To solve the issue of poor performance of present magnetorheological damper electric model for response characteristics, a nonlinear secondary coil model is put forward based on self-design damper. The model parameters are identified with GA algorithm based on square-wave test result. The transfer function of the circuit and damper model is analyzed and a current driver with variable structure is designed. An ADRC controller is designed for the current driver. HIL test results show that, compared with PID algorithm and BUCK circuit, the designed current driver and ADRC controller can improve the response characteristics of magnetorheological damper.

In order to solve the trajectory planning of intelligent vehicle with multi-constraint conditions in unknown environment, a rolling optimization scheme is presented. A feasible trajectory planning method in the local environment is proposed via vehicle dynamic study. With the characteristics of smooth trajectories of six polynomials, the process of trajectory planning is transformed into solving a single polynomial coefficient. The obstacles in the planning window are classified according to the metric function, and the optimal trajectory and vehicle speed are decided according to the optima index. Simulink and CarSim are used to develop the environmental model and the trajectory model. The effectiveness of the planning scheme is verified by simulation study.

The non-uniform contact states of friction pairs are simulated by the different contact ratio. A lubrication model of the friction pair is proposed and corresponding tests are conducted on a wet clutch comprehensive test bench. The results simulations and tests show that, as the contact ratio reduces the surface temperature of friction pair increases obviously, and the local friction coefficient and total friction coefficient increase dramatically, so that the local friction torque grows significantly. In addition, the vibration of the output torque becomes more severe when the contact ratio reduces. Therefore, the non-uniform contact after thermal buckling severely deteriorates the friction properties of the friction pairs, and accelerates the failure of the wet clutches.

A multi-parameters aerodynamic optimization of vehicle body with internal flow is carried out using genetic algorithm. The error of the aerodynamic drag coefficient between the experimental and numerical results is about 6%, indicating that the numerical simulation method is correct. Comparing the obtained intake flow and velocity contour, the difference between the simplified model with internal flow and the real model is small. Therefore, the aerodynamic optimization can be performed using the simplified flow model. Six key parameters of the body surface are selected to build a parameterized model with internal flow. The genetic algorithm based on the roulette wheel selection method and the Taguchi method is used in the aerodynamic optimization. An optimization model with an aerodynamic drag coefficient of 0.298 is obtained. Comparing with the original model, the rear trunk lid dip angle and departure angle obtained with the optimization model both contract inward. The simplified method with internal flow and the multi-parameters aerodynamic optimization method can be used to provide reference for the aerodynamic development of related model of vehicle.

The traffic flows in the junction bottleneck area of urban expressway under normal weather condition, and under snow and ice conditions were investigated by video footage and manual counts. Then, the key values of the urban expressway running characteristics in the on-ramp bottleneck area were obtained, by which the improved Cell Transmission Model can be used to perform simulation of the running state of urban expressway. Furthermore, a variable speed-limit control method under snow and ice conditions was put forward, and simulation was carried out. Simulation results show that during the traffic congestion time, using the proposed method, the traffic efficiency is increased by 16.16 in the junction bottleneck area of urban expressway under snow and ice condition, effectively avoiding serious traffic congestion.

The key of realizing the cross dock is to design the joint of inbound trucks and out bound trucks. A proper sequence makes the cross docking more efficient and needs less makespan. Regarding the number of items in the temporary inventory was proportional to the makespan, this paper transfers the objective function of minimizing the makespan into minimizing the number of items in the temporary inventory. An Improved Harmony Search (IHS) algorithm is proposed to solve the optimization problem. Based on the solving limitation on the parameter value of harmony search algorithm, dynamic adjustment method is used to set parameters. The fixed parameters optimized by Taguchi experiments effectively improve the accuracy of solutions further. The simulation results show that the IHS solutions are more close to the overall optimal solutions than the Harmony Search and Tabu Search. HIS is better in terms of searching for optimal solutions to solve the problem of sequencing among inbound and outbound trucks.

