Aiming at the problems of blurred target boundaries and low image segmentation efficiency in the image segmentation process of ancient murals, a multi-class image segmentation model fused with a lightweight convolutional neural network named MC-DM (Multi-Class DeepLabV3+MobileNetV2 (Mobile Networks Vision 2)) was proposed. In the model, DeepLabV3+ architecture and MobileNetV2 network were combined together, and the unique spatial pyramid structure of DeepLabV3+ was utilized to perform multi-scale fusion of the convolutional features of the mural to reduce the loss of image details during the mural segmentation. First of all, the features of the input image were extracted by MobileNetV2 to ensure the accurate extraction of image information and reduce the time consumption at the same time. Secondly, the image features were processed through the dilated convolution, so that the receptive field was expanded, and more semantic information was obtained without changing the number of parameters. Finally, the bilinear interpolation method was utilized to up-sample the output feature image to obtain a pixel-level prediction segmentation map, so that the accuracy of image segmentation was ensured to the greatest extent. In the JetBrains PyCharm Community Edition 2019 environment, a dataset made of 1 000 mural scanning pictures was used for testing. Experimental results showed that the MC-DM model had a 1% improvement in training accuracy compared with the traditional SegNet (Segment Network)-based image segmentation model, and had a 2% improvement in accuracy compared with the image segmentation model based on PSPNet (Pyramid Scene Parsing Network), and the Peak Signal-to-Noise Ratio (PSNR) of the MC-DM model was 3 to 8 dB higher than those of the experimental comparison models on average, which verified the effectiveness of the model in the field of mural segmentation. The proposed model provides a new idea for the segmentation of ancient mural images.

In order to use computer vision technology to accurately detect the fall of the elderly, aiming at the incompleteness of existing fall detection algorithms caused by artificial designing of features and the problems in the fall detection process such as the difficulty of separating foreground and background, the confusion of objects, the loss of moving objects, and the low accuracy of fall detection, a deep learning fall detection algorithm with the fusion of human motion information was proposed to detect the fall state of human body. Firstly, foreground and background were separated by the improved YOLOv3 network, and human object was marked by minimum bounding rectangle according to the detection results of YOLOv3 network. Then, by analyzing the motion features in the process of human fall, the motion features of human body were vectorized and transformed into the motion weight information between 0 and 1 through the Sigmoid activation function. Finally, in order to classify human falls, the motion features and the features extracted by Convolutional Neural Network (CNN) were spliced and fused through the fully connected layer. The proposed fall detection algorithm was compared with human object detection algorithms such as background difference, Gaussian mixture, VIBE (VIsual Background Extractor), Histogram of Oriented Gradient (HOG) and human fall judgment schemes such as threshold method, grading method, Support Vector Machine (SVM) classification, CNN classification, and tested under different lighting conditions and the interference of mixed daily noise motion. The results show that the proposed algorithm is superior to traditional human fall detection algortihms in environmental adaptability and fall detection accuracy. The proposed algorithm can effectively detect the human body in the video and accurately detect the fall state of human body, which further verifies the feasibility and efficiency of the deep learning recognition method with the fusion of motion information in the video fall behavior analysis.

The traditional A ^* algorithm has the disadvantages of long planning time and large search range in unmanned vehicle path planning. After comprehensively analyzing the calculation process of the A ^* algorithm, the A ^* algorithm was improved from four aspects. Firstly, targeted expansion, that is, different quadrants were selected with target for node expansion according to the relative position of the node to be expanded and the target node. Secondly, target visibility judgment, that is, whether there were obstacles between the node to be expanded and the target point was determined, if there were no obstacles, A ^* algorithm jumped out of the exploration process to reduce redundant searches. Thirdly, the heuristic function of the A ^* algorithm was changed, that is, the cost estimation of the n-th generation parent node of the node to be expanded to the target point was increased, thereby reducing the local optimal situation of the cost estimation to the target point. Fourthly, the selection strategy of the expanded nodes was changed, that is, the traditional method of minimizing the heuristic function to select the expanded nodes was changed, and the simulated annealing method was introduced to optimize the selection method of the expanded nodes, so that the search process was performed as close to the target point as possible. Finally, the Matlab simulation experimental results show that, under the simulated map environment, the improved A ^* algorithm has the running time reduced by 67.06%, the number of experienced grids decreased by 73.53%, and the fluctuation range of the optimized path length is ±0.6%.

