Aiming at the problems of low utilization rate, high error rate, no recognition ability and manual participation in video surveillance processing in coastal and inland waters of China, a new ship tracking and recognition method based on Darknet network model and YOLOv3 algorithm was proposed to realize ship tracking and real-time detection and recognition of ship types, solving the problem of ship tracking and recognition in important monitored waters. In the Darknet network of the proposed method, the idea of residual network was introduced, the cross-layer jump connection was used to increase the depth of the network, and the ship depth feature matrix was constructed to extract advanced ship features for combination learning and obtaining the ship feature map. On the above basis, YOLOv3 algorithm was introduced to realize target prediction based on image global information, and target region prediction and target class prediction were integrated into a single neural network model. Punishment mechanism was added to improve the ship feature difference between frames. By using logistic regression layer for binary classification prediction, target tracking and recognition was able to be realized quickly with high accuracy. The experimental results show that, the proposed algorithm achieves an average recognition accuracy of 89.5% with the speed of 30 frame/s; compared with traditional and deep learning algorithms, it not only has better real-time performance and accuracy, but also has better robustness to various environmental changes, and can recognize the types and important parts of various ships.

The prediction accuracy of link prediction based on node similarity is always reduced without considering the asymmetric information. In order to solve the problem, a novel method for node similarity measurement with asymmetric information was proposed. Firstly, the disadvantage of the similarity measure algorithm based on Common Neighbor (CN) was analyzed, which it only considered the number of CNs without considering the number of all neighbors of each node. Secondly, the similarity measure between nodes was defined as the ratio of the common nodes to all the neighbor nodes. Then, the symmetric similar information and the asymmetric similar information between nodes were combined, and the similarity between nodes was described in detail. Finally, the proposed method was applied to predict the link relationship in complex networks. The experimental results on the real datasets show that, compared with the previous common neighbor-based similarity measurement methods such as CN, Adamic Adar (AA) and Resource Allocation (RA), the proposed method can improve the accuracy of node similarity measurement and improve the accuracy of link relationship prediction in complex networks.

Cloud computing data are vulnerable to be theft illegally and tampered maliciously. To solve these problems, a Dynamic Updating Ciphertext-Policy Attribute-Based Encryption (DU-CPABE) scheme which enables both data dynamic updating and security protection was proposed. Firstly, by using linear partitioning algorithm, data information was divided into fixed size blocks. Secondly, the data blocks were encrypted by using Ciphertext-Policy Attribute-Based Encryption (CP-ABE) algorithm. Finally, based on conventional Merkle Hash Tree (MHT), an Address-MHT (A-MHT) was proposed for the operation of dynamically updating data in cloud computing. The theoretical analysis proved the security of the scheme, and the simulation in ideal channel showed that, for five updates, compared with CP-ABE method, the average time overhead of data update was decreased by 14.6%. The experimental results show that the dynamic updating of DU-CPABE scheme in cloud computng services can effectively reduce data update time and system overhead.

The existing image completion methods have the problems of structural distortion on visual connectivity and easy to overfitting in the process of training. In order to solve the problems, a new image completion method of Generative Adversarial Network (GAN) based on two discrimination networks was proposed. One completion network, one global discrimination network and one local discrimination network were used in the completion model of the proposed method. The broken area of image to be completed was filled by a similar patch as input in the completion network, which greatly improved the speed and quality of the generation images. The global marginal structure information and feature information were used comprehensively in the global discrimination network to ensure that the completed image of completion network conformed visual connectivity. While discriminating the output image, the assisted feature patches found from multiple images were used to improve the generalization ability of discrimination in the local discrimination network, which solved the issue that the completion network was easily overfitting with too concentrated features or single feature. The experimental results show that, the proposed completion method has good completion effect on face images, and has good applicability in different kinds of images. The Peak Signal-to-Noise Ratio (PSNR) and Structural SIMilarity (SSIM) of the proposed method are better than those of the state-of-the-art methods based on deep learning.