Traditional pedestrian detection system uses global template for search and recognition, resulting in large useless search and reducing the realtime pedestrian detection performance. In addition, it is difficult to apply to unmanned vehicle and pedestrian safety technology in the traffic environment. Based on the difference of motion between pedestrian and vehicle moving background, a fat pedestrian area estimation method is proposed using optical flow clustering. This method uses the combination of local and global optical flow to obtain the optical flow field. According to the difference between background and foreground, the optical flow clustering algorithm and background estimation method are established. Then, the image segmentation algorithm is used to realize the segmentation of optical flow vector. Finally, according to the characteristics of the human body, we can judge whether the candidate region of the moving object is an effective area for pedestrians and realize the pedestrian area estimation. Experiments are carried out based on JLU-PDS and Daimler international pedestrian sharing test library, and experimental environments include the crowed, shelter, streets, fields and other auxiliary working conditions. Results show that the algorithm has good detection effect and greatly reduces the sear space of pedestrian recognition. In addition, it provides technical support and basis for research and application of vehicle active pedestrian protection technology and unmanned intelligent vehicles.

Set covering method is applied to determine the location of gas station for network expansion. A novel gas station location model was built with variable coverage radius. A two-stage approximation greedy algorithm was proposed to determine the size and location of new gas station so that capacity balance between the new station and existing station was achieved. The uncertainty of major parameters was considered and solved well by using of Mont Carlo simulation. The ‘different areas with different coverage radius’ strategy was proposed. Computational results demonstrate that the proposed approach performed very well in keeping with the relationship between the refueling demand and the service capacity in the coverage area. It was adaptive better to the changes of population density and vehicle density in the planning area from city center to city periphery.

To study the influence of unconfined compressive strength of fiber soil with different porosity, the porosity, fiber length and fiber content were taken as the test variables. According to the results of single factor experiment, BBD response surface experiment design method was employed to conduct the unconfined compressive strength tests, and fitting equations of different variables and the response values were established. Test results show that fiber reinforced viscous effects of unconfined compressive strength of the soil is the largest when the fiber length is 9 mm, fiber content is 3‰ and fiber soil porosity is 0.4. Obvious interaction between fiber length and porosity is observed. The results may provide reference for fiber soil engineering application, such as sub grade engineering, foundation engineering design and construction.

In this paper, the effects of injection pressure on the droplet size distribution characteristics in spray front field, spray front and full range droplet size of gasoline direct injection, as well as space distribution characteristics of the droplet size in the spray front field are investigated the using laser diffraction technique. The results indicate that, first, the increase in injection pressure can improve the atomization quality, and promote the droplet size distribution of the spray front field moving to the small diameter direction; second, the spray front droplet size (SMD and Dv90) is bigger than the full range droplet size (SMD and Dv90); third, in the spray front field, the SMD near the near the spray plume axis is bigger than that in the marginal region and the Dv90 near the spray plume axis is smaller than that in the marginal region.

The effects of setting pressure and motored speed on the movement characteristics are studied and the reaction deviation is analyzed using direct-acting relief valve, which is low cost and has simple structure to realize intake valve closing and continuously variable valve lift hydraulic valve train. Results show that the maximum valve lift increases with the setting pressure under the same engine speed. The difference of opening degree under different setting pressure is small and it gradually delays as the engine speed increases. Under the same setting pressure, the maximum valve lift increases with the engine speed, and at setting pressure of 4.5 MPa the increase trend slows down. The contribution of the mechanism inertia to the lift improvement is negligible. The Chi-square tests show that the related parameters follow normal distribution and the values of the parameters are estimated mean value and the deviation is the mean square error. The valve action has good consistency. With the increase in the hydraulic oil temperature, the maximum lift decreases, and the closing time gradually advances.