In Internet of Vehicles(IoV)environment,a single vehicle cannot meet all the time-sensitive driving safety requirements because of limited capability on information acquiring and processing. Cooperation among vehicles to enhance information sharing and channel access ability is inevitable. In order to solve these problems,a cooperative feedback control algorithm based dynamic reinforcement model for driving safety was proposed. Firstly,a virtual fleet cooperation model was proposed to improve the precision and expand the range of global traffic sensing,and a stable cooperation relationship was constructed among vehicles to form cooperative virtual fleet while avoiding channel congestion. Then,a joint optimization model focusing on message transmission and driving control was implemented,and the deep fusion of heterogeneous traffic data was used to maximize the safety utility of IoV. Finally,an adaptive feedback control model was proposed according to the prediction on spatial-temporal change of traffic flow,and the driving safety strategy was able to be adjusted in real-time. Simulation results demonstrate that the proposed model can obtain good performance indexes under different traffic flow distribution models, can effectively support driving assisted control system, and reduce channel congestion while maintaining driving safety.

The scalar neurons of convolutional neural networks cannot express the feature location information,and have poor adaptability to the complex vehicle driving environment,resulting in low traffic sign recognition rate. Therefore,an intelligent traffic sign recognition method based on capsule network was proposed. Firstly,the very deep convolutional neural network was used to improve the feature extraction part. Then,a pooling layer was introduced in the main capsule layer. Finally,the movement index average method was used for improving the dynamic routing algorithm. The test results on the GTSRB dataset show that the improved capsule network method improves the recognition accuracy in special scenes by 10. 02 percentage points. Compared with the traditional convolutional neural network,the proposed method has the recognition time for single image decreased by 2. 09 ms. Experimental results show that the improved capsule network method can meet the requirement of accurate and real-time traffic sign recognition.

Aiming at the problem of low reconstruction quality of the existing Compressed Sensing Magnetic Resonance Imaging (CSMRI) algorithms at low sampling rates, an imaging method combining deep priors and non-local similarity was proposed. Firstly, a deep denoiser and Block Matching and 3D filtering (BM3D) denoiser were used to construct a sparse representation model that can fuse multiple priori knowledge of images. Secondly, the undersampled k-space data was used to construct a compressed sensing magnetic resonance imaging optimization model. Finally, an alternative optimization method was used to solve the constructed optimization problem. The proposed algorithm can not only use the deep priors through the deep denoiser, but also use the non-local similarity of the image through the BM3D denoiser to reconstruct the image. Compared with the reconstruction algorithms based on BM3D, experimental results show that the proposed algorithm has the average peak signal-to-noise ratio of reconstruction increased about 1 dB at the sampling rates of 0.02, 0.06, 0.09 and 0.13. Compared with the existing MRI algorithm WaTMRI (Magnetic Resonance Imaging with Wavelet Tree sparsity),DLMRI (Dictionary Learning for Magnetic Resonance Imaging), DUMRI-BM3D (Magnetic Resonance Imaging based on Dictionary Updating and Block Matching and 3D filtering), etc, the images reconstructed by the proposed algorithm contain a lot of texture information, which are the closest to the original images.

To improve the performance of vehicle classification by making full use of image information from traffic surveillance, Histogram of Oriented Gradient Convolutional (HOG-C) features extraction method was added on the capsule network, and a Capsule Network model fusing with HOG-C features (HOG-C CapsNet) was proposed. Firstly, the gradient data in the images were calculated by the gradient statistical feature extraction layer, and then the Histogram of Oriented Gradient (HOG) feature map was plotted. Secondly, the color information of the image was extracted by the convolutional layer, and then the HOG-C feature map was plotted with the extracted color information and HOG feature map. Finally, the HOG feature map was input into to the convolutional layer extract its abstract features, and the abstract features were encapsulated through a capsule network into capsules with the three-dimensional spatial feature representation, so as to realize the vehicle classification by dynamic routing algorithm. Compared with other related models on the BIT-Vehicle dataset, the proposed model has the accuracy of 98.17%, the Mean Average Precision (MAP) of 97.98%, the Mean Average Recall (MAR) of 98.42% and the comprehensive evaluation index of 98.20%. Experimental results show that the vehicle classification in traffic surveillance scenarios can be achieved with better performance by using HOG-C CapsNet.

