In view of the increasing demand for user experience quality， the difficulty in obtaining link status caused by highly mobile vehicles， and the time-varying problem of heterogeneous edge nodes providing resources to vehicles in Vehicle Edge Computing （VEC）， a VEC scheme based on Joint Task Offloading and Resource Optimization （JTO-RO） was developed. Firstly， without loss of the generality， a Vehicle-to-Infrastructure （V2I） transmission model was proposed by considering the intra-edge and inter-edge interference comprehensively. In the model， by introducing Non-Orthogonal Multiple Access （NOMA） technology， edge nodes did not rely on link status information and improved the channel capacity at the same time. Secondly， in order to enhance performance and efficiency of the system， a Multi-Agent Twin Delayed Deep Deterministic policy gradient （MATD3） algorithm was designed to formulate task offloading strategies， which were able to be adjusted dynamically through interactive learning with the environment. Thirdly， the synergies of the two strategies were considered jointly， and an optimization scheme was formulated with the goal of maximizing the task service ratio to meet the increasing user experience quality requirements. Finally， simulation was carried out on a real vehicle trajectory dataset. The results show that compared with three current representative schemes （the schemes using Random Offloading （RO） algorithm， D4PG （Distributed Distributional Deep Deterministic Policy Gradient） algorithm， and MADDPG （Multi-Agent Deep Deterministic Policy Gradient） algorithm as task offloading algorithms as task offloading algorithm， respectively）， the proposed scheme has the average service ratio improved by more than 20%， 10%， and 29%， respectively， in three scenarios （normal scenario， task-intensive scenario and delay-sensitive scenario）， verifying the advantages and effectiveness of the scheme.

In view of the problems of current tile defect detection mainly relying on manual detection， such as strong subjectivity， low efficiency， and high labor intensity， an improved lightweight algorithm for detecting small defects in large-format ceramic tile images based on YOLOv8 was proposed. Firstly， the high-resolution large-format image was cropped， and HorBlock was introduced into the backbone network to enhance model’s capture capability. Secondly， Large Separable Kernel Attention （LSKA） was incorporated to improve C2f for improving the detection performance of the model and model’s feature extraction capability was enhanced by introducing SA （Shuffle Attention）. Finally， Omni-Dimensional Dynamic Convolution （ODConv） was introduced to further enhance model’s capability to handle with small defects. Experimental results on Alibaba Tianchi tile defect detection dataset show that the improved model not only has lower parameters than the original YOLOv8n， but also has an increase of 8.2 percentage points in mAP@0.5 and an increase of 7 percentage points in F1 score compared to the original YOLOv8n. It can be seen that the improved model can identify and process small surface defects of large-format tiles more accurately， and improve the detection effect significantly while maintaining lightweight.

Parallel imaging techniques can help solving problems of radiofrequency energy deposition and image inhomogeneity， reducing scan time， lowering motion artifacts， and accelerating data acquisition in ultra-high field Magnetic Resonance Imaging （MRI）. To enhance feature extraction ability to MRI complex-valued data and reduce wrap-around artifacts caused by under-sampling in parallel imaging， a Residual Complex convolution scan-specific Robust Artificial-neural-networks for K-space Interpolation （RCRAKI） was proposed. In the algorithm， the raw under-sampled MRI scan data was taken as input， and the advantages of both linear and nonlinear reconstruction methods were combined with a residual structure. In the residual connection part， convolution was used to create a linear reconstruction baseline， while multiple layers of complex convolution were utilized in the main path to compensate for baseline defects， ultimately reconstructing Magnetic Resonance （MR） images with fewer artifacts. Experiments were conducted on data acquired from a 7T ultra-high field MR device developed by the Institute of Energy of Hefei Comprehensive National Science Center， and RCRAKI was compared with residual scan-specific Robust Artificial-neural-networks for K-space Interpolation （rRAKI） under a sampling rate of 40 Automatic Calibration Signals （ACSs） and 8 speedup ratio for mouse imaging quality across different anatomical planes. Experimental results show that in sagittal plane， the proposed algorithm has the Normalized Root Mean Squared Error （NRMSE） decreased by 59.74%， the Structural SIMilarity （SSIM） increased by 0.45%， and the Peak Signal-to-Noise Ratio （PSNR） increased by 13.04%； in axial plane， the proposed algorithm has the NRMSE decreased by 7.97%， the SSIM improved slightly （by 0.005%）， and the PSNR increased by 1.09%； in coronal plane， the proposed algorithm has the NRMSE decreased by 35.03%， the PSNR increased by 5.60%， and the SSIM increased by 0.98%. It can be seen that RCRAKI performs well on all the different anatomical planes of MRI data， can reduce the influence of noise amplification at high speedup ratio， and reconstruct MR images with clearer details.