The effects of different direct injection methods of compressed natural gas(CNG) fuel into the cylinder on the microscopic physical fields, the flame propagation mechanisms and the NO formation mechanism were simulated and analyzed using software FIRE. The results show that, after injection the formation mechanism of CNG fuel mixture depends on its separation speed relative to the air in the cylinder, and the separation speed of CNG is controlled effectively by the injection timing according to the kinetic energy provided by gas fuel during the injection, which determines the distribution of concentration fields. During the compression stroke, the concentration distribution rises from the bottom to the top if the fuel was injected early and vice versa. The NO formation rate depends on the concentration and temperature of local mixture. The NO formation rate is almost 0 regardless of equivalence ratio when temperature is lower than 2000 K. The amount of NO production depends on the NO reaction rate, the size of production area and the reaction duration.

In the simulation of hydrodynamics lubrication using CFD based methods, cavitation model has remarkable influence on the simulation results. In this paper, a full-cavitation model is introduced to improve the current CFD based methods. The influences of different cavitation treatment methods on the distributions of pressure and temperature are studied. Further, the applicability and accuracy between Reynolds and CFD based methods are compared. The results show that the accuracy of the maximum pressure is improved up to 12.1%, and the overall pressure distribution is also more close to the experimental data. The temperature distribution with the full-cavitation model is more realistic. However, in modeling the bubble collapsing process, the current full-cavitation model needs to be improved when calculating the temperature distribution in the cavitation area.

To study the steam cooling in thick-wall ribbed channel, a conjugate calculation model was established on the basis of experiment data. The influence of the wall thickness on the flow and heat transfer characteristics of thick-wall ribbed channel with 3D heat conduction effect was investigated. The flow and heat transfer performance of steam in thick-wall ribbed channel for rib height-to-hydraulic diameter ratio (e/D) ranging from 0.047 to 0.188 and rib angle-attack ranging from 30° to 90° was also analyzed. Then, the relationship of Nu with wall thickness, e/D and rib angle was developed. The results show that the influence of the wall thickness on the flow and heat transfer performance of the ribbed channel is small, and the heat transfer coefficients of the channel decrease with the increase in the wall thickness. When e/D is 0.141 and 0.047, the comprehensive performance factor of the thick-wall ribbed channel reaches it maximum and minimum, respectively. When the Nu on the rib walls is not considered, the comprehensive performance factor of 60° thick-wall channel is the highest. When the average Nu is obtained from the whole walls, the comprehensive performance factor of 45° thick-wall ribbed channel is the highest.

Pressure fluctuation in high pressure common rail has great influence on injection quantity deviation of multiple injections. To eliminate the pressure fluctuation at the inlet of the injector, a new damping filter with damping hole and vessel to inhibit pressure fluctuation is studied. Through the uniform design method, it is found that the amplitude of pressure fluctuation is the minimum when the damping hole diameter is 0.4 mm and volume of the cavity is 1720 mm³. Based on this, experiments are conducted to verify the filter performance under different rail pressure and control strategy of multi-injection. Results show that, when the common rail pressures are 40, 60 and 100 MPa, respectively, the filter can decrease the amplitude of pressure fluctuation by 51% to 54% under single injection with a pulse width of 0.4 ms. In pre-injection mode, the filter can decrease the fluctuation amplitude of fuel injection quantity by 62.4% to 715. In main-post mode, accordingly, it can be reduced by 86% to 86.3%. Therefore, the filter can suppress fuel quantity deviation and provide reference to optimize the control precision of the common rail fuel injection system.

Based on the characteristics that the first failure time of motorized spindle obeys Weibull distribution, a reliability test time design model is established considering the first failure time corresponding to acceptable reliability. Then, Combined with the modified Miner linear cumulative damage law and the program load spectrum of motorized spindle, the accelerating factor model of the accelerated life test of motorized spindle is established. On the basis, the acceleration factor and the test time of the accelerated life test are quantitatively determined, and the dependability of the test time design is evaluated by calculating the prior distribution and posterior distribution of the failure rate. Experimental results show that the proposed method is feasible and can provide the basis for the scheme design of accelerated life test of motorized spindle.

