The parameters of driving intention identification were defined and the membership functions of the defined parameters and the fuzzy inference rules were established. The driving intentions were identified by fuzzy inference. The hybrid electric vehicle control strategy was formulated based on the driving intention identification results and its simulations were done. The simulation results show that the fuzzy inference system identifies well the driving intentions and the control strategy based on the driving intention identification improves the fuel economy of the hybrid electric vehicle.

The integrated control for active steering (AFS), direct yaw moment (DYC) and active suspension (AS) of X-by-wire electric vehicle was studied based on the model predictive control theory to enhance the vehicle handling stability. Adopting the hierarchical integrated control structure, a model predictive controller was designed. The driving torque control allocation rules and AS control method were studied and the horizontal integrated control of AFS/DYC and the overall integrated control of AFS/DYC/AS were achieved. A simulation test was carried out to verify the proposed algorithm. The results show that the algorithm can make the vehicle follow the expectation effectively and enhance the vehicle stability in extreme conditions and its active safety.

To discuss the operation of the 6-wheel articulated electric wheel truck(6WAEWT) and analyze its control strategy, considering the structure feature of 6WAEWT, a reference model of the 6WAEWT was derived from the 4-wheel Ackermann model. The driving state of the 6WAEWT was judged by monitoring the working state parameters of the wheel motor, and the control strategies at steering and slipping were designed.

The online and offline parameter identification approaches were built with extended Kalman filter (EKF) and the least square algorithm respectively based on the proposed electro-chemical polarization (ECP) model for the lithium-ion power battery pack used in the electric vehicle. Validation results based on the hybrid pulse power characterization test showed that the parameter identification approaches with the ECP model can ensure the maximum relative error within 1% and accurately simulate the dynamic voltage behavior of the power battery pack. By using the proposed parameter identification approaches, the operating efficiency of the battery management system can be greatly improved becacuse the time-consuming, laborious, even error-prone experiments and periodical calibration for model parameters before the first operation of the battery pack are avoided effectively.

A forward vehicle collision warning and avoidance system was proposed based on the electronic hydraulic brake device to prevent the rear-end collision. The system real-timely judges the dangerous level of rear-end collision with a time-to-collision algorithm based on the running situations of the front vehicle and this vehicle. When the collision possibility is detected, the system warns the driver with vision and hearing warnings and achieves the vehicle active collision avoidance by the electronic hydraulic brake. An experiment platform was established on a real vehicle to verify the brake pressure control effect. Introducing the system in a test vehicle, the test results show that the system provided timely the stepped collision warning and the brake assistance, enhancing the vehicle driving safety.

In order to control and optimize the frontal crashworthiness of the passenger car in the conceptual design phase, taking a B-class car as an example, a method of setting the important parameters G₁ and G₂ for the target simplified 2-step impact pulse in the conceptual design phase was proposed. The sub-structure performance targets were decomposed effectively under the overall goal condition based on the energy management technique, providing a sufficient basis for further structure sections design and overall goal achievement.

Taking the 2-column type rollover protective structure(ROPS) of the loader of type ZL80 as an example, based on the dynamic simulation and test, according to the requirements of the international standard for ROPS, ISO 3471, a ROPS design method which focuses on minimizing the human body injury was proposed. The design method analyzed the movement and damage of the operator and vehicle in the way of accident reconstruction, and on the basis of this analysis, the ROPS security design was performed. The results show that the ROPS meeting the static test standard could not guarantee the safety of the operator wearing a seat belt. On the premise to assure the human body living space in the dynamic rollover process, the smaller the stiffness of ROPS, the smaller the damage value and damage probability of the dummy. For the 2-column type ROPS, the column and outrigger should have appropriate stiffness.

In order to enhance the heat transfer performance and inprove the temperature field in the engine cabin to keep the vehicle performance stable, a 3-D physical model was built inside and outside the engine cabin according to the system drawing. A heat exchanger model was used instead of the radiator module in the engine cabin, the CFD numerical analysis was used to simulate the heat transfer performance of the engine cabin in a virtual wind tunnel at standard working conditions. The temperature fields and flow fields in the radiator, inside and outside the engine cabin were obtained. The comparison between the experiment data and the simulation results shows that the discrepancy is acceptable, proving the effectiveness of the model and the accuracy of CFD numerical analysis.