In recent years， Generative Adversarial Networks （GANs） have been widely used for data augmentation， which can solve the problem of insufficient training samples effectively and has important research significance for model training. However， the existing GAN models for data augmentation have problems such as high requirements for datasets and unstable model convergence， which can lead to distortion and deformation of the generated images. Therefore， a lightweight GAN based on dynamic-upsample — DU-FastGAN （Dynamic-Upsample-FastGAN） was proposed for data augmentation. Firstly， a generator was constructed through a dynamic-upsample module， which enables the generator to use upsampling methods of different granularities based on the size of the current feature map， thereby reconstructing textures， and enhancing overall structure and local detail quality of the synthesis. Secondly， in order to enable the model to better obtain global information flow of images， a weight information skip connection module was proposed to reduce the disturbance of convolution and pooling operations on features， thereby improving the model’s learning ability for different features， and making details of the generated images more realistic. Finally， a feature loss function was given to improve the quality of the model generation by calculating relative distance between the corresponding feature maps during the sampling process. Experimental results show that compared with methods such as FastGAN， MixDL （Mixup-based Distance Learning）， and RCL-master （Reverse Contrastive Learning-master）， DU-FastGAN achieves a maximum reduction of 23.47% in FID （Fréchet Inception Distance） on 10 small datasets， thereby reducing distortion and deformation problems in the generated images effectively， and improving the quality of the generated images. At the same time， DU-FastGAN achieves lightweight overhead with model training time within 600 min.

In view of the problem that the existing exam paper generation technologies pay too much attention to the difficulty of generated exam papers， while ignoring other related objectives， such as quality， score distribution， and skill coverage， a multi-objective exam paper generation method guided by reinforcement learning and matrix completion was proposed to optimize the specific objectives in the field of exam paper generation. Firstly， deep knowledge tracking method was used to model the interaction information among students and response logs in order to obtain the skill proficiency of the student group. Secondly， matrix factorization and matrix completion methods were used to predict the scores of students' undone exercises. Finally， based on the multi-objective exam paper generation strategy， in order to improve the Q network update efficiency， an Exam Q-Network function approximator was designed to select the appropriate question set automatically for update of the exam paper composition. Experimental results show that compared with the models such as DEGA （Diseased-Enhanced Genetic Algorithm） and SSA-GA （Sparrow Search Algorithm - Genetic Algorithm）， it is verified that the proposed model has significant effect in solving multiple dilemmas of exam paper generation scenarios in terms of three indicators — difficulty， rationality and accuracy. The effect of verifying the models mentioned in the solution of the test papers is significantly effective.

In the existing optimization of Electric power material Vehicle Routing Problem （EVRP）， the objective function is relatively single， the constraints are not comprehensive enough， and the traditional solution algorithms are not efficient. Therefore， a multi-objective routing optimization model and solution algorithm for electric power material distribution based on Deep Reinforcement Learning （DRL） was proposed. Firstly， the electric power material distribution area constraints such as the distribution of gas stations and the fuel consumption of material transportation vehicles were fully considered to establish a multi-objective power material distribution model with the objectives of the minimum total length of the power material distribution routings， the lowest cost， and the highest material demand point satisfaction. Secondly， a power material distribution routing optimization algorithm DRL-EVRP was designed on the basis of Deep Reinforcement Learning （DRL） to solve the proposed model. In the algorithm， the improved Pointer Network （Ptr-Net） and the Q-learning algorithm were combined to form the Deep Q-Network （DQN）， which was used to take the sum of the negative value of the cumulative incremental routing length and the satisfaction as the reward function. After DRL-EVRP algorithm was trained and learned， it can be directly used for the planning of electric power material distribution routings. Simulation results show that the total length of the power material distribution routing solved by DRL-EVRP algorithm is shorter than those solved by the Extended Clarke and Wright （ECW） saving algorithm and Simulated Annealing （SA） algorithm， and the calculation time of the proposed algorithm is within an acceptable range. Therefore， the power material distribution routing can be optimized more efficiently and quickly by the proposed algorithm.