As temperature increases the viscosity of the heavy machine hydrostatic turntable oil film decreases, resulting in the decrease in the bearing capacity of the table. Therefore, temperature is an important factor affecting the speed and bearing capacity of the turntable. The Reynolds equation considering the centrifugal force and the convection dominated energy equations are simultaneously established in polar coordinates. Finite difference method is applied to establish the relationship between temperature distribution and bearing capacity of the hydrostatic turntable under different speed. The established model is verified in a 5 meter vertical lathe hydrostatic bearing test rig. The results show the oil temperature increases with the rotation speed of the table; when the speed reaches 40r/min, the temperature rises rapidly and the bearing capacity decreases sharply. The oil gasket seal edge temperature rise is different, in the high speed side the temperature rise is more obvious, resulting in local high temperature region. This study may provide reference for further analysis of thermal compensation and structural design optimization of hydrostatic turntable.

In the performance testing of air compressor there exist several issues owing to the nonlinearity characteristics, such as time delay and large inertia in the formation process of the back pressure. These issues may influence the stability and reliability of the air compressor. To solve these problems, a fuzzy self-adaptive PID controller is designed to control the back pressure of the air compressor. Simulation model diagram of back pressure control system is designed on the basis of air cylinder model. Simulation results show that the fuzzy self-adaptive PID controller is better with small overshoot and shorter transition time and superior stability than the traditional PID controller.

The dimensionless similarity analysis method is adopted to calculate the dynamic similarity relationship between a three-span bridge and a small-scale test bench based on Magneto-rheological Bearing (MRB). On the basis of the relationship, the test bench model s designed. Dynamics of the two systems are calculated by Matlab respectively. Simulation results show that the acceleration, displacement of the prototype and similar models have good similarity. Further, according to the model parameters, a small-scale test bench is built, and the experimental results are in good agreement with the theoretical results. The acceleration peak and displacement peak of the bridge deck can be reduced by 42% and 36%, respectively. Therefore, the model experimental study based on similarity theory can reveal the vibration isolation performance of the controllable prototype.

Self rotating grinding with cup type diamond wheel is a typical ultra precision grinding process for silicon wafer. In this paper, a mechanical model of the micro contact between the wheel micro unit and silicon wafer was established for the stable ductile grinding process, and the mechanism of self rotating grinding was studied using force decomposition method. On the normal direction, the load and stress distribution of the corresponding elastic and plastic stages were obtained using the Hertz theory and the cavity model respectively. On the tangential direction, the sliding friction force was obtained using micro tribology theory. The load in the two directions was synthesized and the total stress condition was obtained. The analysis results were verified by corresponding experiment and simulation.

In order to realize simulation and engineering preview of pipe laying on land, a set of real-time semi-physical simulation system for pipe laying operation was established using visual simulation technology, network communication technology and dynamic modeling technology based on the actual data of “Offshore Oil 201” pipe-laying ship. The mathematic model of the initial pipeline and cable system was established considering the influence of current load and dynamic positioning system. The influence of dynamic current load on the shape and stress of the pipeline and cable during the initial laying process was discussed. The pipeline shape will shift under current load. The tension and initial length of the cable can be adjusted according to the current load in the initial pipe laying process, and the initial pipeline can be accurately lowered to the predetermined area of seabed. The real-time semi-physical simulation system can be used for engineering preview in actual initial pipe laying process. It can also be used to train pipe laying operators.

A high speed camera was used to record the locomotion of water strider to investigate its motion characteristics. The locomotion was observed in five situations, including walking on water surface without leg handling, with forelegs removal, with middle legs removal, with hind legs removal, and moving on paper. The joint angles of water strider legs were calculated, and the function of each leg was investigated based on the calculated angles. The results of linear motion on water surface show that the foreleg maintains the body balance by contacting with water; the middle leg provides driving force by paddling; and the hind leg provides support and maintains balance through its swing, and reduces water resistance. When water strider moves on paper, its middle legs provide driving force as it moves on water surface, and its hind legs swing slightly to control the balance during the caprioling and landing. The water strider can move stably through the adjustment of the rest legs after missing some legs.

