Sleep disorders are receiving more and more attention， and the accuracy and generalization of automated sleep stage classification are facing more and more challenges. However， due to the very limited human sleep data publicly available， the sleep stage classification task is actually similar to a few-shot scenario. And due to the widespread individual differences in sleep features， it is difficult for existing machine learning models to guarantee accurate classification of data from new subjects who have not participated in the training. In order to achieve accurate stage classification of new subjects’ sleep data， existing studies usually require additional collection and labeling of large amounts of data from new subjects and personalized fine-tuning of the model. Based on this， a new sleep stage classification model， Meta Transfer Sleep Learner （MTSL）， was proposed. Inspired by the idea of Scale & Shift based weight transfer strategy in transfer learning， a new meta transfer learning framework was designed. The training phase included two steps： pre-training and meta transfer training， and many meta-tasks were used for meta transfer training. In the test phase， the model could be easily adapted to the feature distribution of new subjects by fine-tuning with only a few new subjects’ data， which greatly reduced the cost of accurate sleep stage classification for new subjects. Experimental results on two public sleep datasets show that MTSL model can achieve higher accuracy and F1-score under both single-dataset and cross-dataset conditions. This indicates that MTSL is more suitable for sleep stage classification tasks in few-shot scenarios.

In multi-agent systems， there are multiple cooperative tasks that change with time and multiple conflict optimization objective functions. To build a multi-agent system， the dynamic multiobjective multi-agent cooperative scheduling problem becomes one of critical problems. To solve this problem， a probability-driven dynamic prediction strategy was proposed to utilize the probability distributions in historical environments to predict the ones in new environments， thus generating new solutions and realizing the fast response to environmental changes. In detail， an element-based representation for probability distributions was designed to represent the adaptability of elements in dynamic environments， and the probability distributions were gradually updated towards real distributions according to the best solutions found by optimization algorithms in each iteration. Taking into account continuity and relevance of environmental changes， a fusion-based prediction mechanism was built to predict the probability distributions and to provide a priori knowledge of new environments by fusing historical probability distributions when the environment changes. A new heuristic-based sampling mechanism was also proposed by combining probability distributions and heuristic information to generate new solutions for updating out-of-date populations. The proposed probability-driven dynamic prediction strategy can be inserted into any multiobjective evolutionary algorithms， resulting in probability-driven dynamic multiobjective evolutionary algorithms. Experimental results on 10 dynamic multiobjective multi-agent cooperative scheduling problem instances show that the proposed algorithms outperform the competing algorithms in terms of solution optimality and diversity， and the proposed probability-driven dynamic prediction strategy can improve the performance of multiobjective evolutionary algorithms in dynamic environments.

Aiming at the problems of detail information loss and low segmentation accuracy in the segmentation of day and night ground-based cloud images， a segmentation network called CloudResNet-UNetwork （CloudRes-UNet） for day and night ground-based cloud images based on improved Res-UNet （Residual network-UNetwork） was proposed， in which the overall network structure of encoder-decoder was adopted. Firstly， ResNet50 was used by the encoder to extract features to enhance the feature extraction ability. Then， a Multi-Stage feature extraction （Multi-Stage） module was designed， which combined three techniques of group convolution， dilated convolution and channel shuffle to obtain high-intensity semantic information. Secondly， Efficient Channel Attention Network （ECA?Net） module was added to focus on the important information in the channel dimension， strengthen the attention to the cloud region in the ground-based cloud image， and improve the segmentation accuracy. Finally， bilinear interpolation was used by the decoder to upsample the features， which improved the clarity of the segmented image and reduced the loss of object and position information. The experimental results show that， compared with the state-of-the-art ground-based cloud image segmentation network Cloud-UNetwork （Cloud-UNet） based on deep learning， the segmentation accuracy of CloudRes-UNet on the day and night ground-based cloud image segmentation dataset is increased by 1.5 percentage points， and the Mean Intersection over Union （MIoU） is increased by 1.4 percentage points， which indicates that CloudRes-UNet obtains cloud information more accurately. It has positive significance for weather forecast， climate research， photovoltaic power generation and so on.

