Image fusion technology plays an important role in computer-aided diagnosis. Detail extraction and energy preservation are two key issues in image fusion, and the traditional fusion methods address them simultaneously by designing the fusion method. However, it tends to cause information loss or insufficient energy preservation. In view of this, a fusion method was proposed to solve the problems of detail extraction and energy preservation separately. The first part of the method aimed at detail extraction. Firstly, the Non-Subsampled Shearlet Transform (NSST) was used to divide the source image into low-frequency and high-frequency subbands. Then, an improved energy-based fusion rule was used to fuse the low-frequency subbands, and an strategy based on the intuitionistic fuzzy set theory was proposed for the fusion of the high-frequency subbands. Finally, the inverse NSST was employed to reconstruct the image. In the second part, an intensity enhancement method was proposed for energy preservation. The proposed method was verified on 43 groups of images and compared with other eight fusion methods such as Principal Component Analysis (PCA) and Local Laplacian Filtering (LLF). The fusion results on two different categories of medical image fusion (Magnetic Resonance Imaging (MRI) and Positron Emission computed Tomography (PET), MRI and Single-Photon Emission Computed Tomography (SPECT)) show that the proposed method can obtain more competitive performance on both visual quality and objective evaluation indicators including Mutual Information (MI), Spatial Frequency (SF), Q value, Average Gradient (AG), Entropy of Information (EI), and Standard Deviation (SD), and can improve the quality of medical image fusion.

Aspect-level sentiment analysis aims to classify the sentiment of text in different aspects. In the aspect-level sentiment analysis of long text, the existing aspect-level sentiment analysis algorithms do not fully extract the features of aspect related information in the long text due to the redundancy and noise problems, leading to low classification accuracy. On the datasets with coarse and fine aspects, existing solutions do not take advantage of the information in the coarse aspect. In view of the above problems, an algorithm named TFN+BERT-Pair-ATT was proposed based on text filtering and improved Bidirectional Encoder Representation from Transformers (BERT). First, the Text Filter Network (TFN) based on Long Short-Term Memory (LSTM) neural network and attention mechanism was used to directly select part sentences related to the coarse aspect from the long text. Next, the related sentences were associated with others in order, and after combining with fine aspects, the sentences were input into the BERT-Pair-ATT, which is with the attention layer added to the BERT, for feature extraction. Finally, the sentiment classification was performed by using Softmax. Compared with the classical Convolutional Neural Network (CNN) based models such as Gated Convolutional network with Aspect Embedding (GCAE) and LSTM based model Interactive Attention Network (IAN), the proposed algorithm improves the related evaluation index by 3.66% and 4.59% respectively on the validation set, and improves the evaluation index by 0.58% compared with original BERT. Results show that the algorithm based on text filtering and improved BERT has great value in the aspect-level sentiment analysis task of long text.

To overcome the randomness of RRT-Connect and slow convergence of B-RRT ^*(asymptotically-optimal Bidirectional Rapidly-exploring Random Tree) in path generation, an efficient path planning algorithm based on B-RRT ^*, abbreviated as EB-RRT ^*, was proposed. Firstly, an intelligent sampling function was intriduced to achieve more directional expansion of random tree, which could improve the smoothness of path and reduce the seek time. A rapidly-exploring strategy was also added in EB-RRT ^* by which RRT-Connect exploration mode was adopted to ensure rapidly expanding in the free space and improved asymptotically-optimal Rapidly-exploring Random Tree (RRT ^*) algorithm was adopted to prevent trapped in local optimum in the obstacle space. Finally, EB-RRT ^* algorithm was compared with Rapidly-exploring Random Tree (RRT), RRT-Connect, RRT ^* and B-RRT ^* algorithms. The simulation results show that the improved algorithm is superior to other algorithms in the efficiency and smoothness of path planning. It reduced the path planning time by 68.3% and the number of iterations by 48.6% compared with B-RRT ^* algorithm.