Combined with the research results of bionic teeth, a three-dimensional virtual model of the working device of a certain type backhoe hydraulic excavator was built in Pro/E software. The kinematics and dynamics of the model were analyzed in ADAMS software. The parameters of the backhoe hydraulic excavator working device were obtained, such as the maximum excavation height, the maximum excavation depth, the maximum unloading height, the minimum turning radius, the maximum excavation force of the stick and the maximum digging force of the bucket. Optimization results of the parameters of the boom of the working device show that the angle between the arc outer line and center line was 65°, and the arc radius of the boom was 210 mm.

Numerical simulation and experiment were performed to study the bending-forming process of bidirectional trapezoidal sandwich panel with different deformation degrees. The deforming characteristics, regularity and main forming defects of the panel were investigated. The results show that the stress and strain of the weld region between the face sheet and the core plate are obviously less than that of the suspended area. The strain and equivalent stress of the top face sheet in the weld region are respectively 15.2% and 76.2% of that in the suspended region. The strain and equivalent stress in weld region of the bottom face sheet are respectively 11.5% and 67% of that in the suspended region. Each cell of the core is mainly deformed by changing the angle between the inclined plane and the bottom plane. Moreover, the deformation of the core cell is mostly converged in the transition curved region. The depression, straight plane and wrinkle are the main forming defects in the forming process of the sandwich panel. Face sheet thickness and deformation degree are the main factors affecting the forming defects. The characteristics of the formed part of the sandwich panel are tested by multi-point bidirectional bending test, and the correctness of numerical simulation of the forming defects is verified.

Coating layer with the same composition as the FV520B steel was deposited on the base material by laser cladding and Tungsten Inert Gas (TIG) cladding. Then, TIG cladding was tempered at 470 °C and 615 °C, respectively. The microstructure, hardness, erosion resistance of the cladding layers were investigated, and the surface quality, micro-hardness and residual stress were also tested after erosion. The results show that laser cladding with compact and symmetrical microstructure has the highest hardness, and TIG cladding with tempering has similar microstructure compared with the base material. With the increase in hardness, the erosion peak will move towards the high angle. When the impacting angle is less than 30°, the anti-erosion performance and surface quality of the laser cladding are better. At higher impacting angle, anti-erosion performance the laser cladding is worse than that of TIG cladding layer. For all the layers, erosion hardness is the lowest at the impacting angle with highest erosion rate, and erosion hardness is highest at the impacting angle of 90°. Both residual extend stress and residual compressive stress, which are caused by erosion, are maximal for laser cladding layer with highest hardness.

The side-slip angle of electric vehicle can not be directly and conveniently measured by using sensors. To solve this problem, a Square-Root Unscented Kalman Filter (SR-UKF) algorithm was proposed to estimate the side-slip angle. First, the lateral dynamics model of the vehicle and the nonlinear dynamics model of the tire were established. Then, based on the models and the estimated side-slip angle, the cornering stiffness of the tire was estimated by using least square method. Experiments were carried out to verify the effectiveness of the algorithm in estimating the side-slip angle of electric vehicle. Results show that the proposed method has high precision and can provide effective vehicle state information for the design of the stability control system of the vehicle.

In order to meet the requirement of aerial photoelectric reconnaissance system that the performance of rapid and high precision need to be processed by Fast Steer Mirror (FSM), a control method of adding Zero Phase Error Tracking Control (ZPETC) to the original PID control system of FSM is proposed. As a feedforward control method, ZPETC is specially designed for tracking control. It is able to improve the performance of the control system by cancelling the zero pole or phase error of the controlled target. Matlab simulation results demonstrate that using ZPTEC FSM can achieve ideal performance. The experiment results prove that, comparing with the conventional PID controller, the performance of the FSM is improved.

A Disease-Symptom Semantic Net (DSSN) for misdiagnosis prompt is constructed. First, symptom words are recognized from medical corpus, and a symptom ontology based on semantic relations between symptom words is established. Then, the relations between diseases and symptoms and the misdiagnosed relations between diseases were test mined and extracted to construct DSSN. DSSN contains Disease Ontology (DO), new established symptom ontology, misdiagnosed relations between diseases and differential diagnosis knowledge. Finally, a use case in clinical diagnosis is used to illustrate that DSSN is helpful to prompt misdiagnosis in clinical assistance diagnosis system.