Gliomas are the most common primary cranial tumors arising from cancerous changes in the glia of the brain and spinal cord， with a high proportion of malignant gliomas and a significant mortality rate. Quantitative segmentation and grading of gliomas based on Magnetic Resonance Imaging （MRI） images is the main method for diagnosis and treatment of gliomas. To improve the segmentation accuracy and speed of glioma， a 3D-Ghost Convolutional Neural Network （CNN） -based MRI image segmentation algorithm for glioma， called 3D-GA-Unet， was proposed. 3D-GA-Unet was built based on 3D U-Net （3D U-shaped Network）. A 3D-Ghost CNN block was designed to increase the useful output and reduce the redundant features in traditional CNNs by using linear operation. Coordinate Attention （CA） block was added， which helped to obtain more image information that was favorable to the segmentation accuracy. The model was trained and validated on the publicly available glioma dataset BraTS2018. The experimental results show that 3D-GA-Unet achieves average Dice Similarity Coefficients （DSCs） of 0.863 2， 0.847 3 and 0.803 6 and average sensitivities of 0.867 6， 0.949 2 and 0.831 5 for Whole Tumor （WT）， Tumour Core （TC）， and Enhanced Tumour （ET） in glioma segmentation results. It is verified that 3D-GA-Unet can accurately segment glioma images and further improve the segmentation efficiency， which is of positive significance for the clinical diagnosis of gliomas.

Current Video Super-Resolution （VSR） algorithms cannot fully utilize inter-frame information of different distances when processing complex scenes with large motion amplitude， resulting in difficulty in accurately recovering occlusion， boundaries， and multi-detail regions. A VSR model based on frame straddling optical flow was proposed to solve these problems. Firstly， shallow features of Low-Resolution frames （LR） were extracted through Residual Dense Blocks （RDBs）. Then， motion estimation and compensation was performed on video frames using a Spatial Pyramid Network （SPyNet） with straddling optical flows of different time lengths， and deep feature extraction and correction was performed on inter-frame information through RDBs connected in multiple layers. Finally， the shallow and deep features were fused， and High-Resolution frames （HR） were obtained through up-sampling. The experimental results on the REDS4 public dataset show that compared with deep Video Super-Resolution network using Dynamic Upsampling Filters without explicit motion compensation （DUF-VSR）， the proposed model improves Peak Signal-to-Noise Ratio （PSNR） and Structure Similarity Index Measure （SSIM） by 1.07 dB and 0.06， respectively. The experimental results show that the proposed model can effectively improve the quality of video image reconstruction.

Concerning the trade-off between convergence and diversity in solving the multi-trip pickup and delivery Vehicle Routing Problem （VRP）， a hybrid Non-dominated Sorting Genetic Algorithm Ⅱ （NSGA-Ⅱ） combining Adaptive Large Neighborhood Search （ALNS） algorithm and Adaptive Neighborhood Selection （ANS）， called NSGA-Ⅱ-ALNS-ANS， was proposed. Firstly， considering the influence of the initial population on the convergence speed of the algorithm， an improved regret insertion method was employed to obtain high-quality initial population. Secondly， to improve global and local search capabilities of the algorithm， various destroy-repair operators and neighborhood structures were designed， according to the characteristics of the pickup and delivery problem. Finally， a Best Fit Decreasing （BFD） algorithm based on random sampling and an efficient feasible solution evaluation criterion were proposed to generate vehicle routing schemes. The simulation experiments were conducted on public benchmark instances of different scales， in the comparison experiments with the MA （Memetic Algorithm）， the optimal solution quality of the proposed algorithm increased by 27%. The experimental results show that the proposed algorithm can rapidly generate high-quality vehicle routing schemes that satisfy multiple constraints， and outperform the existing algorithms in terms of both convergence and diversity.

Due to the ambiguity of text and the lack of location information in training data， current state-of-the-art diffusion model cannot accurately control the locations of generated objects in the image under the condition of text prompts. To address this issue， a spatial condition of the object’s location range was introduced， and an attention-guided method was proposed based on the strong correlation between the cross-attention map in U-Net and the image spatial layout to control the generation of the attention map， thus controlling the locations of the generated objects. Specifically， based on the Stable Diffusion （SD） model， in the early stage of the generation of the cross-attention map in the U-Net layer， a loss was introduced to stimulate high attention values in the corresponding location range， and reduce the average attention value outside the range. The noise vector in the latent space was optimized step by step in each denoising step to control the generation of the attention map. Experimental results show that the proposed method can significantly control the locations of one or more objects in the generated image， and when generating multiple objects， it can reduce the phenomenon of object omission， redundant object generation， and object fusion.

Aspect-level sentiment analysis aims to predict the sentiment polarity of specific target in given text. Aiming at the problem of ignoring the syntactic relationship between aspect words and context and reducing the attention difference caused by average pooling， an aspect-level sentiment analysis model based on Alternating-Attention （AA） mechanism and Graph Convolutional Network （AA-GCN） was proposed. Firstly， the Bidirectional Long Short-Term Memory （Bi-LSTM） network was used to semantically model context and aspect words. Secondly， the GCN based on syntactic dependency tree was used to learn location information and dependencies， and the AA mechanism was used for multi-level interactive learning to adaptively adjust the attention to the target word. Finally， the final classification basis was obtained by splicing the corrected aspect features and context features. Compared with the Target-Dependent Graph Attention Network （TD-GAT）， the accuracies of the proposed model on four public datasets increased by 1.13%-2.67%， and the F1 values on five public datasets increased by 0.98%-4.89%， indicating the effectiveness of using syntactic relationships and increasing keyword attention.