The geometry of contact deformation for the rolling elements of angular contact ball bearing was described by coordinate transformation method. The analytical expression for the stiffness of angular contact ball bearing was derived according to the Hertzian elastic contact theory. The influences of preload, contact angle, etc. on the bearing stiffness were analyzed theoretically. A test bench was developed to measure the bearing stiffness. The comparison was carried out between the theoretical analysis and the experimental results.The results show that the discrepancy between the theoretical calculation and test data was less than 10%, meeting the requirement of engineering calculation.

In view of the problem that it is difficult to determine the difference degree of reliability levels between the evaluation object and similar products when using similar comparison method to predict the reliability level of NC machine tools, a reliability prediction method introducing interval analytic hierarchy process (AHP) was presented. The relationship between the whole machine and its subsystems was analyzed, and a reliability prediction model was built for the whole machine. The reliability data of similar products were used in the course of reliability prediction for subsystems. The differences between the evaluation object and the similar products were comprehensively analyzed and an evaluation model for reliability correction factor was built. The reliability correction factor was obtained by using the interval AHP, realizing the mutual complement of certain and fuzzy informations. The application of this method was illustrated with an example, verifying the feasibility of the method.

To support effectively the production process quality management and its continuous improvement, the relative and absolute evaluation concepts were proposed combining the design output quality and the task execution quality, and considering the influence of the design input on the design task quality. The trapezoidal fuzzy numbers were used for the representation of the heterogeneous informations. Using the subjective fuzzy weights as constraints, taking the minimization of the weighted distance between the individual and the group decision makings as the optimization target, a decision maker weight adjustment model was built. The concrete steps of the fuzzy group evaluation of the design task quality were given based on the TOPSIS method. The calculation examples show that the suggested evaluation method avoided effectively the excessive weight adjustment, identified the weak links of the process.

In view of the limitations of traditional macro-mechanics in the surface coating study, the continuum damage mechanics(CDM) was introduced into the finite element analysis of the surface coating. A material model was built based on CDM, and it was embedded in the solver of the software ABAQUS through the user defined subroutine VUMAT of ABAQUS as the interface. A series of numerical simulations were carried out to analyze the mechanical behavior of the coating. By means of the proposed model not only the stress distribution in the coating could be obtained, but also the damage evolution of the coating under loading could be predicted. By studying the relationship between the reaction force and the coating thickness, a concept of a special reaction force corresponding to the crack penetrating the coating thickness was proposed to evaluate the load carrying capacity of the coating. Besides, a critical thickness derived from the relationship between the coating thickness and the crack depth as a reference for coating thickness design.

The cooling problem of the transition segment of the gas turbine combustor (TSGTC) was studied. The diameter of the cooling hole was chosen as the optimization variable for the size optimization of TSGTC. The quadric regressive orthogonal-composite design of experiment was used to rationally allocate the test points. The least-square method was used to construct the response surface models of the combustor wall temperature and the cold air outlet pressure to the cooling hole diameter. The adaptive response surface methodology was applied to realize the optimal design of the response surface models. The results of numerical simulations provided a valuable reference for form and cooling design of TSGTC.

The effects of methanol injection timing, ignition timing and excess air ratio on formaldehyde and unburned methanol emissions from a spark-ignition direct-injection methanol engine under homogenous combustion mode were investigated experimentally by means of a measurement method in which the gas chromatography and the liquid chromatography were used to separate and measure formaldehyde and methanol. The results show that the methanol injection timing, ignition timing and excess air ratio affect significantly the formaldehyde and unburned methanol emissions from a spark-ignition direct-injection methanol engine. The variations in emitted formaldehyde and unburned methanol show opposite tendencies with the variations in the methanol injection timing, ignition timing and excess air ratio. Retarding methanol injection timing, advancing ignition timing and using lean mixture decrease formaldehyde emission.

An experimental multi-droplet stream creating system was developed for the hydrocarbon fuel based on the ultrasonic atomization and the carrier-gas transport techniques. The speed and global equivalence ratio of the stream could be controlled independently to simulate really the flow and mixing conditions of the droplets at the spray fringes. The stream scattering light was photographed by a high-speed camera at 90? angle using a high intensity LED light source under different air flow rates(10,20,30 and 40 L/min) and different n-heptane supply rates(150,200,250 and 300 mL/h). The pictures showed that there was a stable flow near the nozzle exit and its length increases firstly and then decreases with air flow increasing because of the combined effect of air entrainment and droplet vaporization. When the stream speed at nozzle exit was lower than 5.2 m/s, appeared the obvious air entrainment eddies after the stable sector, and the lower the speed, the more eddy number and the more regular the eddy form. When the stream was far away from the nozzle exit, the multi-droplet stream might not be shot by its scattering at 90癮ngle because of the droplet diffusion and vaporization.

