Image-Based Flow Prediction of Vocal Folds Using 3D Convolutional Neural Networks

In this work,a three dimensional(3D)convolutional neural network(CNN)model based on image slices of various normal and pathological vocal folds is proposed for accurate and efficient prediction of glottal flows.The 3D CNN model is composed of the feature extraction block and regression block.The feature extraction block is capable of learning low dimensional features from the high dimensional image data of the glottal shape,and the regression block is employed to flatten the output from the feature extraction block and obtain the desired glottal flow data.The input image data is the condensed set of 2D image slices captured in the axial plane of the 3D vocal folds,where these glottal shapes are synthesized based on the equations of normal vibration modes.The output flow data is the corresponding flow rate,averaged glottal pressure and nodal pressure distributions over the glottal surface.The 3D CNN model is built to establish the mapping between the input image data and output flow data.The ground-truth flow variables of each glottal shape in the training and test datasets are obtained by a high-fidelity sharp-interface immersed-boundary solver.The proposed model is trained to predict the concerned flow variables for glottal shapes in the test set.The present 3D CNN model is more efficient than traditional Computational Fluid Dynamics(CFD)models while the accuracy can still be retained,and more powerful than previous data-driven prediction models because more details of the glottal flow can be provided.The prediction performance of the trained 3D CNN model in accuracy and efficiency indicates that this model could be promising for future clinical applications.

基于块编码特点的压缩视频质量增强算法

针对现有压缩视频质量增强算法未能充分利用压缩视频特点的问题,研究了视频编码与压缩视频质量增强任务之间的本质关系,并针对性地设计了一种基于三维卷积神经网络(3D convolutional neural network, 3D-CNN)的非对齐压缩视频质量增强算法。实验结果表明:相较于高效视频编码(high efficiency video coding, HEVC)标准H.265,所提算法在低延迟(low delay, LD)配置下且量化参数(quantization parameter, QP)为37时,峰值信噪比(peak signal-to-noise ratio, PSNR)提升了0.465 2 dB;相较于数据压缩会议(data compression conference, DCC)中提出的多帧引导的注意力网络(multi-frame guided attention network, MGANet)方法,该算法PSNR的增长量提升了15.1%。

Mural Anomaly Region Detection Algorithm Based on Hyperspectral Multiscale Residual Attention Network

Mural paintings hold significant historical information and possess substantial artistic and cultural value.However,murals are inevitably damaged by natural environmental factors such as wind and sunlight,as well as by human activities.For this reason,the study of damaged areas is crucial for mural restoration.These damaged regions differ significantly from undamaged areas and can be considered abnormal targets.Traditional manual visual processing lacks strong characterization capabilities and is prone to omissions and false detections.Hyperspectral imaging can reflect the material properties more effectively than visual characterization methods.Thus,this study employs hyperspectral imaging to obtain mural information and proposes a mural anomaly detection algorithm based on a hyperspectral multi-scale residual attention network(HM-MRANet).The innovations of this paper include:(1)Constructing mural painting hyperspectral datasets.(2)Proposing a multi-scale residual spectral-spatial feature extraction module based on a 3D CNN(Convolutional Neural Networks)network to better capture multiscale information and improve performance on small-sample hyperspectral datasets.(3)Proposing the Enhanced Residual Attention Module(ERAM)to address the feature redundancy problem,enhance the network’s feature discrimination ability,and further improve abnormal area detection accuracy.The experimental results show that the AUC(Area Under Curve),Specificity,and Accuracy of this paper’s algorithm reach 85.42%,88.84%,and 87.65%,respectively,on this dataset.These results represent improvements of 3.07%,1.11%and 2.68%compared to the SSRN algorithm,demonstrating the effectiveness of this method for mural anomaly detection.

Behavior recognition algorithm based on the improved R3D and LSTM network fusion

Because behavior recognition is based on video frame sequences,this paper proposes a behavior recognition algorithm that combines 3D residual convolutional neural network(R3D)and long short-term memory(LSTM).First,the residual module is extended to three dimensions,which can extract features in the time and space domain at the same time.Second,by changing the size of the pooling layer window the integrity of the time domain features is preserved,at the same time,in order to overcome the difficulty of network training and over-fitting problems,the batch normalization(BN)layer and the dropout layer are added.After that,because the global average pooling layer(GAP)is affected by the size of the feature map,the network cannot be further deepened,so the convolution layer and maxpool layer are added to the R3D network.Finally,because LSTM has the ability to memorize information and can extract more abstract timing features,the LSTM network is introduced into the R3D network.Experimental results show that the R3D+LSTM network achieves 91%recognition rate on the UCF-101 dataset.