The single frequency modulation or phase modulation waveform based on chaotic sequence has low waveform complexity, which limits predictive probability of chaotic signal, radar intercept probability and anti-interference performance. In order to solve the problems, joint frequency-phase modulation based on chaotic sequence in radar waveform was proposed. Firstly, the radar signal was carried out for the chaotic frequency encoding, which was that a pulse was divided into a series of sub-pulses and different frequency modulation was carried out for different sub-pulses. At the same time, in each frequency encoding sub-pulse, the random initial phase was used in each cycle of waveform. The simulation results show that the maximum value of autocorrelation sidelobe peak of joint frequency-phase modulation based on chaotic radar signal achieved -24.71 dB. Compared with the frequency modulation or phase modulation based on chaotic signal, the correlation performance of the proposed joint frequency-phase modulation has improved. The experimental results show that, the joint frequency-phase modulation chaotic radar waveform combines the advantages of phase modulation and frequency modulation and is an ideal detection signal with the flat power spectrum characteristic of phase modulation and anti-noise-interference ability of frequency modulation.

The current method of image classification which uses the Speed Up Robust Feature (SURF) is low in efficiency and accuracy. To overcome these shortages, this paper proposed an approach for image classification which uses the statistical features of the SURF set. This approach took all dimensions and scale information of the SURF as independent random variables, and split the data with the sign of Laplace response. Firstly, the SURF vector set of the image was got. Then the feature vector was constructed with the first absolute order central moments and weighted first absolute order central moments of each dimision. Finally, the Support Vector Machine (SVM) accomplished the image classification process with this vector. The experimental results show that the precision of this approach is better than that of the methods of SURF histogram and 3-channel-Gabor texture features by increases of 17.6% and 5.4% respectively. By combining this approach with the HSV histogram, a high-level feature fusion method was got, and good classification performance was obtained. Compared with the fused method of the SURF histogram and HSV histogram, the fused method of 3-channel-Gabor texture features and HSV histogram, and the multiple-instance-learning method based on the model of Bag of Visual Word (BoVW), the fused method of this approach and HSV histogram has better precision with the increases of 5.2%, 6.8% and 3.2% respectively.

This article focused on the mobile sink scheduling problem in Wireless Sensor Networks (WSN). A mobile single-sink scheduling algorithm in wireless sensor networks was proposed based on Linear Programming (LP). Firstly, the problem was mathematically modeled and formulated in time domain, and the problem was re-formulated from time to space domain to reduce the complexity. Then a polynomial-time optimal algorithm was proposed based on linear programming. The simulations confirm the efficiency of the algorithm and the results show that the algorithm can significantly improve the network lifetime of wireless sensor networks.

To deal with the poor performance of word sense disambiguation in parsing, a Chinese phrase parsing approach was proposed based on disambiguation of Chinese part of speech. First, it expanded part of speech of TongYiCi CiLin and then substituted the original words in the training set and test set with semantics codes. In this process, it used part of speech of word for word sense disambiguation. The experimental results on Penn Chinese TreeBank (CTB) show that the proposed method achieves precision rate of 80.30%, recall rate of 78.12%, and F-measure of 79.19%. Relative to the no disambiguation system, the presented approach can effectively improve the performance of phrase parsing.

Bursty topics detection on microblog is an import branch of online public opinion analysis, and has attracted much attention from international scholars. In this paper, a new approach of calculating users' influence was proposed based on the analysis of users' behavior characteristics. Combining the user influence with text features and propagation features, this paper defined a concept named Bursty which is used to judge if a word was a burst word. Being judged by Bursty, burst words can be extracted from microblog corpus. Hierarchical clustering algorithm was introduced to cluster the burst words and chose appropriate burst word clusters to describe bursty topics on microblog in order to realize bursty topics detection on microblog. In experiments, the precision, recall and F-measure reached 63.64%,87.5% and 74% respectively. The method is proved effective on bursty topic detection based on mass microblog data.