To solve the problems of uncertain influence factors and indicator quantification difficulty in the risk assessment of industrial control networks， a method based on fuzzy theory and attack tree was proposed， and the proposed method was tested and verified on Chinese Train Control System （CTCS）. First， an attack tree model for CTCS was constructed based on network security threats and system vulnerability. α-cut Triangular Fuzzy Number （TFN） was used to calculate the interval probabilities of leaf nodes and attack paths. Then， Analytic Hierarchy Process （AHP） was adopted to establish the mathematical model for security event losses and get the final risk assessment result. Finally， the experimental result demonstrates that the proposed method implements system risk assessment effectively， predicts the attack paths successfully and reduces the influence of subjective factors. By taking advantage of the proposed method， the risk assessment result would be more realistic and provides reference and basis for the selection of security protection strategies.

Exact cover problems are NP complete problems in combinatorial optimization， and it is difficult to solve them in polynomial time by using classical algorithms. In order to solve this problem， on the open source quantum computing framework qiskit， a quantum circuit solution based on Quantum Approximate Optimization Algorithm （QAOA） was proposed， and Constrained Optimization BY Linear Approximation （COBYLA） algorithm based on the simplex method was used to optimize the parameters in the quantum logic gates. Firstly， the classical Ising model was established through the mathematical model of the exact cover problem. Secondly， the classical Ising model was quantized by using the rotation variable in quantum theory， and then the Pauli rotation operator was used to replace the rotation variable to obtain the quantum Ising model and the problem Hamiltonian， which improved the speed of QAOA in finding the optimal solution. Finally， the expected expression of the problem Hamiltonian was obtained by the accumulation of the product of the unitary transformation with the mixed Hamiltonian as the generator and the unitary transformation with the problem Hamiltonian as the generator， and the generative quantum circuit was designed accordingly. In addition， the classical processor was used to optimize the parameters in the two unitary transformations to adjust the expected value of the problem Hamiltonian， thereby increasing the probability of solution. The circuit was simulated on qiskit， IBM’s open source quantum computing framework. Experimental results show that the proposed scheme can obtain the solution of the problem in polynomial time with a probability of 95.6%， which proves that the proposed quantum circuit can find a solution to the exact cover problem with a higher probability.

To address the problems of poor effects， easily falling into suboptimal solutions， and inefficiency in neural network hyperparameter optimization， an Improved Real Coding Genetic Algorithm （IRCGA） based hyperparameter optimization algorithm for the neural network was proposed， which was named IRCGA-DNN （IRCGA for Deep Neural Network）. Firstly， a real-coded form was used to represent the values of hyperparameters， which made the search space of hyperparameters more flexible. Then， a hierarchical proportional selection operator was introduced to enhance the diversity of the solution set. Finally， improved single-point crossover and variational operators were designed to explore the hyperparameter space more thoroughly and improve the efficiency and quality of the optimization algorithm， respectively. Two simulation datasets were used to show IRCGA’s performance in damage effectiveness prediction and convergence efficiency. The experimental results on two datasets indicate that， compared to GA-DNN（Genetic Algorithm for Deep Neural Network）， the proposed algorithm reduces the convergence iterations by 8.7% and 13.6% individually， and the MSE （Mean Square Error） is not much different； compared to IGA-DNN（Improved Genetic Algorithm for Deep Neural Network）， IRCGA-DNN achieves reductions of 22.2% and 13.6% in convergence iterations respectively. Experimental results show that the proposed algorithm is better in both convergence speed and prediction performance， and is suitable for hyperparametric optimization of neural networks.

Aiming at the correlation problem between sleep stages and sleep apnea hypopnea， a sleep physiological time series classification method based on adaptive multi-task learning was proposed. Single-channel electroencephalogram and electrocardiogram were used for sleep staging and Sleep Apnea Hypopnea Syndrome （SAHS） detection. A two-stream time dependence learning module was utilized to extract shared features under joint supervision of the two tasks. The correlation between sleep stages and sleep apnea hypopnea was modeled by the adaptive inter-task correlation learning module with channel attention mechanism. The experimental results on two public datasets indicate that the proposed method can complete sleep staging and SAHS detection simultaneously. On UCD dataset， the accuracy， MF1（Macro F1-score）， and Area Under the receiver characteristic Curve （AUC） for sleep staging of the proposed method were 1.21 percentage points， 1.22 percentage points， and 0.008 3 higher than those of TinySleepNet； its MF2 （Macro F2-score）， AUC， and recall of SAHS detection were 11.08 percentage points， 0.053 7， and 15.75 percentage points higher than those of the 6-layer CNN model， which meant more disease segments could be detected. The proposed method could be applied to home sleep monitoring or mobile medical to achieve efficient and convenient sleep quality assessment， assisting doctors in preliminary diagnosis of SAHS.

