deep learning

摘要： Based on the artificial intelligence algorithm of RetinaNet,we propose the Ghost-RetinaNet in this paper,a fast shadow detection method for photovoltaic panels,to solve the problems of extreme target density,large overlap,high cost and poor real-time performance in photovoltaic panel shadow detection.Firstly,the Ghost CSP module based on Cross Stage Partial(CSP)is adopted in feature extraction network to improve the accuracy and detection speed.Based on extracted features,recursive feature fusion structure ismentioned to enhance the feature information of all objects.We introduce the SiLU activation function and CIoU Loss to increase the learning and generalization ability of the network and improve the positioning accuracy of the bounding box regression,respectively.Finally,in order to achieve fast detection,the Ghost strategy is chosen to lighten the size of the algorithm.The results of the experiment show that the average detection accuracy(mAP)of the algorithm can reach up to 97.17%,the model size is only 8.75 MB and the detection speed is highly up to 50.8 Frame per second(FPS),which can meet the requirements of real-time detection speed and accuracy of photovoltaic panels in the practical environment.The realization of the algorithm also provides new research methods and ideas for fault detection in the photovoltaic power generation system.

摘要： Nowadays,with the widespread application of the Internet of Things(IoT),mobile devices are renovating our lives.The data generated by mobile devices has reached a massive level.The traditional centralized processing is not suitable for processing the data due to limited computing power and transmission load.Mobile Edge Computing(MEC)has been proposed to solve these problems.Because of limited computation ability and battery capacity,tasks can be executed in the MEC server.However,how to schedule those tasks becomes a challenge,and is the main topic of this piece.In this paper,we design an efficient intelligent algorithm to jointly optimize energy cost and computing resource allocation in MEC.In view of the advantages of deep learning,we propose a Deep Learning-Based Traffic Scheduling Approach(DLTSA).We translate the scheduling problem into a classification problem.Evaluation demonstrates that our DLTSA approach can reduce energy cost and have better performance compared to traditional scheduling algorithms.

摘要： Accurate precipitation nowcasting can provide great convenience to the public so they can conduct corresponding arrangements in advance to deal with the possible impact of upcoming heavy rain.Recent relevant research activities have shown their concerns on various deep learning models for radar echo extrapolation,where radar echo maps were used to predict their consequent moment,so as to recognize potential severe convective weather events.However,these approaches suffer from an inaccurate prediction of echo dynamics and unreliable depiction of echo aggregation or dissipation,due to the size limitation of convolution filter,lack of global feature,and less attention to features from previous states.To address the problems,this paper proposes a CEMA-LSTM recurrent unit,which is embedded with a Contextual Feature Correlation Enhancement Block(CEB)and a Multi-Attention Mechanism Block(MAB).The CEB enhances contextual feature correlation and supports its model to memorize significant features for near-future prediction;the MAB uses a position and channel attention mechanism to capture global features of radar echoes.Two practical radar echo datasets were used involving the FREM and CIKM 2017 datasets.Both quantification and visualization of comparative experimental results have demonstrated outperformance of the proposed CEMA-LSTMover recentmodels,e.g.,PhyDNet,MIM and PredRNN++,etc.In particular,compared with the second-rankedmodel,its average POD,FAR and CSI have been improved by 3.87%,1.65%and 1.79%,respectively on the FREM,and by 1.42%,5.60%and 3.16%,respectively on the CIKM 2017.

摘要： We propose new hybrid Lagrange neural networks called LaNets to predict the numerical solutions of partial differential equations.That is,we embed Lagrange interpolation and small sample learning into deep neural network frameworks.Concretely,we first perform Lagrange interpolation in front of the deep feedforward neural network.The Lagrange basis function has a neat structure and a strong expression ability,which is suitable to be a preprocessing tool for pre-fitting and feature extraction.Second,we introduce small sample learning into training,which is beneficial to guide themodel to be corrected quickly.Taking advantages of the theoretical support of traditional numerical method and the efficient allocation of modern machine learning,LaNets achieve higher predictive accuracy compared to the state-of-the-artwork.The stability and accuracy of the proposed algorithmare demonstrated through a series of classical numerical examples,including one-dimensional Burgers equation,onedimensional carburizing diffusion equations,two-dimensional Helmholtz equation and two-dimensional Burgers equation.Experimental results validate the robustness,effectiveness and flexibility of the proposed algorithm.

摘要： The Earth observation remote sensing images can display ground activities and status intuitively,which plays an important role in civil and military fields.However,the information obtained from the research only from the perspective of images is limited,so in this paper we conduct research from the perspective of video.At present,the main problems faced when using a computer to identify remote sensing images are:They are difficult to build a fixed regular model of the target due to their weak moving regularity.Additionally,the number of pixels occupied by the target is not enough for accurate detection.However,the number of moving targets is large at the same time.In this case,the main targets cannot be recognized completely.This paper studies from the perspective of Gestalt vision,transforms the problem ofmoving target detection into the problem of salient region probability,and forms a Saliency map algorithm to extract moving targets.On this basis,a convolutional neural network with global information is constructed to identify and label the target.And the experimental results show that the algorithm can extract moving targets and realize moving target recognition under many complex conditions such as target’s long-term stay and small-amplitude movement.

