Due to the high similarity of gray values among liver and adjacent organs in Computed Tomography （CT） and Magnetic Resonance Imaging （MRI） images， a 3D liver image segmentation method based on multi-scale feature fusion and grid attention mechanism， namely MAGNet （Multi-scale feature fusion And Grid attention mechanism Network）， was proposed to segment liver automatically and accurately. Firstly， high-level features and low-level features were connected by the attention-guided concatenation module to extract important context information， and the grid attention mechanism was introduced in the attention-guided concatenation module to focus on the segmentation region of interest. Then， the multi-scale feature fusion module was formed by the layered connection in a single feature map according to the number of channels， and this module was used to replace the basic convolutional block to obtain multi-scale semantic information. Finally， the deep supervision mechanism was utilized to solve the problems of vanishing gradient， exploding gradient and slow convergence. Experimental results show that on 3DIRCADb dataset， compared with the U³-Net+DC method， MAGNet improves the Dice Similarity Coefficient （DSC） metric by 0.10 percentage points and reduces the Relative Volume Difference （RVD） metric by 1.97 percentage points； on Sliver07 dataset， compared with the CANet method， MAGNet improves the DSC metrics by 0.30 percentage points， reduces Volumetric Overlap Error （VOE） metrics by 0.68 percentage points， and reduces the Average Symmetric Surface Distance （ASD） and Root Mean Square Symmetric Surface Distance （RMSD） metrics 0.03 mm and 0.22 mm respectively； on the liver MRI dataset of a hospital， MAGNet also has good results on all metrics. Besides， MAGNet was applied to a mixed dataset of 3DIRCADb dataset and the hospital liver MRI dataset above， and a competitive segmentation result was also achieved.

Considering the difficulty of modeling the interaction between labels and solidification of global label relationship in multi-label image classification tasks， a new Multiple-Label image classification method based on Global and Local Label Relationship （ML-GLLR） was proposed by combining self-attention mechanism and Knowledge Distillation （KD） method. Firstly， Convolutional Neural Network （CNN）， semantic module and Dual Layer Self-Attention （DLSA） module were used by the Local Label Relationship （LLR） model to model local label relationship. Then， the KD method was used to make LLR learn global label relationship. The experimental results on the public datasets of MicroSoft Common Objects in COntext （MSCOCO） 2014 and PASCAL VOC challenge 2007 （VOC2007） show that， LLR improves the mean Average Precision （mAP） by 0.8 percentage points and 0.6 percentage points compared with Multiple Label classification based on Graph Convolutional Network （ML-GCN） respectively， and the proposed ML-GLLR increases the mAP by 0.2 percentage points and 1.3 percentage points compared with LLR. Experimental results show that， the proposed ML-GLLR can not only model the interaction between labels， but also avoid the problem of global label relationship solidification.

Session-based recommendation aims to predict the user’s next click behavior based on the click sequence information of the current user’s anonymous session. Most of the existing methods realize recommendations by modeling the item information of the user’s session click sequence and learning the vector representation of the items. As a kind of coarse-grained information， the item category information can aggregate the items and can be used as an important supplement to the item information. Based on this， a Session-based Recommendation model of Multi-granular Graph Neural Network （SRMGNN） was proposed. Firstly， the embedded vector representations of items and item categories in the session sequence were obtained by using the Graph Neural Network （GNN）， and the attention information of users was captured by using the attention network. Then， the items and item category information given by different weight values of attention were fused and input into the Gated Recurrent Unit （GRU）. Finally， through GRU， the item time sequence information of the session sequence was learned， and the recommendation list was given. Experiments performed on the public Yoochoose dataset and Diginetica dataset verify the advantages of the proposed model with the addition of item category information， and show that the model has better effect compared with all the eight models such as Short-Term Attention/Memory Priority （STAMP）， Neural Attentive session-based RecomMendation （NARM）， GRU4REC on the evaluation indices Precision@20 and Mean Reciprocal Rank （MRR）@20.

To realize 10Gbps level wireless access under the condition of terahertz (THz) carrier frequency, a new Medium Access Control (MAC) protocol for THz ultra-high data-rate wireless networks, MAC-T (Medium Access Control for THz) was proposed in this paper. In MAC-T, a new TDMA (Time Division Multiple Access)+CSMA (Carrier Sense Multiple Access) adaptive hybrid MAC access control mechanism and a new superframe structure were designed. Moreover, some key parameters corresponding to terahertz communications were defined. Therefore, MAC-T could make the maximum data transmission rate reach up to 10Gbps or higher. The theoretical analysis and simulation results show that MAC-T can operate normally in terahertz networks and the data rate can reach up to 18.3Gbps which is 2.16 times 5.78Gbps that IEEE 802.15.3c can achieve. Meanwhile, the average access delay of MAC-T is 0.0044s which improves about 42.1% compared with that of IEEE 802.15.3 which is 0.0076s. Thus, MAC-T can provide significant support in the research and application of terahertz ultra-high data-rate wireless networks.

In order to recognize the curve whose feature points are the same but the curvature between the feature points is different, a new method for representing and recognizing the contour curve was proposed. First, feature points of the contour were extracted for the rough matching; then the sampling points of the sub-curve were obtained based on the precision requirement using the given minimal area threshold. A new recognition vector of sample points was defined, and a novel recognition vector matrix was constructed based on the recognition vector of sample points; last the similarity of the corresponding sub-curves was calculated by comparing the recognition vector matrix. The curve was recognized by recognizing their each sub-curve. The matching method was a process from simple to complex, thus many redundancies calculations were avoided. The experimental results show the proposed algorithm is efficient and feasible.