In order to improve the storage security and signature efficiency of digital signature keys in the consortium blockchain Practical Byzantine Fault Tolerance (PBFT) algorithm consensus process, considering the actual application environment of the consortium blockchain PBFT consensus algorithm, a trusted third-party proof signature scheme based on key division and Chinese encryption SM2 algorithm was proposed. In this scheme, by a trusted third-party, the key was generated and split, and the sub-split private key was distributed to the consensus nodes. In each consensus, the identity must be proved to the trusted third-party at first, and then the other half of the sub-split private key was obtained by the verification party to perform identity verification. In this signature scheme, the segmentation and preservation of the private key was realized by combining the characteristics of the consortium chain, and the modular inversion process in the traditional SM2 algorithm was eliminated by using consensus feature and hash digest. The theoretical analysis proved that the proposed scheme was resistant to data tampering and signature forgery, while Java Development Kit (JDK1.8) and TIO network framework were used to simulate the signature process in consensus. Experimental results show that compared with the traditional SM2 algorithm, the proposed scheme is more efficient, and the more consensus nodes, the more obvious the efficiency gap. When the node number reaches 30, the efficiency of the scheme is improved by 27.56%, showing that this scheme can satisfy the current application environment of the consortium blockchain PBFT consensus.

Concerning the fact that common background subtraction methods ignore the spatio-temporal continuity of foreground and the disturbance of dynamic background to foreground extraction, an improved background subtraction model was proposed based on Tensor Robust Principal Component Analysis (TRPCA). The improved tensor nuclear norm was used to constrain the background, which enhanced the low rank of background and retained the spatial information of videos. Then the regularization constraint was performed to the foreground by 3D Total Variation (3D-TV), so as to consider the spatio-temporal continuity of object and effectively suppress the interference of dynamic background and target movement on the foreground extraction. Experimental results show that the proposed model can effectively separate the foreground and background of videos. Compared with High-order Robust Principal Component Analysis (HoRPCA), Tensor Robust Principal Component Analysis with Tensor Nuclear Norm (TRPCA-TNN) and Kronecker-Basis-Representation based Robust Principal Component Analysis (KBR-RPCA), the proposed algorithm has the F-measure values all optimal or sub-optimal. It can be seen that, the proposed model effectively improves the accuracy of foreground and background separation, and suppresses the interference of complex weather and target movement on foreground extraction.

In order to solve the problems of too many parameters, long training time and slow detection speed of the existing human skeleton key point detection models, a detection method combining the human skeleton key point detection model called Convolutional Pose Machines (CPMs) and the lightweight convolutional neural network model called SqueezeNet was proposed. Firstly, the CPMs with 4 stages (CPMs-Stage4) was used to detect the key points of the human images. Then, the Fire Module network structure of SqueezeNet was introduced into CPMs-Stage4 to reduce the model parameters greatly, and thus to obtain a new lightweight human skeleton key point detection model called SqueezeNet15-CPMs-Stage4. The verification results on the extended Leeds Sports Pose (LSP) dataset show that, compared with CPMs, SqueezeNet15-CPMs-Stage4 model has the training time reduced by 86.68%, the detection time of single image reduced by 44.27%, and the detection accuracy of 90.4%; and the proposed model performs the best in training time, detection speed and accuracy compared with three reference models improved VGG-16, DeepCut and DeeperCut. The experimental results show that the proposed model achieves high detection accuracy with short training time and fast detection speed, and can effectively reduce the training cost of the human skeleton key point detection model.

Aiming at the limitations of traditional noise reduction methods for acoustic signals in low Signal-to-Noise Ratio (SNR) environment, a real-time noise reduction algorithm was proposed by combining adaptive threshold Voice Activity Detection (VAD) algorithm and Minimum Mean-Square Error Log-Spectral Amplitude estimation (MMSE-LSA). Firstly, the background noise was estimated in VAD algorithm by probability statistics based on the maximum value of the energy probability, and the obtained background noise was updated in real time and saved. Then, the background noise updated in real time was used as the reference noise of MMSE-LSA, and the noise amplitude spectrum was updated adaptively. Finally, the noise reduction processing was performed. The experimental results on four kinds of acoustic signals in real scenes show that the proposed algorithm can guarantee the real-time processing of low SNR acoustic signals; and compared with the traditional MMSE-LSA algorithm, it has the SNR of the noise reduction signal increased by 10-15 dB without over-subtraction. It can be applied to practical engineering.