摘要： 3D human pose estimation is a major focus area in the field of computer vision,which plays an important role in practical applications.This article summarizes the framework and research progress related to the estimation of monocular RGB images and videos.An overall perspective ofmethods integrated with deep learning is introduced.Novel image-based and video-based inputs are proposed as the analysis framework.From this viewpoint,common problems are discussed.The diversity of human postures usually leads to problems such as occlusion and ambiguity,and the lack of training datasets often results in poor generalization ability of the model.Regression methods are crucial for solving such problems.Considering image-based input,the multi-view method is commonly used to solve occlusion problems.Here,the multi-view method is analyzed comprehensively.By referring to video-based input,the human prior knowledge of restricted motion is used to predict human postures.In addition,structural constraints are widely used as prior knowledge.Furthermore,weakly supervised learningmethods are studied and discussed for these two types of inputs to improve the model generalization ability.The problem of insufficient training datasets must also be considered,especially because 3D datasets are usually biased and limited.Finally,emerging and popular datasets and evaluation indicators are discussed.The characteristics of the datasets and the relationships of the indicators are explained and highlighted.Thus,this article can be useful and instructive for researchers who are lacking in experience and find this field confusing.In addition,by providing an overview of 3D human pose estimation,this article sorts and refines recent studies on 3D human pose estimation.It describes kernel problems and common useful methods,and discusses the scope for further research.

摘要： Sofar,slope collapse detectionmainlydepends onmanpower,whichhas the followingdrawbacks:(1)lowreliability,(2)high risk of human safe,(3)high labor cost.To improve the efficiency and reduce the human investment of slope collapse detection,this paper proposes an intelligent detection method based on deep learning technology for the task.In thismethod,we first use the deep learning-based image segmentation technology to find the slope area from the captured scene image.Then the foreground motion detection method is used for detecting the motion of the slope area.Finally,we design a lightweight convolutional neural network with an attentionmechanismto recognize the detected motion object,thus eliminating the interference motion and increasing the detection accuracy rate.Experimental results on the artificial data and relevant scene data show that the proposed detection method can effectively identify the slope collapse,which has its applicative value and brilliant prospect.

摘要： Machine learning represents a growing subfield of artificial intelligence with much promise in the diagnosis,treatment,and tracking of complex conditions,including neurodegenerative disorders such as Alzheimer’s and Parkinson’s diseases.While no definitive methods of diagnosis or treatment exist for either disease,researchers have implemented machine learning algorithms with neuroimaging and motion-tracking technology to analyze pathologically relevant symptoms and biomarkers.Deep learning algorithms such as neural networks and complex combined architectures have proven capable of tracking disease-linked changes in brain structure and physiology as well as patient motor and cognitive symptoms and responses to treatment.However,such techniques require further development aimed at improving transparency,adaptability,and reproducibility.In this review,we provide an overview of existing neuroimaging technologies and supervised and unsupervised machine learning techniques with their current applications in the context of Alzheimer’s and Parkinson’s diseases.

摘要： The entry into a hyper-connected society increases the generalization of communication using SNS.Therefore,research to analyze big data accumulated in SNS and extractmeaningful information is being conducted in various fields.In particular,with the recent development of Deep Learning,the performance is rapidly improving by applying it to the field of Natural Language Processing,which is a language understanding technology to obtain accurate contextual information.In this paper,when a chatbot system is applied to the healthcare domain for counseling about diseases,the performance of NLP integrated withmachine learning for the accurate classification ofmedical subjects from text-based health counseling data becomes important.Among the various algorithms,the performance of Bidirectional Encoder Representations from Transformers was compared with other algorithms of CNN,RNN,LSTM,and GRU.For this purpose,the health counseling data of Naver Q&A service were crawled as a dataset.KoBERT was used to classify medical subjects according to symptoms and the accuracy of classification results was measured.The simulation results show that KoBERTmodel performed high performance by more than 5%and close to 18%as large as the smallest.

摘要： COVID-19 virus is certainly considered as one of the harmful viruses amongst all the illnesses in biological science. COVID-19 symptoms are fever, cough, sore throat, and headache. The paper gave a singular function for the prediction of most of the COVID-19 virus diseases and presented with the Convolutional Neural Networks and Logistic Regression which might be the supervised learning and gaining knowledge of strategies for most of COVID-19 virus diseases detection. The proposed system makes use of an 8-fold pass determination to get a correct result. The COVID-19 virus analysis dataset is taken from Microsoft Database, Kaggle, and UCI websites gaining knowledge of the repository. The proposed studies investigate Convolutional Neural Networks (CNN) and Logistic Regression (LR) about the usage of the UCI database, Kaggle, and Google Database Datasets. This paper proposed a hybrid method for COVID-19 virus, most disease analyses through reducing the dimensionality of capabilities the usage of Logistic Regression (LR), after which making use of the brand new decreased function dataset to Convolutional Neural Networks and Logistic regression. The proposed method received the accuracy of 78.82%, sensitiveness of 97.41%, and specialness of 98.73%. The overall performance of the proposed system is appraised thinking about performance, accuracy, error rate, sensitiveness, particularity, correlation and coefficient. The proposed strategies achieved the accuracy of 78.82% and 97.41% respectively through Convolutional Neural Networks and Logistic Regression.

摘要： 本发明公开了基于拓扑结构和DeepLearning‑AIS的网络威胁感知及免疫方法，原始数据流量数据格式转换后输入到流量数据可视化程序中，所得输入改进的ViT训练模型结合参数进行特征提取训练，得到基于ViT模型的自动化特征提取模型，而后处理所有会话数据，得到数据的特征表示，将数据映射到向量空间，形成高维空间内的拓扑结构；利用拓扑结构，挖掘节点之间的关联性信息，执行LBP，迭代训练节点分类器至其收敛；将流量数据与分类结果输入AIS进行进化学习；选取自动生成的各类初始加密流量特征抗原种群和记忆抗体，存入各类抗体数组和记忆抗体数组；对流量特征抗原进行人工免疫系统训练，得到所有的抗原集合，进一步对AIS免疫网络系统进行训练，直到搜索完毕。