Concerning the current lack of effective development and deployment tools for deep learning applications， a component-based development framework for deep learning applications was proposed. The framework splits functions according to the type of resource consumption， uses a review-guided resource allocation scheme for bottleneck elimination， and uses a step-by-step boxing scheme for function placement that takes into account high CPU utilization and low memory overhead. The real-time license plate number detection application developed based on this framework achieved 82% GPU utilization in throughput-first mode， 0.73 s average application latency in latency-first mode， and 68.8% average CPU utilization in three modes （throughput-first mode， latency-first mode， and balanced throughput/latency mode）. The experimental results show that based on this framework， a balanced configuration of hardware throughput and application latency can be performed to efficiently utilize the computing resources of the platform in the throughput-first mode and meet the low latency requirements of the applications in the latency-first mode. Compared with MediaPipe， the use of this framework enabled ultra-real-time multi-person pose estimation application development， and the detection frame rate of the application was improved by up to 1 077%. The experimental results show that the framework is an effective solution for deep learning application development and deployment on CPU-GPU heterogeneous servers.

In the existing Dual LAN （Local Area Network） Terahertz Wireless LAN （Dual-LAN THz WLAN） related MAC （Medium Access Control） protocol， some nodes may repeatedly send the same Channel Time Request （CTRq） frame within multiple superframes to apply for time slot resources and idle time slots exist in some periods of network operation， therefore an efficient MAC protocol based on spontaneous data transmission SDTE-MAC （high-Efficiency MAC Protocol based on Spontaneous Data Transmission） was proposed. SDTE-MAC protocol enabled each node to maintain one or more time unit linked lists to synchronize with the rest of the nodes in the network running time， so as to know where each node started sending data frames at the channel idle time slot. The protocol optimized the traditional channel slot allocation and channel remaining slot reallocation processes， improved network throughput and channel slot utilization， reduced data delay， and could further improve the performance of Dual-LAN THz WLAN. The simulation results showed that when the network saturates， compared with the new N-CTAP （Normal Channel Time Allocation Period） slot resource allocation mechanism and adaptive shortening superframe period mechanism in the AHT-MAC （Adaptive High Throughout multi-pan MAC protocol）， the MAC layer throughput of the SDTE-MAC protocol was increased by 9.2%， the channel slot utilization was increased by 10.9%， and the data delay was reduced by 22.2%.

Multi-modal abstractive summarization is commonly based on the Sequence-to-Sequence （Seq2Seq） framework， and the objective function optimizes the model at the character level， which searches locally optimal results to generate words and ignores the global semantic information of the summary samples. It may cause a problem of semantic deviation between the summary and multimodal information， resulting in factual errors. In order to solve the above problems， a multi-modal summarization model based on semantic relevance analysis was proposed. Firstly， the summary generator based on Seq2Seq framework was trained to generate candidate summaries with semantic multiplicity. Secondly， a summary evaluator based on semantic relevance analysis was applied to learn the semantic differences among candidate summaries and the evaluation mode of ROUGE （Recall-Oriented Understudy for Gisting Evaluation） from a global perspective， so that the model could be optimized at the level of summary samples. Finally， the summary evaluator was used to carry out reference-free evaluation of the candidate summaries， making the finally selected summary sample as similar as possible to the source text in semantic space. Experiments on benchmark dataset MMSS show that the proposed model can improve the evaluation indexes of ROUGE-1， ROUGE-2 and ROUGE-L by 3.17， 1.21 and 2.24 percentage points respectively compared with the current optimal MPMSE （Multimodal Pointer-generator via Multimodal Selective Encoding） model.