MSF-Net: A Multilevel Spatiotemporal Feature Fusion Network Combines Attention for Action Recognition

An action recognition network that combines multi-level spatiotemporal feature fusion with an attention mechanism is proposed as a solution to the issues of single spatiotemporal feature scale extraction,information redundancy,and insufficient extraction of frequency domain information in channels in 3D convolutional neural networks.Firstly,based on 3D CNN,this paper designs a new multilevel spatiotemporal feature fusion(MSF)structure,which is embedded in the network model,mainly through multilevel spatiotemporal feature separation,splicing and fusion,to achieve the fusion of spatial perceptual fields and short-medium-long time series information at different scales with reduced network parameters;In the second step,a multi-frequency channel and spatiotemporal attention module(FSAM)is introduced to assign different frequency features and spatiotemporal features in the channels are assigned corresponding weights to reduce the information redundancy of the feature maps.Finally,we embed the proposed method into the R3D model,which replaced the 2D convolutional filters in the 2D Resnet with 3D convolutional filters and conduct extensive experimental validation on the small and medium-sized dataset UCF101 and the largesized dataset Kinetics-400.The findings revealed that our model increased the recognition accuracy on both datasets.Results on the UCF101 dataset,in particular,demonstrate that our model outperforms R3D in terms of a maximum recognition accuracy improvement of 7.2%while using 34.2%fewer parameters.The MSF and FSAM are migrated to another traditional 3D action recognition model named C3D for application testing.The test results based on UCF101 show that the recognition accuracy is improved by 8.9%,proving the strong generalization ability and universality of the method in this paper.

CoolVox:Advanced 3D convolutional neural network models for predicting solar radiation on building facades

Data-driven models have become increasingly prominent in the building,architecture,and construction industries.One area ideally suited to exploit this powerful new technology is building performance simulation.Physics-based models have traditionally been used to estimate the energy flow,air movement,and heat balance of buildings.However,physics-based models require many assumptions,significant computational power,and a considerable amount of time to output predictions.Artificial neural networks(ANNs)with prefabricated or simulated data are likely to be a more feasible option for environmental analysis conducted by designers during the early design phase.Because ANNs require fewer inputs and shorter computation times and offer superior performance and potential for data augmentation,they have received increased attention for predicting the surface solar radiation on buildings.Furthermore,ANNs can provide innovative and quick design solutions,enabling designers to receive instantaneous feedback on the effects of a proposed change to a building's design.This research introduces deep learning methods as a means of simulating the annual radiation intensities and exposure level of buildings without the need for physics-based engines.We propose the CoolVox model to demonstrate the feasibility of using 3D convolutional neural networks to predict the surface radiation on building facades.The CoolVox model accurately predicted the radiation intensities of building facades under different boundary conditions and performed better than ARINet(with average mean square errors of 0.01 and 0.036,respectively)in predicting the radiation intensity both with(validation error=0.0165)and without(validation error=0.0066)the presence of boundary buildings.

Improving Wind Forecasts Using a Gale-Aware Deep Attention Network

Numerical weather prediction of wind speed requires statistical postprocessing of systematic errors to obtain reliable and accurate forecasts.However,use of postprocessing models is often undesirable for extreme weather events such as gales.Here,we propose a postprocessing algorithm based on a gale-aware deep attention network to simultaneously improve wind speed forecasts and gale area warnings.Specifically,the algorithm includes both a galeaware loss function that focuses the model on potential gale areas,and an observation station supervision strategy that alleviates the problem of missing extreme values caused by data gridding.The effectiveness of the proposed model was verified by using data from 235 wind speed observation stations.Experimental results show that our model can produce wind speed forecasts with a root-mean-square error of 1.1547 m s^(-1),and a Hanssen–Kuipers discriminant score of 0.517,performance that is superior to that of the other postprocessing algorithms considered.