In order to solve the problems that the accuracy is low or the head pose estimation cannot be performed by traditional head pose estimation methods when the key feature points of the face cannot be located due to partial occlusion or too large angle, a multi-angle head pose estimation method based on optimized LeNet-5 network was proposed. Firstly, the depth, the size of the convolution kernel and other parameters of the Convolutional Neural Network (CNN) were optimized to better capture the global features of the image. Then, the pooling layers were improved, and a convolutional operation was used to replace the pooling operation to increase the nonlinear ability of the network. Finally, the AdaBound optimizer was introduced, and the Softmax regression model was used to perform the pose classification training. During the training, hair occlusion, exaggerated expressions and wearing glasses were added to the self-built dataset to increase the generalization ability of the network. Experimental results show that, the proposed method can realize the head pose estimation under multi-angle rotations, such as head up, head down and head tilting without locating key facial feature points, under the occlusion of light, shadow and hair, with the accuracy of 98.7% on Pointing04 public dataset and CAS-PEAL-R1 public dataset, and the average running speed of 22-29 frames per second.

In order to prevent photo and video attacks in the face recognition system, an interactive liveness detection algorithm was proposed which combines the head pose and facial expression. Firstly, the number of convolution kernels, network layers, and regularization of VGGNet were adjusted and optimized, and a multi-layer convolutional head pose estimation network was constructed. Secondly, the methods such as global average pooling, local response normalization and convolutional replacement pooling were introduced to improve VGGNet and build an expression recognition network. Finally, the above two networks were fused to realize an interactive liveness detection system, which sends random instructions to users to complete liveness detection in real time. The experimental results show that the proposed head pose estimation network and expression recognition network achieve 99.87% and 99.60% accuracy on CAS-PEAL-R1 dataset and CK+ dataset respectively, and the liveness detection system has the comprehensive accuracy reached 96.70%, the running speed reaches 20-28 frames per second, which make the generalization ability of the system outstanding in the practical application.

In order to solve the problem of insufficient available training data in the classification task of breast mass and calcification, a multi-view model based on secondary transfer learning was proposed combining with imaging characteristics of mammogram. Firstly, CBIS-DDSM (Curated Breast Imaging Subset of Digital Database for Screening Mammography) was used to construct the breast local tissue section dataset for the pre-training of the backbone network, and the domain adaptation learning of the backbone network was completed, so the backbone network had the essential ability of capturing pathological features. Then, the backbone network was secondarily transferred to the multi-view model and was fine-tuned based on the dataset of Mianyang Central Hospital. At the same time, the number of positive samples in the training was increased by CBIS-DDSM to improve the generalization ability of the network. The experimental results show that the domain adaption learning and data augmentation strategy improves the performance criteria by 17% averagely and achieves 94% and 90% AUC (Area Under Curve) values for mass and calcification respectively.

In order to solve the problems of copyright registration, rampant piracy and copyright disputes faced by digital intellectual property under Internet ecology, a Digital Copyright Identifier (DCI) control model of digital works without trusted third party was proposed. Firstly, the Peer-to-Peer (P2P) architecture based on the concept of de-centralization of blockchain was constructed. The blockchain replaced the traditional database as the core of storage mechanism. Through the creation of transactions, construction of blocks, legitimacy validation and link of blocks a digital work blockchain transaction information storage structure was built, guaranteeing the copyright information not be tampered and traceable. Secondly, the digital distribution protocol based on smart contract was proposed, three types of contracts include copyright registration, inquiry and transfer were designed, and the transactions were generated by automatically executing the preset instructions to ensure the transparency and high efficiency of models. Theoretical analysis and simulation show that the probability of forged block attack is close to zero in the digital work blockchain network, compared with the traditional copyright authentication mechanism based on trusted third party, the model has better architectural security. The experimental results show that the model simplifies the threshold of digital copyright registration, enhances the authority of copyright certification and has better real-time and robustness.

Aiming at the problem that the traditional entity alignment algorithm may lead to bad performance in entity alignment task of Chinese heterogeneous encyclopedia knowledge base, an entity alignment method based on entity attributes and the features of context topics was proposed. First, a Chinese heterogeneous encyclopedia knowledge base was constructed based on Baidu encyclopedia and Hudong encyclopedia data. Next, the Resource Description Framework Schema (RDFS) vocabulary list was made to normalize the entity attributes. Then the entity context information was extracted and the Chinese word segmentation was used on the contexts. The contexts were modelled by using the topic model and the parameters were computed by Gibbs sampling method. After that the topic-word probability matrix, the characteristic word collection and the corresponding feature matrix were calculated. Last, the Longest Common Subsequence (LCS) algorithm was used to compute the entity attribute similarity. When the similarity was between the lower and the upper bounds, the topic features of the entities' context were combined to resolve the entity alignment problem. Finally, according to the standard method, an entity alignment data set of Chinese heterogeneous encyclopedia was constructed for simulation experiments. In comparison with the traditional property similarity algorithm, weighted-property algorithm, context term frequency feature model and topic model algorithm, the experimental results show that the proposed method achieves 97.8% accuracy, 88.0% recall, 92.6% F-score in people class and 98.6% accuracy, 73.0% recall, 83.9% F-score in movie class. It outperformed the other entity alignment algorithms. The experimental results also indicate that the proposed method can improve the entity alignment results in constructing the Chinese heterogeneous encyclopedia knowledge base, and it can be applied to the entity alignment tasks with context information.