Aiming at the low data-transmitting efficiency of the traditional message-oriented middleware in power information system, a firm real-time data-transmitting system based on data stream-transmitting mechanism was proposed. Queue caching mechanism was adopted to realize the asynchronous sending and batch confirmation of message. Data stream-transmitting mechanism was designed to eliminate the cache latency and the cost of cache resources of the data on transit node to improve the timeliness and concurrency of data transmission. Distributed and data routing thought was used-data to make the node network to the third-party system transparently and achieve a data routing distribution function. The simulation results of a provincial electric power information system data exchange scene, verified the system performance. Concurrent data exchange capacity is 3000 concurrent. Transmission speed in the gigabit bandwidth system environment is 980 MB/s. Switching delay is kept in milliseconds.

k-Nearest Neighbor ( kNN) algorithm is commonly used in data imputation. It is of poor efficiency because of the similarity computation between every tow records. To solve the efficiency problem, an improved kNN data imputation algorithm combined with Locality Sensitive Hashing (LSH) named LSH- kNN was proposed. First, all the complete records were indexed in LSH way. Then corresponding LSH ways for nominal, numeric and mixed-type incomplete data were put forward, and LSH values for all the incomplete records were computed in the proposed way to find candidate similar records. Finally, the incomplete records' real distance to candidate similar records were calculated, and the top- k similar records for kNN imputation were found. The experimental results show that the proposed method LSH- kNN has higher efficiency than traditional kNN as well as keeping almost the same accuracy.

Concerning the problem of unsatisfactory feature extraction and low recognition rate caused by redundant words in clustering dictionary in the practical monitoring video for two-person interaction recognition, a Bag Of Word (BOW) model based on improved Spatio-Temporal Interest Point (STIP) feature was proposed. First of all, foreground movement area of interaction was detected in the image sequences by the intractability method of information entropy, then the STIPs were extracted and described by 3-Dimensional Scale-Invariant Feature Transform (3D-SIFT) descriptor in detected area to improve the accuracy of the detection of interest points. Second, the BOW model was built by using the improved Fuzzy C-Means (FCM) clustering method to get the dictionary, and the representation of the training video was obtained based on dictionary projection. Finally, the nearest neighbor classification method was chosen for the two-person interaction recognition. Experimental results showed that compared with the recent STIPs feature algorithm, the improved method with intractability detection achieved 91.7% of recognition rate. The simulation results demonstrate that the intractability detection method combined with improved BOW model can greatly improve the accuracy of two-person interaction recognition, and it is suitable for dynamic background.

In the Wireless Sensor Network (WSN) routing, if the node does not fully consider the path node residual energy and link status of the route, some nodes will be dead early, hence the lifetime of the network will be shorten seriously. To resolve this problem, a uneven clustering routing algorithm for wireless sensor network was proposed based on ant colony optimization algorithm. Firstly, the method clustered nodes using uneven clustering algorithm which considered the node energy. Then considering the node need to transmit data as source node, the sink node as destination node, ant colony optimization algorithm was used to do multipath searching, and the searching process fully considered the factors such as transmission energy consumption, path minimum residual energy, transmission distance and transmission hops, time delay and bandwidth of selected link. Several optimal paths that met the conditions were given to complete the information transmission between source and the destination nodes at last. The experimental results show that the lifetime of WSN can be effectively prolonged while fully considering the path transmission energy consumption, path minimum residual energy and transmission hops.

To solve the system integration problem caused by heterogeneous sensor networks, this paper proposed a heterogeneous sensor networks integration system ISHSN (Integration System for Heterogeneous Sensor Network). ISHSN consisted of the gateway of the Internet of Things (IoT) and the access Agent. The gateway converted the data to the same format and converted the command to the customized format according the target sensor network. The access Agent dealt with data collection, link merge and command forward, and balance loading of the access Agents with the scheduling algorithm. The experiment proves that the ISHSN has good scalability and availability in sensor networks data collection and sensor network control.