In this paper, we propose an automatic music composition algorithm based on Long Short Term Memory-Recurrent Neural Network (LSTM-RNN). In this algorithm, we first divide music set into set which consists sequences of unit by length, and in the preprocessing we get the Mel frequency cepstrum coefficient as the feature of audio music. Second, the proposed training samples are trained and predicted by LSTM-RNN. Finally, the generated music sequences are joined to get a new music. In order to verify the effectiveness of the algorithm, we carry out an anonymous evaluation of the original music and the music generated by the algorithm. The results show that the algorithm can work well on automatic music composition.

In feature selection, with the increasing number of selected features, the relevance between candidate features and class labels is dynamically changed. This paper presents a new definition of relevance, called Conditional Relevance (CR). That is, we give a new definition of the relevance between candidate features and class labels when each selected feature is given. Consequently, we propose a novel Conditional Relevance Feature Selection (CRFS) method based on information theory. First, the superiority of the CR is verified in theory. Then, the new feature selection algorithm is compared with seven feature selection algorithms on two different classifies and on 10 real data sets. The results show the highest accuracy and the average highest accuracy of the 10 data sets on two classifiers. Experimental results show that the new algorithm can effectively improve the classification performance.

CSNB binary compression sort numbers are proposed to reduce the spatial complexity of ordering sequence, and the original sequence is restored correctly by 01 check, parity check and decompression algorithm of CSNA check. The uniqueness of the decompression algorithm result is theoretically and experimentally tested. CSNB can correctly describe any sort sequences through experimental results, and a unique solution can be obtained by CSNB decompression algorithm, which can restore the original sequence. Test results show that CSNB compression algorithm has high compression ratio for documents of integer types. Data compression can not only save data storage space, but also enhance the safety and efficiency in the transmission process.

Game theory is a network performance optimization method. For inter-cluster energy efficiency optimization based on cooperative game model with non-transferable earnings, this paper analyzes wireless channel allocation algorithms with constraint to balance the routing protocol. The impacts of game algorithm and greedy algorithm on the throughput are compared using Minimax Nash equilibrium channel allocation strategy. According to request of internet network access protocol in Mesh networks, fair routing protocol between two clusters reasonably distributes channel resource management to cluster header nodes. So each node enjoys its corresponding bandwidth weight, and gets inter-cluster fair routing and channel assignment model based on non-transferable earnings and cooperative game. The simulation results of NS3 show that this method is superior to other algorithms on throughput and effectively improves the network performance.

To solve the issues of low efficiency and less inliers of existing matching binary feature search algorithms, a Fast Calculating Bit Map algorithm (FCBM) and a Bit Map Locality Sensitive Hashing algorithm (BMLSH) are proposed. First, the bit vectors of the binary features extracted from left image is calculated and the FCBM is used to calculate bitmap of each bit vector. Second, the bitmaps of the bit vectors are taken as the key words, and each keyword and its corresponding binary feature identifier are used to construct local sensitive hash table, then the maps are stored in the hash table. Furthermore, the keywords in the hash table are stored in the bit set. Finally, the bit set is used to judge whether the bitmaps of right image binary feature exists or not in the hash table, and the query of matching binary features are optimized to improve the search efficiency and quality. Experiments show that the proposed BMLSH can improve the search efficiency and increase the number of inlier points.

Traditional DNALA algorithm has the shortcomings of low efficiency in terms of running time and unsatisfactory accuracy of the data mining results. To overcome these shortcomings, an improved DNALA algorithm is proposed based spectral clustering matrix, called DNALA-I, which can improve the time efficiency and accuracy of the results. The calculation of the distance matrix by multi sequence ratios in the traditional DNALA algorithm is improved by calculation of the spectral clustering distance matrix in the DNALA-I algorithm. Experiment results show that, compared with the traditional DNALA algorithm, the proposed DNALA-I algorithm can effectively reduce the time for sequence alignment meanwhile remain the data mining accuracy.