Since the segmentation performance of existing binary segmentation algorithm for video is excessively low, a binary probability segmentation algorithm in real-time based on Graphics Processing Unit (GPU) was proposed. The algorithm implemented a probabilistic segmentation based on the Quadratic Markov Measure Field (QMMF) model by regularizing the likelihood of each pixel of frame belonging to forground class or background class. In this algorithm, first two kinds of likelihood models, Static Background Likelihood Model (SBLM) and Unstable Background Likelihood Model (UBLM) were proposed. Secondly, the probability of each pixel belonging to background was computed by tonal transforming, cast shadow detecting and camouflage detecting algorithm. Finally, the probability of background which makes the energy function have a minimum value was computed by Gauss-Seidel model iteration and the binary value of each pixel was calculated. Moreover, illumination change, cast shadow and camouflage were included to improve the accuracy of segmentation algorithm. In order to fulfill the real-time requirement, a parallel version of our algorithm was implemented in a NVIDIA GPU. The accuracy and GPU execution time of the segmentation algorithm were analyzed. The experimental results show that the average missing rate and false detection rate of ViBe+ and GMM+ are 3 and 6 times those of QMMF, the average execution time of GPU of ViBe+ and GMM+ is about 1.3 times that of QMMF. Moreover, the average speedup of algorithm was computed and it is about 76.8.

At present, in the field of target recognition, the highest accuracy algorithm is the Deformable Part Model (DPM) for human detection. Aiming at the disadvantage of large amount of calculation, a parallel solution method based on Graphics Processing Unit (GPU) was proposed. In this paper, with the GPU programming model of OpenCL, the details of the whole DPM algorithm were implemented by the parallel methods,and optimization of the memory model and threads allocation was made. Through the comparison of the OpenCV library and the GPU implementation, under the premise of ensuring the detection effect, the execution efficiency of the algorithm was increased by nearly 8 times.

E-mail security scanning software is being widely used. With the rapid increase of user number and system flow, there are insurmountable problems in terms of performance, robustness, maintainability and scalability in the traditional tightly-coupled and synchronous IMHS (Inter-scan Message Hosted Security) system. With regards to the system bottleneck by the mass users, a loosely-coupled, asynchronous, stateless principle was proposed for system design. Through the integration of cloud computing and Service Oriented Architecture (SOA) technique, a P2P-based E-mail secure cloud service system was designed and implemented. The system supports dynamic collaborative process, and effectively improves the efficiency of resource use and system scalability. The results and analysis of typical operation tests in real system verify the feasibility and effectiveness of the system architecture.

Focusing on the problem of high efficiency resource scheduling in cloud computing environment, since current research has been less concerned about the cost of the services of the cloud service provider, an improved Quantum Genetic Algorithm (QGA) was proposed to reduce the minimum service cost of cloud service provider. This algorithm converted quantum-bits encoded by binary number to real-coded quantum-bits as chromosome represented by binary-coded quantum-bits cannot describe the resource scheduling matrix, and used rotation strategy and mutation operator to guarantee the convergence of the algorithm. Comparative experiments were conducted among the improved QGA, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) through simulation platform, the populations number is 1 and 100 with 100 iteration times. The experimental results show that the improved QGA can obtain smaller minimum service cost.

To cope with Multi-objective Programming on Manufacturing Cloud Service Composition (MOP-MCSC) problem in cloud manufacturing (CMfg) system, a mathematical model and a solution algorithm were proposed and studied. Firstly, inspired by the resource service composition technology in manufacturing grid, a QoS-aware MOP-MCSC model in CMfg system had been explored and described. Secondly, by analyzing the characteristics of manufacturing cloud services according to the domain knowledge of manufacturing, an eight-dimensional QoS evaluation criterion with corresponding quantitative calculation formulas was defined. Then, the QoS expression of manufacturing cloud service was eventually formulated. Lastly, the MOP-MCSC model was built, and an Adaptive Mutation Particle Swarm Optimization (AMPSO) was designed to realize this model. The simulation experimental results suggest that the proposed algorithm could solve the MOP-MCSC problem efficiently and effectively with a better performance than conventional particle swarm optimization.