Aiming at the problems of insufficient generalization ability, poor stability and difficulty in meeting the real-time requirement of facial expression recognition, a real-time facial expression recognition method based on multi-scale kernel feature convolutional neural network was proposed. Firstly, an improved MSSD (MobileNet+Single Shot multiBox Detector) lightweight face detection network was proposed, and the detected face coordinates information was tracked by Kernel Correlation Filter (KCF) model to improve the detection speed and stability. Then, three linear bottlenecks of three different scale convolution kernels were used to form three branches. The multi-scale kernel convolution unit was formed by the feature fusion of channel combination, and the diversity feature was used to improve the accuracy of expression recognition. Finally, in order to improve the generalization ability of the model and prevent over-fitting, different linear transformation methods were used for data enhancement to augment the dataset, and the model trained on the FER-2013 facial expression dataset was migrated to the small sample CK+ dataset for retraining. The experimental results show that the recognition rate of the proposed method on the FER-2013 dataset reaches 73.0%, which is 1.8% higher than that of the Kaggle Expression Recognition Challenge champion, and the recognition rate of the proposed method on the CK+ dataset reaches 99.5%. For 640×480 video, the face detection speed of the proposed method reaches 158 frames per second, which is 6.3 times of that of the mainstream face detection network MTCNN (MultiTask Cascaded Convolutional Neural Network). At the same time, the overall speed of face detection and expression recognition of the proposed method reaches 78 frames per second. It can be seen that the proposed method can achieve fast and accurate facial expression recognition.

In view of the fact that the low rank and sparse methods generally regard the foreground as abnormal pixels in the background, which makes the foreground detection precision decrease in the complex scene, a new foreground detection method combining weighted Schatten- p norm with 3D Total Variation (3D-TV) was proposed. Firstly, the observed data were divided into low rank background, moving foreground and dynamic disturbance. Then 3D total variation was used to constrain the moving foreground and strengthen the prior consideration of the spatio-temporal continuity of the foreground objects, effectively suppressing the random disturbance of the anomalous pixels in the discontinuous dynamic background. Finally, the low rank performance of video background was constrained by weighted Schatten- p norm to remove noise interference. The experimental results show that, compared with Robust Principal Component Analysis (RPCA), Higher-order RPCA (HoRPCA) and Tensor RPCA (TRPCA), the proposed model has the highest F-measure value, and the optimal or sub-optimal values of recall and precision. It can be concluded that the proposed model can better overcome the interference in complex scenes, such as dynamic background and severe weather, and its extraction accuracy as well as visual effect of moving objects is improved.

In order to reduce the harm caused by bursty events in social media, a multi-feature based microblog bursty events detection algorithm was proposed. The algorithm combines text emotion filtering and user influence calculation methods. Firstly, the microblog text with negative emotion was obtained through noise filtering and emotion filtering. Then the proposed user influence calculation method was combined with the burst word extraction algorithm to extract the characteristics of burst words. Finally, a cohesive hierarchical clustering algorithm was introduced to cluster bursty word sets, and extract bursty events from them. In the experimental test, the accuracy is 66.84%, which proves that the proposed method can effectively detect bursty events.

Aiming at the problems that traditional short-time traffic flow prediction method in highway domain is suitable for small scale data, which limits the efficiency on massive data, and the spatio-temporal relationship of data is neglected, a short-term traffic flow prediction method for big data with combining K-Nearest Neighbors ( KNN) in highway domain was proposed. Firstly, the K value and distance metric of model were tuned, and the model parameters were compared by using cross validation. Secondly, considering inherent spatio-temporal association of data, feature vectors based on spatio-temporal characteristics were modeled. Finally, a regression prediction model was established under big data environment, and the prediction was realized with the model of optimal parameters. The experimental results show that compared with traditional time series model, the proposed model works on all toll stations at one time, has high speed of single running and improves the efficiency by 77%. The method significantly reduces Mean Absolute Percentage Error (MAPE) and Median Absolute Percentage Error (MDAPE) and it also has good horizontal expansibility.

Current popular prediction methods of foreign trade product sales only respectively study prediction problems from angles of the third party platform or big data, lacking consideration of dynamic evolution prediction on product sales based on Internet platform, big data and cross-border e-commerce. To improve the efficiency of export sales prediction, to achieve scalability and dynamic evolution of prediction systems, with mining controllable relevance big data of cross-border e-commerce export sale based on "Internet+foreign trade" surroundings, personalized prediction mechanism and smart prediction algorithms, improving corresponding algorithms such as distributed quantitative calculation and centralized qualitative calculation, a dynamic prediction model on export sales based on "Internet+foreign trade"-driven controllable relevance big data of cross-border e-commerce was proposed. Finally, this model was verified and analyzed. The performance analysis results show that the model integrates fully openness and extensibility of "Internet+" and dynamic prediction advantages of big data, achieving dynamic, smart, quantitative, and qualitative prediction on export sales based on "Internet+foreign trade"-driven controllable relevance big data of cross-border e-commerce. The comprehensive prediction efficiency of the proposed model is obviously better than those of traditional models, and it has stronger dynamic evolution and higher utility.