The static gesture recognition speed is higher than that of dynamic gesture recognition for RGB-D (RGB-Depth) data, but redundancy gestures and repeated gestures lead to low recognition accuracy. In order to solve the problem, a static gesture recognition method based on locking mechanism was proposed. First, RGB data flow and the Depth data stream were obtained through Kinect equipment, then two kinds of data flow were integrated into human body skeleton data flow. Second, the locking mechanism was used to identify static gestures, and comparison and calculation were done with the established bone point feature model gesture library before. Finally, an "advanced programmers road" brain-training Web game was designed for application and experiment. In the experiments of six different movement gestures, compared with the static gesture recognition method, the average recognition accuracy of the proposed method was increased by 14.4%; compared with the dynamic gesture recognition method, the gesture recognition speed of the proposed method was improved by 14%. The experimental results show that the proposed method keeps the high speed of static recognition method, realizes the real-time recognition; and also improves the identification accuracy through eliminating redundant repeated gestures.

Focused on the issue that interference would exist in the analysis and prediction of airport energy data because of the outliers, a prediction model based on improved Fuzzy Support Vector Regression (FSVR) was established for the demand of airport energy. Firstly, a fuzzy statistical method was selected to make an analysis on test sample sets, parameters and the outputs of models, and a basic membership function form consistent with the data distribution would be derived from this analysis. Secondly, relearning of membership function would be performed with respect to expert experiences, then the parameter values a and b of the normal membership function, the boundary parameter values of semi-trapezoid membership function and the parameter values p and d of triangular membership function would gradually be refined and improved, so as to eliminate or reduce the outliers which were not conducive to data mining and reserved the key points. Finally, combined with Support Vector Regression (SVR) algorithm, a prediction model was established and its feasibility was verified subsequently. The experimental result shows that, compared with Back Propagation (BP) neural network, the prediction accuracy of the FSVR increases 2.66% and the recognition rate of outliers increases 3.72%.

Due to using a single point in time for the processing unit, current network alerts information fusion methods cannot adapt to the network attacks with high concealment and long duration. Aiming at this problem, a network alerts depth information fusion method based on time confrontation was proposed. In view of multi-source heterogeneous alerts data flow, firstly, the alerts were collected and saved in a long time window. Then the alerts were clustered using a clustering algorithm based on sliding window. Finally, the alerts were fused by introducing window attenuation factor. The experimental results on real data set show that, compared with Basic-DS and EWDS (Exponential Weight DS), the proposed method has higher True Positive Rate (TPR) and False Positive Rate (FPR) as well as lower Data to Information Rate (DIR) because of longer time window. Actual test and theoretical analysis show that the proposed method is more effective on detecting network attacks, and can satisfy real-time processing with less time delay.

By analyzing and comparing the existing network security situation evaluation methods, it is found that they can not accurately reflect the features of large-scale, coordination, multi-stage gradually shown by network attack behaviors. Therefore, a network security situation evaluation method based on attack pattern recognition was proposed. Firstly, the causal analysis of alarm data in the network was made, and the attack intention and the current attack phase were recognized. Secondly, the situation evaluation based on the attack phase was realized. Lastly the State Transition Diagram (STG) of attack phase was created to realize the forecast of network security situation by combining with vulnerability and configuration information of host. A simulation experiment for the proposed network security situation evaluation model was performed by network examples. With the deepening of the attack phase, the value of network security situation would increase. The experimental results show that the proposed method is more accurate in reflecting the truth of attack, and the method does not need training on the historical sequence, so the method is more effective in situation forecasting.

As the session controlling protocol of application-layer, SIP has the features of simple, expansibility and dilatancibility. At the basis of simple introduction of SIP protocol, the JAIN SIP exploring construction for the fulfillment of SIP communication of SUN Co was discussed in detail. To use Java language and take JAIN SIP as the core, all kinds of communication entity basic method in the fulfillment of SIP communication were described and simple model for SIP communication was built.