The aim of the Quality of Service (QoS) routing in Mobile Ad Hoc NETwork (MANET) is to determine an efficient route with enough available mobile nodes to satisfy a request of a source node, and the selected nodes of Multi-Point Relay (MPR) are on the optimal route that needs evaluating by the routing protocol. In order to steadily look for the optimal QoS routing path with maximum bandwidth and minimum delay from a source node to a destination node with shorter time, a new flexible link-state QoS routing protocol called FLSQR was proposed, which used a new link-state method in which each node' cache stores an Effective Decision Table (EDT) for routing calculation. The FLSQR used the MPR1 and MPR2 selection according to the Effective Distance (ED) in EDT to select the optimal and suboptimal routing paths, and further chose the path of optimal bandwidth and delay by presented metric model. The experimental results show that the proposed FLSQR protocol can acquire better improvement in finding an optimal routing path in MANET than the OLSR and QOLSR-MPR protocol.

Concerning the thresholds selection in image segmentation, this paper proposed a method that used dynamic threshold to divide the histogram of original image into two new independent histograms. The two new histograms correspond to two fictitious images whose sizes are the same to original one，and pixels of the same probability are similar pixels of original image. The cross entropy can be assembled and calculated between probability distribution of the two new histograms. By analyzing the relationship between peak and valley of the maximum of entropy functional curve,the best multi-thresholds for segmentation image can be achieved. This method is simple and clear, and the experiment shows this method is effective.

To meet the need of instruction detection system used in complex natural environment, such as coastal mudflats, an improved barrier coverage model, a multi-line barrier coverage scheduling protocol named k-MLBCSP, a coverage layout algorithm and a coverage adjustment algorithm were proposed. The k-MLBCSP protocol divided the network lifetime into three phases. In the initialization phase, the coverage layout algorithm guaranteed reasonable network settings. In the adjustment phase, the coverage adjustment algorithm provided an effective way for the sink and alive senosrs to further negotiate coverage layout strategies. The theoretical analysis and simulations show that compared with LBCP and RIS, k-MLBCSP increases the sensor network's coverage probability and lifetime. Furthermore, k-MLBCSP reduces the time complexity and the network load.

To detect the critical nodes that can lead to topology partition in a Mobile Ad Hoc Network (MANET), a Cross-link-tolerant Partition Detection Algorithm (CPDA) was proposed. Through utilizing the information of adjacent nodes, CPDA could eliminate the cross-links' impact on the elementary loop. Therefore, it solved the problem that the existing algorithm of Distributed Partition Detection Protocol (DPDP) based on elementary-loop could not address cross links, which improved the accuracy of detection of critical nodes. The performance results show that CPDA has no limitation on network topology and outperforms DPDP in terms of detection accuracy and overhead.

Ranking function based annotation may cause more or fewer labels according to the probability predicted by the model in multi-label image annotation. Therefore, a Convolutional Neural Network with THreshold OPtimization (CNN-THOP) model was proposed. The model consists of Convolutional Neural Network (CNN) and threshold optimization. Firstly, CNN was used to train a model, which was used to predict the image, so as to obtain the prediction probability, and Batch Normalization (BN) layer was added to the CNN to effectively accelerate the convergence. Secondly, threshold optimization was performed by the prediction probabilities of the test set images obtained by the proposed model. After the threshold optimization process, an optimal threshold was obtained for each kind of label, so as to obtain a set of optimal thresholds. Only when the prediction probability of this kind of label was greater than or equal to the best threshold of this kind of label, the image would be labeled with this label. In the labeling process, the CNN model and a set of optimal thresholds were added to achieve more flexible multi-label labeling of the image to be labeled. Through the verification on 8 000 images in the natural scene image dataset, experimental results show that CNN-THOP has about 20 percentage points improvement on average precision compared to Ranking Support Vector Machine (Rank-SVM), and is about 6 percentage points and 4 percentage points higher respectively than Convolutional Neural Network using Mean Square Error function (CNN-MSE) in average recall and F1 value respectively, and has the Complete Matching Degree (CMD) reached 64.75%, which proves that the proposed method is effective in automatic image annotation.