An end-to-end online large-scale 3D scene reconstruction method is proposed. This method uses robustness to estimate the rotation attitude of the camera and constructs a hierarchical optimization framework for the fusion of depth data input. Then, the information of each frame is optimized according to the global pose of the camera, and the algorithm limits the target tracking time and completes real-time tracking of the frame. Experimental results show that the average time to reconstruct the algorithm reaches 399 ms and the average number of estimated Iterative Closest Point (ICP) times is 20, which needs 100 ms to complete each frame transformation. The system is robust to the reconstruction of large-scale scenes and has better real-time performance. This method is a real-time three-dimensional reconstruction system with corresponding sparseness, dense structure information and camera illustration uniformity.

To solve the large delay problem of video coding by using the chaotic encryption, a fast security encryption algorithm for H.264/AVC video is proposed. First, a key generation scheme is designed. Four groups of chaotic sequences are generated with four-dimensional Quantum Cellular neural Network (QCNN) iteration, the first three groups are used for chaotic sequence pool, and the other one is used to generate the initial value of the Logistic map to increase the dynamics of the initial value. Then, to avoid repeatedly solving the high dimensional chaotic systems, the sequence of the calculation key is selected from the chaotic sequence pool according to the index generated by the Logistic map. After calculation, the encryption key is obtained, which realizes the fast generation of the key. Finally, some key elements, such as partial sign bits and information bits in video coding, are effectively encrypted for reducing the amount of encrypted data. Experiments show that the proposed encryption algorithm for 70 frames of video coding delay is less than 0.6%. Through the key space analysis, NIST SP 800-22 test, known plaintext attack and inference attack experiment analysis, it is shown that the proposed video encryption algorithm has the advantages of large key space, good random key, strong attack resistance and high security. In addition, the algorithm does not change the video compression ratio to conform video encoding format specification. By quickly generating secure keys with only encrypting the key video syntax elements, the H.264/AVC video encryption delay is effectively reduced.

To increase the speed and accuracy of vehicle recognition, an improved hierarchical Histogram of Oriented Gradient (HOG) symmetry algorithm was proposed. First, the HOG features are improved to get the hierarchical HOG symmetry features which are fused with Local Binary Pattern (LPB) features to get the fusion features. Second, the fusion features are taken as the training sample of Support Vector Machine (SVM) classifier. Meanwhile, the Principal Component Analysis (PCA) is used to reduce the dimensions for decreasing the complexity of the classifier. Finally, the SVM is used to recognize the appearance features of the vehicle. Simulation results show that the proposed vehicle type recognition algorithm can not only increase the feature extraction speed but also improve the detection accuracy, enhancing the real-time recognition of original vehicle images. The mean processing speed is about 26.2 frames/s and the accuracy is about 94.58%, increasing by 7.98% in comparison to traditional HOG feature extraction algorithm. The method can effectively increase the accuracy of vehicle recognition and also reduce the computing time caused by high dimensional features.

The proposed Bacterial foraging optimization algorithm (BFO) based on support vector machine (SVM) presents an effective intelligent early-warning model (IBFO-SVM) for the somatization disorder severity prediction for Community Correction patients. In IBFO-SVM, opposition-based learning strategy is introduced into the BFO to increase the diversity of the bacterial population and uniformly distributed as far as possible for initial population, which improves the convergence rate of BFO and reduces the probability which falls into the local optimal solution. The proposed improved bacterial foraging optimization algorithm (IBFO) is used to determine the two optimal parameters of SVM (penalty coefficient and kernel width). For comparison purpose, the original BFO, genetic algorithm (GA), particle swarm optimization (PSO) has also been proposed to optimize the parameters of SVM. In this study, IBFO-SVM, BFO-SVM, GA-SVM and PSO-SVM model were compared by 10-fold cross-validation method in psychological evaluation data. The experimental results demonstrate that the proposed IBFO-SVM method has better performance in predicting the severe somatization disorder and mild somatization disorder in terms of classification accuracy, Mathews correlation coefficient (MCC), sensitivity and specificity than other methods.

