing spectral and spatial features makes it challenging to utilize the differentiated information across adjacent spectral channels.Thus,we proposed a multi-branch interleaved iterative upsampling hyperspectral image super-resolution reconstruction net-work(MIIUSR)to address the above problems.We reinforce spatial feature extraction by integrat-ing detailed features from different receptive fields across adjacent channels.Furthermore,we pro-pose an interleaved iterative upsampling process during the reconstruction stage,which progres-sively fuses incremental information among adjacent frequency bands.Additionally,we add two parallel three dimensional(3D)feature extraction branches to the backbone network to extract spectral and spatial features of varying granularity.We further enhance the backbone network's construction results by leveraging the difference between two dimensional(2D)channel-grouping spatial features and 3D multi-granularity features.The results obtained by applying the proposed network model to the CAVE test set show that,at a scaling factor of×4,the peak signal to noise ratio,spectral angle mapping,and structural similarity are 37.310 dB,3.525 and 0.943 8,respec-tively.Besides,extensive experiments conducted on the Harvard and Foster datasets demonstrate the superior potential of the proposed model in hyperspectral super-resolution reconstruction.
查看更多>>摘要：The wet multi-disc clutches are extensively used in various transmission systems,with one of the most prevalent failure modes being the buckling deformation of friction components.An improved Hilbert-Huang transform method(IHHT)is proposed to address the limitations of tradi-tional time-domain vibration analyses,such as low accuracy and mode mixing.This paper first clas-sifies the buckling degree of the friction components.Next,wavelet packet transform(WPT)is applied to the vibration signals of different buckling plates to partition them into distinct fre-quency bands.Then,the instantaneous features are extracted by empirical mode decomposition(EMD)and Hilbert transform(HT)to discarding extraneous intrinsic mode function(IMF)com-ponents.Comparative analyses of Hilbert spectral entropy and time-domain features confirm the enhanced precision of IHHT under specific classifiers,which is better than traditional methods.
查看更多>>摘要：Moderate exercise contributes to health,but excessive exercise may lead to physical injury or even endanger life.It is pressing for a device that can detect the intensity of exercise.Therefore,in order to enable real-time detection of exercise intensity and mitigate the risks of harm from excessive exercise,a exercise intensity monitoring system based on the heart rate variability(HRV)from electrocardiogram(ECG)signal and linear features from phonocardiogram(PCG)signal is proposed.The main contributions include:First,accurate analysis of HRV is crucial for subsequent exercise intensity detection.To enhance HRV analysis,we propose an R-peak detector based on encoder-decoder and temporal convolutional network(TCN).Experimental results demonstrate that the proposed R-peak detector achieves an F1 score exceeding 0.99 on real high-intensity exercise ECG datasets.Second,an exercise fatigue monitoring system based on multi-sig-nal feature fusion is proposed.Initially,utilizing the proposed R-peak detector for HRV extraction in exercise intensity detection,which outperforms traditional algorithms,with the system achiev-ing a classification performance of 0.933 sensitivity,0.802 specificity,and 0.960 accuracy.To fur-ther improve the system,we combine HRV with the linear features of PCG.Our exercise intensity detection system achieves 90.2%specificity,96.7%recall,and 98.1%accuracy in five-fold cross-vali-dation.
查看更多>>摘要：Human posture estimation is a prominent research topic in the fields of human-com-puter interaction,motion recognition,and other intelligent applications.However,achieving high accuracy in key point localization,which is crucial for intelligent applications,contradicts the low detection accuracy of human posture detection models in practical scenarios.To address this issue,a human pose estimation network called AT-HRNet has been proposed,which combines convolu-tional self-attention and cross-dimensional feature transformation.AT-HRNet captures significant feature information from various regions in an adaptive manner,aggregating them through convolu-tional operations within the local receptive domain.The residual structures TripNeck and Trip-Block of the high-resolution network are designed to further refine the key point locations,where the attention weight is adjusted by a cross-dimensional interaction to obtain more features.To vali-date the effectiveness of this network,AT-HRNet was evaluated using the COCO2017 dataset.The results show that AT-HRNet outperforms HRNet by improving 3.2%in mAP,4.0%in AP75,and 3.9%in APM.This suggests that AT-HRNet can offer more beneficial solutions for human posture estimation.
NETL
NSTL
万方数据