The existing multi-tampering type image forgery detection algorithms using noise features often can not effectively detect the feature difference between tampered areas and non-tampered areas， especially for copy-move tampering type. To this end， a dual-stream image tampering forensics network fusing residual feedback and self-attention mechanism was proposed to detect tampering artifacts such as unnatural edges of RGB pixels and local noise inconsistence respectively through two streams. Firstly， in the encoder stage， multiple dual residual units integrating residual feedback were used to extract relevant tampering features to obtain coarse feature maps. Secondly， further feature reinforcement was performed on the coarse feature maps by the improved self-attention mechanism. Thirdly， the mutual corresponding shallow features of encoder and deep features of decoder were fused. Finally， the final features of tempering extracted by the two streams were fused in series， and then the pixel-level localization of the tampered area was realized through a special convolution operation. Experimental results show that the F1 score and Area Under Curve （AUC） value of the proposed network on COVERAGE dataset are better than those of the comparison networks. The F1 score of the proposed network is 9.8 and 7.7 percentage points higher than that of TED-Net （Two-stream Encoder-Decoder Network） on NIST16 and Columbia datasets， and the AUC increases by 1.1 and 6.5 percentage points， respectively. The proposed network achieves good results in copy-move tampering type detection， and is also suitable for other tampering type detection. At the same time， the proposed network can locate the tampered area at pixel level accurately， and its detection performance is superior to the comparison networks.

With the development of Artificial Intelligence （AI） technology， the Deep Neural Network （DNN） models have been applied to various mobile and edge devices widely. However， the model deployment becomes challenging and the popularization and application of the models are limited due to the facts that the computing power of edge devices is low， the memory capacity of edge devices is small， and the realization of model acceleration requires in-depth knowledge of edge device hardware. Therefore， a DNN acceleration and deployment method based on Tensor Virtual Machine （TVM） was presented to accelerate the Convolutional Neural Network （CNN） model on Field-Programmable Gate Array （FPGA）， and the feasibility of this method was verified in the application scenarios of distracted driving classification. Specifically， in the proposed method， the computational graph optimization method was utilized to reduce the memory access and computational overhead of the model， the model quantization method was used to reduce the model size， and the computational graph packing method was adopted to offload the convolution calculation to the FPGA in order to speed up the model inference. Compared with MPU （MicroProcessor Unit）， the proposed method can reduce the inference time of ResNet50 and ResNet18 on MPU+FPGA by 88.63% and 77.53% respectively. On AUC （American University in Cairo） dataset， compared to MPU， the top1 inference accuracies of the two models on MPU+FPGA are only reduced by 0.26 and 0.16 percentage points respectively. It can be seen that the proposed method can reduce the deployment difficulty of different models on FPGA.

Visual localization and mapping system is affected by dynamic objects in a dynamic environment， so that it has increase of localization and mapping errors and decrease of robustness. And motion segmentation of input images can significantly improve the performance of visual localization and mapping system in dynamic environment. Dynamic objects in dynamic environment can be divided into moving objects and potential moving objects. Current dynamic object recognition methods have problems of chaotic moving subjects and poor real-time performance. Therefore， motion segmentation strategies of visual localization and mapping system in dynamic environment were reviewed. Firstly， the strategies were divided into three types of methods according to preset conditions of the scene： methods based on static assumption of image subject， methods based on prior semantic knowledge and multi-sensor fusion methods without assumption. Then， these three types of methods were summarized， and their accuracy and real-time performance were analyzed. Finally， aiming at the difficulty of balancing accuracy and real-time performance of motion segmentation strategy of visual localization and mapping system in dynamic environment， development trends of the motion segmentation methods in dynamic environment were discussed and prospected.

Aiming at the problem that traditional defense schemes against Distributed Denial of Service （DDoS） attacks in Software Defined Network （SDN） tend to ignore the importance of reducing the workload of SDN， as well as do not consider the timeliness of attack mitigation， an efficient collaborative defense scheme against DDoS attacks in SDN was proposed. Firstly， the overhead of the control plane was reduced and the data plane’s resources were entirely used by offloading some of the defense tasks into the data plane. Then， if an anomaly was detected， eXpress Data Path （XDP） rules were generated to mitigate the attack promptly， and the statistical information of the data plane was handed over to the control plane to further detect and mitigate the attack， thereby improving the accuracy and further reducing the controller overhead. Finally， the rules of XDP were updated according to the anomaly source determined by the control plane. To validate the effectiveness of the proposed scheme， the Hyenae attack tool was used to generate three different types of attack data. Compared with the Support Vector Machine （SVM） scheme that relies on the control plane， the new architecture defense scheme， and the cross-plane collaborative defense scheme， the proposed scheme has the timeliness of defense improved by 33.33%， 28.57%， and 21.05%， respectively； the proposed scheme has the Central Processing Unit （CPU） consumption reduced by 33， 11， and 4 percentage points. Experimental results show that the proposed scheme can defend against DDoS attacks well and has a low performance overhead.