In recent years, Sparse Representation based Classifier (SRC) has become a hot issue which has been great successful in face recognition. However, when the SRC reconstructed test samples, it is possible to use the training samples with large difference from the test samples, meanwhile, SRC tends to lose locality information and thus produces unstable classification results. A Self-Paced Learning Weighted Sparse Representation based Classifier (SPL-WSRC) was proposed. It could effectively eliminate the training samples with large difference from the test samples. In addition, locality information between the samples was considered by weighting to improve the classification accuracy and stability. The experimental results on three classical face databases show that the proposed SPL-WSRC algorithm is better than the original SRC algorithm. The effect is more obvious, especially when the training samples are enough.

For the question of parallel test task scheduling, an innovative solution based on graph coloring theory and genetic-bee colony algorithm was proposed. Firstly, a relation model of test tasks was established based on graph coloring theory, in which the occupation of device resource by test task could be represented by graph. Based on this relation model of test task, the optimum solution was searched via combining the artificial bee colony algorithm and the crossover operation and mutation operation which are unique in genetic algorithm to avoid the prematurity of the algorithm as well as accelerate convergence. Eventually, a grouping scheme was generated with maximized parallelism degree. Verified by the simulation, the proposed method can effectively realize the parallel test, improve the test efficiency of automatic test system.

The single frequency modulation or phase modulation waveform based on chaotic sequence has low waveform complexity, which limits predictive probability of chaotic signal, radar intercept probability and anti-interference performance. In order to solve the problems, joint frequency-phase modulation based on chaotic sequence in radar waveform was proposed. Firstly, the radar signal was carried out for the chaotic frequency encoding, which was that a pulse was divided into a series of sub-pulses and different frequency modulation was carried out for different sub-pulses. At the same time, in each frequency encoding sub-pulse, the random initial phase was used in each cycle of waveform. The simulation results show that the maximum value of autocorrelation sidelobe peak of joint frequency-phase modulation based on chaotic radar signal achieved -24.71 dB. Compared with the frequency modulation or phase modulation based on chaotic signal, the correlation performance of the proposed joint frequency-phase modulation has improved. The experimental results show that, the joint frequency-phase modulation chaotic radar waveform combines the advantages of phase modulation and frequency modulation and is an ideal detection signal with the flat power spectrum characteristic of phase modulation and anti-noise-interference ability of frequency modulation.

In order to reduce the negative impacts of sparse data, a new collaborative filtering recommendation algorithm was put forward based on the number of common rating items among users and the similarity of user interests. The similarity calculations were made to be more credible by combing the number of common rating items among users with the similarity of user interests, so as to provide better recommendation results for the target user. Compared with the method based on Pearson similarity, the new algorithm provides better recommendation results with smaller Mean Absolute Error (MAE). In conclusion, the new algorithm is effective and feasible.

To solve the problem of location verification caused by collusion attack in Vehicular Ad Hoc NETworks (VANET), a multi-round vote location verification based on weight and difference was proposed. In the mechanism, a static frame was introduced and the Beacon messages format was redesigned to alleviate the time delay of location verification. By setting malicious vehicles filtering process, the position of the specific region was voted by the neighbors with different degrees of trust, which could obtain credible position verification. The experimental results illustrate that in the case of collusion attack, the scheme achieves a higher accuracy of 93.4% compared to Minimum Mean Square Estimation (MMSE) based location verification mechanism.

Heavy land clutter with antenna scan is the main interference for Foreign Object Debris (FOD) detection. However, traditional Constant False Alarm Rate (CFAR) in space-domain is ineffective to detect targets on runways. To solve this problem, a cell-average clutter-map CFAR was proposed. First of all, an echo model based on the characteristics of FOD surveillance radar system was built. Then, the range-bearing two-dimensional CFAR detection could be realized by using clutter-map cells dividing, cell averaging and recursive filtering. Further analysis of the main factors that affected the detection performance of this method was studied in the end. The simulation results show that, the proposed algorithm can effectively detect the weak target and obtain high detection probability with a low signal-to-clutter ratio.