A space vector model is proposed to overcome the lack of spatial location information in the bag-of-words model. The model turns visual words into vector model using the space location information of visual words to represent image, thereby achieves better classification performance. Experiments are carried out on two standard image datasets Caltech-101 and Caltech-256, respectively, with Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) classifiers. The results show that the space vector model can effectively improve the Average Classification Accuracy (ACA) and Average Category Precision (ACP), and has a good classification effect.

The classical MUSIC algorithm based on the subspace estimation is dependent on covariance matrix estimation, which is built from a lot of snapshots. However, the number of the available snapshots is indeterminacy in the practical application. In order to solve the problem, we propose an improved MUSIC algorithm for single snapshot, called ISS-MUSIC algorithm. This algorithm improves the performance of the classical MUSIC algorithm, and it combines the method of constructing pseudo-covariance and the method of conjugate enhancement. The advantage of ISS-MUSIC algorithm is that it can be used in the case of single snapshot. Meanwhile, its performance outperforms the method of single snapshot MUSIC algorithm based on pseudo-covariance without enhancing conjugate.

Normalized Difference Vegetation Index (NDVI) time series exactly derived from remote sensing data are usually contaminated by different types of noise. Only by adopting the local or global filtering algorithm can not eliminate all kinds of noise from NDVI time series and maintain their local or global characters at the same time. A hybrid filtering algorithm, called Hyb-F, is proposed based on Savitzky-Golay filter, Asymmetric Gaussian algorithm and the Grubbs test method. Then, Hyb-F, which is a four step method, is used to reconstruct the Landsat NDVI time series for eight land cover types within Xianghai National Natural Reserve. First, the obvious outliers are detected and removed by setting thresholds for Normalized Difference Snow Index (NDSI), standard deviation and boundaries of NDVI. Second, via combing Grubbs test and Savitzky-Golay filter, the local outliers are removed. Third, via combing Grubbs test and Assymmetric Gaussian algorithm, the global outliers are removed. Finally, the final NDVI time series are built by using the Savitzky-Golay filter. The shape of fitting curves, statistical indicators and regional application effect of Hyb-F are analyzed. The results of the proposed Hyb-F method are compared with that of only Savitzky-Golay filter or Asymmetric Gaussian algorithm. For the land cover types with vegetation, Hyb-F performs well to the samples of shrub land, grassland, forest and cropland. Results of the mentioned types show large correlation with the clean reference series. The correlation coefficients are from 0.8488 to 0.9215 and the root mean square errors are from 0.0429 to 0.1057. The results of the first two land cover types achieve the highest accuracy among all the methods discussed. For the types without vegetation, such as saline, construction land and water area, the Hyb-F method can smooth the curves well. Compared with other methods, Fyb-F takes all local and global characters and dynamic trends into consideration. It can effectively recognize different types of outliers. Besides, it can weaken the loss of peak values and model the annual growing pattern of land cover types with vegetation. When applied to the whole study region, the indicator of efficiency of reconstruction is 0.59, demonstrating the reliability of Fyb-F. Moreover, the work flow the proposed method is clear and it is easy for application.

This paper presents a low power pipeline Analog-to-Digital Converter (ADC) used in CMOS image sensor. We propose a new fully differential ring amplifier, and employ a SAR ADC to assist the pipeline architecture. Comparing with conventional Operational Transconductance Amplifier (OTA), the proposed amplifier has higher energy-efficiency. This design is implemented in 0.18 μm 1P5M 1.8 V/3.3 V process. The post-simulation results indicate that it can achieve a maximum differential nonlinearity of +0.58 LSB/-0.33 LSB, a maximum integral nonlinearity of +0.55 LSB/-0.57 LSB, Spurious-Free Dynamic Range (SFDR) of 64.9 dB and Signal-to-Noise and Distortion Ratio (SNDR) of 57.8 dB for a 19.94 MHz input at sampling rate. The total power dissipation of ADC core is 3.78 mW, the FoM of ADC is 148 fJ/conv-step.