According to the shape features of mechanical design images, an asynchronous retrieval method based on weighted distance was proposed. The algorithm firstly got preliminary results from the image database by using the circumcircle distance feature, and then calculated the weighted distances between the input image and the preliminary results, by considering both the formal output positions and the Hu invariant moments feature. The experiments show that compared with the traditional methods, the proposed method gets higher precision and recall ratio.

For improving the real-time performance of recovery algorithm in Compressed Sensing (CS) of Wireless Sensor Networks (WSN) data, a quadratic programming based network data recovery algorithm was proposed in this paper. The CS recovery was transformed to bound-constrained quadratic programming, and then the network data was recovered by solving the quadratic programming problem based on the Armijo rule. The analysis and experimental results demonstrate that the proposed algorithm can significantly reduce the complexity and ensure the accuracy of recovery, thus improving the real-time performance of data recovery in WSN.

The basic ray-based method is time consuming and only uses the information of triangle facets. An extended ray-based method was proposed based on the principle of non-intersecting pencil of planes. The key points of this method were as follows: firstly, a group of rays was scattered evenly from the center of the 3D model to intersect with triangle facets, and the non-intersecting pencil of planes determined by the rays was used to get the intersection points; secondly, the retrieval model was established to improve the 3D model retrieval effectiveness, according to the distances from the center to those intersection points and the vertices of the 3D model. Applying this method on ten categories of 3D models in PSB (Princeton Shape Benchmark), the results show that this approach not only reduces the processing time, but improves the retrieval accuracy.

Concerning that the edges of the endocardium and epicardium of the left ventricle in the cardiac Magnetic Resonance Imaging (MRI) images have different directions, a new directional active contour model in curve evolution framework was proposed for segmentation of endocardium and epicardium of the left ventricle. The curve evolution equation included a hybrid geometric flow with edge and region gray characteristics that were obtained from the image itself. The edge-based term in the geometric flow borrowed from extended Dynamic Directional Gradient Vector Flow (DDGVF) with fast marching method was utilized to guide the curve evolution towards the object boundaries with different direction. The region-based term borrowed from Chan-Vese (CV) model was utilized to prevent the curve from leakage under the influence of other edge components. The final curve evolution equation was dealt with level set method. The experimental results for gray and cardiac MRI images show that the proposed method can get better segmentation effects. It has certain application value for realizing myocardium auto-segmentation, evaluation and analysis of heart function based on cardiac MRI images.

A query algorithm of semantic relativity was proposed in this paper, with regard to many meaningless nodes contained in the present results of XML keywords retrieval. Based on the characteristics of semi-structure and self-description of XML files, the concept of Smallest Lowest Entity Sub-Tree (SLEST), in which only physical connection exists between keywords, was put forward by making full use of semantic correlation between nodes. Based on Smallest Interrelated Entity Sub-Tree (SIEST), an algorithm, in which the result was represented by SLEST and SIEST instead of Smallest Lowest Common Ancestor (SLCA), was proposed to capture the IDREF relation between keywords. The result shows that the algorithm proposed in this paper can increase the precision of XML keyword retrieval.

A storing systematic configuration of spatial metadata based on XDR Schema was proposed and a XML data reduced schema was created. The spatial metadata expressed by XML was mapped to SQL Server 2000 RDBMS.  The annotated XDR schema corresponded with XML view, so we could query database using annotated XDR schema and get result in XML form.

An improved genetic algorithm,evolving algorithm GASA was proposed,in which genetic algorithm was combined with simulated annealing algorithm. It avoided the premature convergence problem existed in Genetic Algorithm by useing Boltamann,and enhanced the global convergence. Genetic operators was redesigned, such as selection operator, cross operator and variation operator,on genetic algorithm. New cross operator and variation operator was proposed, which could dynamically regulate genetic operator according to evolving situation of groups. The algorithm was used in the best path problem of emergency decision support system,and it is proved to be reasonable and efficient.