On the premise of ensuring the properties of nodes in the network， the purpose of attribute network representation learning is to learn the low-dimensional dense vector representation of nodes by combining structure and attribute information. In the existing attribute network representation learning methods， the learning of attribute information in the network is ignored， and the interaction of attribute information with the network topology is insufficient， so that the network structure and attribute information cannot be fused efficiently. In response to the above problems， a Dual auto-Encoder Network Representation Learning （DENRL） algorithm was proposed. Firstly， the high-order neighborhood information of nodes was captured through a multi-hop attention mechanism. Secondly， a low-pass Laplacian filter was designed to remove the high-frequency signals and iteratively obtain the attribute information of important neighbor nodes. Finally， an adaptive fusion module was constructed to increase the acquisition of important information through the consistency and difference constraints of the two kinds of information， and the encoder was trained by supervising the joint reconstruction loss function of the two auto-encoders. Experimental results on Cora， Citeseer， Pubmed and Wiki datasets show that DENRL algorithm has the highest clustering accuracy and the lowest algorithm running time on three citation network datasets compared with DeepWalk， ANRL （Attributed Network Representation Learning） and other algorithms， achieves these two indicators of 0.775 and 0.460 2 s respectively on Cora datasets， and has the highest link prediction precision on Cora and Citeseer datasets， reaching 0.961 and 0.970 respectively. It can be seen that the fusion and interactive learning of attribute and structure information can obtain stronger node representation capability.

Aiming at the problems that existing image deblurring algorithms suffer from diffusion and artifacts when dealing with edge loss and the use of full-frame deblurring in video processing does not meet real-time requirements， an Adaptive DeBlurring Generative Adversarial Network （ADBGAN）algorithm based on active discrimination mechanism was proposed. Firstly， an adaptive fuzzy discrimination mechanism was proposed， and an adaptive fuzzy processing network module was developed to make a priori judgment of fuzziness on the input image. When collecting the input， the blurring degree of the input image was judged in advance， and the input frame which was clear enough was eliminated to improve the running efficiency of the algorithm. Then， the incentive link of the attention mechanism was introduced in the process of fine feature extraction， so that weight normalization was carried out in the forward flow of feature extraction to improve the performance of the network to recover fine-grained features. Finally， the feature pyramid fine feature recovery structure was improved in the generator architecture， and a more lightweight feature fusion process was adopted to improve the running efficiency. In order to verify the effectiveness of the algorithm， detailed comparison experiments were conducted on the open source datasets GoPro and Kohler. Experimental results on GoPro dataset show that the visual fidelity of ADBGAN is 2.1 times that of Scale-Recurrent Network （SRN） algorithm， the Peak Signal-to-Noise Ratio （PSNR） of ADBGAN is improved by 0.762 dB compared with that of SRN algorithm， and ADBGAN has good image information recovery ability； in terms of video processing time，the actual processing time is reduced by 85.9% compared to SRN.The proposed algorithm can generate deblurred images with higher information quality efficiently.

Accurate prediction of taxi demands between urban regions can provide decision support information for taxi guidance and scheduling as well as passenger travel recommendation， so as to optimize the relation between taxi supply and demand. However， most of the existing models only focus on modeling and predicting the taxi demands within a region， do not consider enough the spatial-temporal correlation between regions， and pay less attention to the more fine-grained demand prediction between regions. To solve the above problems， a prediction model for taxi demands between urban regions — Origin-Destination fusion with Spatial-Temporal Network （ODSTN） model was proposed. In this model， complex spatial-temporal correlations between regions was captured from spatial dimensions of the regions and region pairs respectively and three temporal dimensions of recent， daily and weekly periods by using graph convolution and attention mechanism， and a new path perception fusion mechanism was designed to combine the multi-angle features and finally realize the taxi demand prediction between urban regions. Experiments were carried out on two real taxi order datasets in Chengdu and Manhattan. The results show that the Mean Absolute Error （MAE）， Root Mean Square Error （RMSE） and Mean Absolute Percentage Error （MAPE） of ODSTN model are 0.897 1， 3.527 4， 50.655 6% and 0.589 6， 1.163 8， 61.079 4%， respectively， indicating that ODSTN model has high accuracy in taxi demand prediction tasks.