A simulation model was built for the thermodynamic system of the ground source heat pump (GSHP) based on the G-function theory of the underground heat exchanger (UGHE), and an analysis method for the long-term operation of the GSHP system was proposed. Taking the severe cold region as example, the system winter heating operation conditions in the last 3 years were studied. The fluid temperature variations at the inlet and the outlet of the UGHE, the coefficient of performance and the energy consumption of the heat pump and the system were analyzed simulationally and the results were compared with the measured data. The results show that the model and method are feasible, providing an effective approach for further forecast and analysis.

Taking a wind-solar hybrid power system for the off-net household in Sunite Youqi, Inner Mongolia as an example, the average daily generating capacity of the wind turbine, the photovoltaic generator and the whole system in each month were tested for analysis. The results show that the system still can provide users with sufficient electricity from July to September when the generating capacity of the wind turbine is the lowest, so it does not need to enhance the photovoltaic modules. It can be concluded that the matching of the system is good, the design of the system is reasonable.

To deal with the deficiencies of the existing prediction models based on dynamic recurrent neural network, considering the characteristics of the traffic flow itself such as complexity, nonlinearity and uncertainty, a short-term traffic flow prediction method using Elman neural network with variable gain was proposed. The real-time updates of the network were achieved by introducing a variable gain factor based on the real-time error analysis. The method was verified by the field measured data from Renmin Street, Changchun City. The results show that comparing with the existing Elman network based models the proposed method is superior on the aspects of convergence time and prediction accuracy.

To reveal the vehicle queue formation and dissipation behavior on a road section with a single lane, the similarity between the growth and dissipation process of vehicle queuing and the chain reaction process of nuclear fission was described. According to the theory of nuclear chain reaction, a chain reaction model system of vehicle queuing on a road section with a single lane was built, including the vehicle queue formation model, the vehicle queue dissipation model, and the vehicle queue model for signalized intersections. By using the field survey data from a typical road section in Dalian city, vehicle queue formation and dissipation models were validated. Also, the numerical simulations for traffic streams under different conditions were made by applying MATLAB software. Based on the simulation results, the sensitivities of the propagation speed of block and dissipation signal to arrival flow rate and running speed were further analyzed. The results show that the proposed models were characterized by high precision, which can be used to calculate the propagation velocity and boundary of the chain reaction of vehicle queuing, and explain the vehicle queue behavior on a road section with a single lane. The research achievements provide a foundation for explaining the vehicle queuing behavior on a multilane road section and the chain reaction mechanism of road jam, give a theoretical basis for solving urban traffic congestion and jam problems and offer a novel idea and a method for studying traffic queue.

To boost the rationality of the decision-making for the intermodal hub project construction scheduling and maximize the social benefits of the project investment, a method was proposed to optimize the construction scheduling for the intermodal hub planning. The 4 importance indicators including the location importance indicator, the passenger flow importance indicator, the node importance indicator, and the comprehensive importance indicator were defined, and the last indicator is equal to the weighted sum of the former 3 indicators, and it is the key factor to optimize the project construction scheduling of the intermodal hub by the dynamic optimization process. A case study in Beijing City was provided to demonstrate the procedure of the proposed method.

Complicated process of order-to-delivery in automobile manufacture enterprise deepens the difficulty in mathematical modeling and solving. Therefore, a simulation test method was proposed to analyzed quantitatively and evaluate different complex order-to-delivery modes. Simulation models of information processing and product delivery were built for 3 order-to-delivery modes, i.e. locate to order (LTO) mode, build to stock(BTS) mode and build to order (BTO) mode. By establishing the mechanism of matching between the order and the product and introducing a control variable and evaluation indexes, the dynamic monitoring and simulation evaluation of the system operation parameters were realized. Simulation results showed that LTO mode is characterized by better stability and adaptability in the inventory control than modes BTS and BTO. LTO mode also provides the selectivity among multi customer order decoupling points and is superior in the informaion processing mode and speed.