In view of the limited accuracy of imaging algorithm，the nonlinear Contrast Source Inversion (CSI) algorithm combined with regularization and Concurrent Frequency (CF) was proposed for reconstructing a hybrid target in an anechoic chamber. The experimental data were obtained using multi-frequency multi-bistatic measurements. The reconstructed position, shape and contrast value of the target were presented, verifying the accuracy of the extended CSI algorithm for reconstructing the complicated 2-D hybrid targets.

To enhance the performance of Shuffled Frog Leaping Algorithm (SFLA) in solving optimization problems，this paper proposed an improved shuffled frog leaping algorithm. By adding mutation operator to the original algorithm, the improved algorithm regulated the scale of mutation operator via fuzzy controller, made a dynamic adjustment of mutation operator in the searching range of solution space with different phase and candidate solution distribution of evolution process. The simulation results of four typical functions of optimization problems show that the proposed algorithm can attain above twice improvement on accuracy, convergent speed and success rate, and it demonstrates a better optimization capability especially in solving the high dimensional complex optimization problem, in comparison with the basic shuffled frog leaping algorithm and the known improved algorithm.

The traditional collaborative filtering recommender algorithms based on Internet pattern research merely E-commerce recommender problem from one angle, and their recommender quality is evidently not high. To improve recommender efficiency, and to achieve scalability and utility of recommendation systems, with studying user behavior similarity measure formula, grade function and correlation rule function based on cloud pattern, a correlation clustering method was put forward. To improve the corresponding algorithms, a cloud pattern collaborative filtering recommender algorithm based on user behavior correlation clustering was proposed. Finally, the improved algorithms were validated by local and global experiments using MovieLens and Alibaba cloud testing data. The experimental results show that the recommender efficiency of the proposed algorithm is obviously higher than those of traditional algorithms, and it has stronger scalability and higher utility.

Using active contour model to get the bone of feathers is affected by other strong edge, and the computation is too much. According to the characteristics of feathers, a method of describing the object contour using the centerline and width was proposed. Two-dimensional contour described in the model was converted into two independent one-dimensional functions, and according to it, the energy function of the model was modified. The improved algorithm made use of symmetry to avoid the interference of strong edges, reduced the computation scale, and it could achieve fully automatic segmentation. The experimental results show that the improved algorithm is robust to noise; it realizes good feather image segmentation and meets industry needs.

To improve bottom-layered information resource addressing efficiency for Internet of Things (IoT), with researching trust evaluation criteria on bottom-layered addressing services for IoT in cloud, and improving trust-driven algorithms, an information resource addressing model based on trust-driven cloud for IoT was presented. First the key addressing features were analyzed, then the addressing model was constructed by designing and using specific constraint conditions, trust steepness function, cloud trust evaluation criteria and trust constraint coefficients. Finally, the model was validated by an IoT system designing instance. The experimental results show the proposed model has satisfactory bottom-layered resource addressing efficiency in comparison with traditional models or algorithms.

An adaptive filter to remove mixed Gaussian noise and impulse noise of the corrupted images based on their minimum entropies was proposed. The proposed filter calculated the noise rate of the corrupted image based on the minimum local entropy, and utilized the noise rate to determine the size of filter windows and the number of pixels to be brushed off in the filter windows of Alphatrimmed mean filter, and then used the average of leaved pixels' gray as the output of the filter. The performance of the proposed filter was evaluated by experiments. Experimental results show that the superiority of this filter including the ability of removing noises and that of preserving the partial details of images in comparison with some existing methods (median filter and mean filter), and this filter is an adaptive Alpha-trimmed mean filter.

Aiming at the lack of the function of 3D Modeling of ComGIS,the idea of 3D Modeling based on ComGIS and CAD’ Integration was put forward,and an integration method of 3D Modeling was introduced.The idea was that GIS data’ import,3D Modeling and raster picture of 3D perspective drawing were realized by Visual LISP programming of AutoCAD,the command of WIN32 API ran AutoCAD in background,and Command Scripts realized loading and running of Visual LISP,and at last,GIS displayed 3D Model.

By the study of several typic methods of determining convexo-concave features for vertices of polygon,the computation time complexity of these methods was analysed, and those methods were implemented in (VC++).

Based on high effective locations algorithm and error estimate, improvement of the public-key scheme based on Chebyshev maps proposed recently was presented and a key agreement protocol was given. Experiment results and analytic results show that it is very secure and practical.