The purpose of the task scheduling strategy for multiple data centers is to allocate computing tasks to different servers in each data center to improve the resource utilization and energy efficiency. Therefore， a deep reinforcement learning-based integrated scheduling strategy for multiple data center was proposed， which is divided into two stages： data center selection and task allocation within the data centers. In the multiple data centers selection stage， the computing power resources were integrated to improve the overall resource utilization. Firstly， a Deep Q Network （DQN） with Prioritized Experience Replay （PER-DQN） was used to obtain the communication paths to each data center in the network with data centers as nodes. Then， the resource usage cost and network communication cost were calculated， and the optimal data center was selected according to the principle that the sum of the two costs is minimum. In the task allocation stage， firstly， in the selected data center the computing tasks were divided and added to the scheduling queue according to the First-Come First-Served （FCFS） principle. Then， combining the computing device status and ambient temperature， the task allocation algorithm based on Double DQN （Double DQN） was used to obtain the optimal allocation strategy， thereby selecting the server to perform the computing task， avoiding the generation of hot spots and reducing the energy consumption of refrigeration equipment. Experimental results show that the average total cost of PER-DQN-based data center selection algorithm is reduced by 3.6% and 10.0% respectively compared to those of Computing Resource First （CRF） and Shortest Path First （SPF） path selection methods. Compared to Round Robin scheduling （RR） and Greedy scheduling （Greedy） algorithms， the Double DQN-based task deployment algorithm reduces the average Power Usage Effectiveness （PUE） by 2.5% and 1.7% respectively. It can be seen that the proposed strategy can reduce the total cost and data center energy consumption effectively， and realize the efficient operation of multiple data centers.

Internet of Vehicles （IoV） traffic monitoring requires the transmission， storage and analysis of private data of users， making the security guarantee of private data particularly crucial. However， traditional security solutions are often hard to guarantee real-time computing and data security at the same time. To address the above issue， security protocols， including two initialization protocols and a periodic reporting protocol， were designed， and a Software Guard Extensions （SGX）-based IoV traffic monitoring Secure Data Processing Framework （SDPF） was built. In SDPF， the trusted hardware was used to enable the plaintext computation of private data in Road Side Unit （RSU）， and efficient operation and privacy protection of the framework were ensured through security protocols and hybrid encryption scheme. Security analysis shows that SDPF is resistant to eavesdropping， tampering， replay， impersonation， rollback， and other attacks. Experiment results show that all computational operations of SDPF are at millisecond level， specifically， all data processing overhead of a single vehicle is less than 1 millisecond. Compared with PFCF （Privacy-preserving Fog Computing Framework for vehicular crowdsensing networks） based on fog computing and PPVF （Privacy-preserving Protocol for Vehicle Feedback in cloud-assisted Vehicular Ad hoc NETwork （VANET）） based on homomorphic encryption， SDPF has the security design more comprehensive： the message length of a single session is reduced by more than 90%， and the computational cost is reduced by at least 16.38%.

Control logic injection attack against Programmable Logic Controller （PLC） manipulate the physical process by tampering with the control program， thereby achieving the purpose of affecting the control process or destroying the physical facilities. Aiming at PLC control logic injection attacks， an intrusion detection method based on automatic whitelist rules generation was proposed， called PLCShield （Programmable Logic Controller Shield）. Based on the fact that PLC control program carries comprehensive and complete physical process control information， the proposed method mainly includes two stages： firstly， by analyzing the PLC program’s configuration file， instruction function， variable attribute， execution path and other information， the detection rules such as program attribute， address， value range and structure were extracted； secondly， combining actively requesting a “snapshot” of the PLC’s running and passively monitoring network traffic was used to obtain real-time information such as the current running status of PLC and the operation and status in the traffic， and the attack behavior was identified by comparing the obtained information with the detection rules. Four PLCs of different manufacturers and models were used as research cases to verify the feasibility of PLCShield. Experimental results show that the attack detection accuracy of the proposed method can reach more than 97.71%. The above prove that the proposed method is effective.

With the deepening of population aging， fall detection has become a key issue in the medical and health field. Concerning the low accuracy of fall detection algorithms in complex scenes， an improved fall detection model PDD-FCOS （PVT DRFPN DIoU-Fully Convolutional One-Stage object detection） was proposed. Pyramid Vision Transformer （PVT） was introduced into the backbone network of baseline FCOS algorithm to extract richer semantic information without increasing the amount of computation. In the feature information fusion stage， Double Refinement Feature Pyramid Networks （DRFPN） were inserted to learn the positions and other information of sampling points between feature maps more accurately， and more accurate semantic relationship between feature channels was captured by context information to improve the detection performance. In the training stage， the bounding box regression was carried out by the Distance Intersection Over Union （DIoU） loss. By optimizing the distance between the prediction box and the center point of the object box， the regression box was made to converge faster and more accurately， which improved the accuracy of the fall detection algorithm effectively. Experimental results show that on the open-source dataset Fall detection Database， the mean Average Precision （mAP） of the proposed model reaches 82.2%， which is improved by 6.4 percentage points compared with that of the baseline FCOS algorithm， and the proposed algorithm has accuracy improvement and better generalization ability compared with other state-of-the-art fall detection algorithms.