Concerning the large-scale concurrent video stream scheduling problem of low resource utilization and load imbalance under cloud environment, a Video-on-Demand (VOD) scheduling policy based on Ant Colony Optimization (ACO) algorithm named VodAco was proposed. The correlation of video stream expected performance and server idle performance was analyzed, and a mathematical model was built based on the definition of comprehensive matching degree, then ACO method was adopted to hunt the best scheduling schemes. The contrast experiments with Round Robin (RR) and greedy schemes were tested on CloudSim. The experimental results show that the proposed policy has more obvious advantages in task completion time, platform resources occupancy and node load balancing performance.

The critical issues in design of routing protocol for Wireless Sensor Network (WSN) are energy awareness and maximizing the lifetime. Focus on those challenges, a new routing algorithm named CTOR was proposed based on time synchronization sleeping schedule with low duty cycle. In CTOR, a node selected several proper forwarders in order to gain assured delivery ratio, and then broadcasted the packets to forwarders; in order to reduce replication packets and yield much gain efficiently, forwarders that received packets forwarded packets according to a probability. Then the sink node broadcasted control messages to make all nodes time synchronized, the other nodes turned into sleeping mode according to the permanent duty cycle. This mechanism can reduce the power consumption and hence the network can work longer. The experimental results show that CTOR can alleviate the routing hole problem, prolong the lifetime of the network and increase the delivery ration of packets.

Embedded Real-time Control System (ERCS) is widely used in all kinds of control systems. Its software is different from other common software. Besides to meet the real-time requirements, the software needs to be reliable. At first, abstract formalization for the software of embedded real-time control system was defined. Then, reliability modeling was given for software modules which can not be subdivided, and the reliability modeling for software of embedded real-time control system was also provided by applying Copula function. The reliability of specific system software was calculated by using the model. The results show that the reliability model established with Copula function takes account of the correlation between software modules, therefore the reliability of software modules of embedded real-time control system was improved compared with the independent modules.

The products of embedded system often have hardware failure while in use, which affects the safety and reliability of the system. In this paper, its reliability was researched from the embedded system hardware level. At first, it defined the goal of the embedded system hardware, and introduced the theory of Markov process. And then, a single IP hardcore and embedded system hardware Markov model was set up. At last, the reliability of the embedded controller hardware automatic block was calculated and analyzed through the established model. The results show that the Markov model can accurately describe the embedded system hardware state change, and can calculate and analyze their reliability. Therefore, the model has certain practical value.

Concerning the partitioning problems of the embedded system software and hardware, a method based on π-nets was proposed to partition the software and hardware of the embedded system. This paper gave a brief introduction to the definition and π-nets rules, and then described and defined the target, and established the division Embedded-system Software and Hardware Partition Model (ESHPM) applying the π-nets of software and hardware of embedded system. Finally this paper analyzed the consistency, deadlock and compatibility; at the same time, optimized the ERSHPM. The ESHPM established in this paper satisfied the consistency and no deadlock between the various processes. And the interaction between the various processes was compatible. The ESHPM effectively improves the accuracy of division, and a more reasonable division method of software and hardware has been got.

An integral method on how to protect software was presented.The method was based on register code software protection.The method of "one computer one register code",and RSA(Rivest-Shamir-Adelman) algorithm were used to encrypt the register code and save different encrypt keys in a database.The method can prevent the illegal register code and illegal register program.In the end,the software protection of the method was realized with(VC++) 6.0.

Signatures indentification of security and attack forecast are integrant function parts of network security field, and the description and definition of attack models and other security signatures request special language. But there exist many questions on the current such languages, such as solitary function of language and weak adaptablility; lack of openness and semantic coherence, and absent reuse ablility. In order to improve this state, the ontology's modeling means were used. It was demonstrated that ontology's language is fit for the description and definition of attack models and other security signatures by a representative attack.

The network structure design for Ad hoc is different from those of the traditional ones. The node structure was expounded and some typical network structures were compared firstly. Then a kind of protocol stack for Ad hoc was given. Different from other models, there was a middleware layer between transmission layer and net layer in this model, which shielded the OS and network’s low layers details and enhanced the reliability and security of communications at the same time.