High Utility Itemsets Mining （HUIM） is able to mine the items with high significance from transaction database， thus helping users to make better decisions. In view of the fact that the application of intelligent optimization algorithms can significantly improve the mining efficiency of high utility itemsets in massive data， a survey of intelligent optimization algorithm-based HUIM methods was presented. Firstly， detailed analysis and summary of the intelligent optimization algorithm-based HUIM methods were performed from three aspects： swarm intelligence optimization-based， evolution-based and other intelligent optimization algorithms-based methods. Meanwhile， the Particle Swarm Optimization （PSO）-based HUIM methods were sorted out in detail from the aspect of particle update methods， including traditional update strategy-based， sigmoid function-based， greedy-based， roulette-based and ensemble-based methods. Additionally， the swarm intelligence optimization algorithm-based HUIM methods were compared and analyzed from the perspectives of population update methods， comparison algorithms， parameter settings， advantages and disadvantages， etc. Next， the evolution-based HUIM methods were summarized and outlined in terms of both genetic and bionic aspects. Finally， the next research directions were proposed for the problems of the existing intelligent optimization algorithm-based HUIM methods.

The traditional classifiers are difficult to cope with the challenges of complex types of data streams with concept drift， and the obtained classification results are often unsatisfactory. Aiming at the methods of dealing with concept drift in different types of data streams， classification methods for complex data streams with concept drift were summarized from four aspects： imbalance， concept evolution， multi-label and noise-containing. Firstly， classification methods of four aspects were introduced and analyzed： block-based and online-based learning approaches for classifying imbalanced concept drift data streams， clustering-based and model-based learning approaches for classifying concept evolution concept drift data streams， problem transformation-based and algorithm adaptation-based learning approaches for classifying multi-label concept drift data streams and noisy concept drift data streams. Then， the experimental results and performance metrics of the mentioned concept drift complex data stream classification methods were compared and analyzed in detail. Finally， the shortcomings of the existing methods and the next research directions were given.

To remove Gibbs artifacts in Magnetic Resonance Imaging （MRI）， a Self-attention connection UNet based on Self-Distillation training （SD-SacUNet） algorithm was proposed. In order to reduce the semantic gap between the encoding and decoding features at both ends of the skip connection in the UNet framework and help to capture the location information of artifacts， the output features of each down-sampling layer at the UNet encoding end was input to the corresponding self-attention connection module for the calculation of the self-attention mechanism， then they were fused with the decoding features to participate in the reconstruction of the features. Self-distillation training was performed on the network decoding end， by establishing the loss function between the deep and shallow features， the feature information of the deep reconstruction network was used to guide the training of the shallow network， and at the same time， the entire network was optimized to improve the level of image reconstruction quality. The performance of SD-SacUNet algorithm was evaluated on the public MRI dataset CC359， with the Peak Signal-to-Noise Ratio （PSNR） of 30.261 dB and the Structure Similarity Index Measure （SSIM） of 0.917 9. Compared with GRACNN （Gibbs-Ringing Artifact reduction using Convolutional Neural Network）， the proposed algorithm had the PSNR increased by 0.77 dB and SSIM increased by 0.018 3； compared with SwinIR （Image Restoration using Swin Transformer）， the proposed algorithm had the PSNR increased by 0.14 dB and SSIM increased by 0.003 3. Experimental results show that SD-SacUNet algorithm improves the image reconstruction performance of MRI with Gibbs artifacts removal and has potential application values.

To solve the problem of limited computing resources and storage space of edge nodes in the Edge Computing （EC） network， an Edge Computing and Service Offloading （ECSO） algorithm based on improved Deep Reinforcement Learning （DRL） was proposed to reduce node processing latency and improve service performance. Specifically， the problem of edge node service offloading was formulated as a resource-constrained Markov Decision Process （MDP）. Due to the difficulty of predicting the request state transfer probability of the edge node accurately， DRL algorithm was used to solve the problem. Considering that the state action space of edge node for caching services is too large， by defining new action behaviors to replace the original actions， the optimal action set was obtained according to the proposed action selection algorithm， so that the process of calculating the action behavior reward was improved， thereby reducing the size of the action space greatly， and improving the training efficiency and reward of the algorithm. Simulation results show that compared with the original Deep Q-Network （DQN） algorithm， Proximal Policy Optimization （PPO） algorithm and traditional Most Popular （MP） algorithm， the total reward value of the proposed ECSO algorithm is increased by 7.0%， 12.7% and 65.6%， respectively， and the latency of edge node service offloading is reduced by 13.0%， 18.8% and 66.4%， respectively， which verifies the effectiveness of the proposed ECSO algorithm and shows that the ECSO can effectively improve the offloading performance of edge computing services.