A multi-support motion generation program was developed in Fortran based on the power spectrum input according to harmonic composition method. Shanghai Minpu Bridge, which is the longest double-deck long-span highway cable-stayed bridge, was taken as example to study the seismic response of the bridge under multi-support excitations. A fine finite element model with beams and shells was built on the basis of platform Abaqus. The eigenvalue analysis and the time history analysis under nonlinear multi-support excitations were performed based on the geometrical nonlinear static analysis to get the dynamic characteristics as well as the variations of the internal forces in the main towers and the subsidiary piers and the axial forces of the web members (vertical, oblique and squint web members). The results show that the lateral shear force in the pier column below the subsidiary pier beam distribute uniformly. It increases with the distance from the main tower, so does the lateral moment. The axial forces of the web members of side span between upper and lower decks vary significantly, especially the vertical and oblique members near the pier top. The axial forces in the squint web members are well distributed and those at the pier top and at the mid-span are relatively small. The axial forces in the vertical and oblique web members at mid-span are small, and those in the other web members vary little.

The limitations of the mainstream method for the sieve analysis based on the digital image processing were analyzed and the advantages of industrial X-ray CT technique were pointed out. 5 specimens were produced according to the coarse aggregate size range of asphalt mixture. The volumetric parameters of the coarse aggregate were obtained by industrial CT scanning. The accuracies of the 3 methods for calculating the volumetric parameters were compared. The 3-dimensional surface contours were characterized by the spherical harmonic transformation, thus the L, W dimension positions were got, and the control planes passing the sieve were intercepted. The size of the square sieve that the aggregate particle can pass was determined by a series of digital image processing techniques. The validation results show that by means of the new method based on CT scan images the 3-dimensional virtual sieve analysis of the coarse aggregate could be realized more accurately.

The Ni-based WC composite (Ni-WC) coating on mould steel 2738 was prepared by CO₂ laser cladding. The substrate 2738 steel and the Ni-based WCCC were tested under dry sliding friction conditions to investigate their friction and wear behavior. The wear volume of specimen was measured using a 3-D surface profilometer, and the worn surface topography was observed by the scanning electronic microscopy. The results show that the microhardness of the Ni-WC coating was enhanced significantly, its surface hardness exceeded 1200 HV, ensuring its wear resistance. The average friction factor was about 0.25, it decreased by about 44% in comparison with that of the substrate 2738 steel of 0.43. WC hard phase enhanced the bearing capacity of the friction surface, reduced the wear rate by 96.7%. The abrasive wear was the main wear mechanism for the Ni-WC coating.

Through the theoretical analysis, the forming loading displacement method for the spherical part was studied based on the flexible roller stretch forming device. Using geometric analysis method, the relationship between the displacement of each roller and the elongation of the metal sheet was established for spherical part forming. Simulation analysis of the material of 2024-O was performed by loading displacement, the result was compared with the analytical result of the force loading, it was found that the distributions of both the equivalent stress and equivalent plastic strain gained by the loading displacement are relatively uniform. Utilizing the flexible roller stretch forming device to form the spherical part,the test results show that the displacement loading method is feasible. Comparing with the force loading, the geometric annalysis method could quickly calculate the required displacement of each roller for the load, saving enormous debugging time, providing some guidance for the stretch forming of the skin.

The loading speed of the connecting rod fracture splitting(FS) machine is one of the important factor affecting the FS quality. In order to investigate the FS machine loading speed characteristics to realize the optimization of loading speed, the virtual prototype simulations were performed. A mechanical and hydraulic model was built for the FS machine using ADAMS/View and ADAMS/Hydraulics modules. The control system models were built for the constant input control system and the input control system with displacement feedback signal. The simulation parameters were set, and the dynamic simulations were accomplished. The loading speed characteristics were analyzed based on the simulation results. In the constant input simulations, it was found that the FS cylinder loading speed 118 mm/s was good for FS quality control as well as for the hydraulic system design. In the displacement feedback input simulations, it revealed that the elliptic input control curve made the most obvious deceleration during the latter half of FS cylinder stroke. The actual loading speed of FS cylinder can be controlled according to the above mentioned curve.

Motivated by the requirement of the geographical information system and ambient intelligence, the Semi-Qualitative Constraint Satisfaction Problem (SQCSP) is proposed, and some preliminary results are obtained. The SQCSP of interval algebra and its extension model can be determined Qualitative Constraint Satisfaction Problem (QCSP). A polynomial-time decision algorithm of RCC5 is given. The SQCSP of RCC8 is a Non-deterministic Polynomial time (NP)-complete problem. A polynomial-time decision algorithm with constriction is given. The correctness proofs and instance algorithms for the above problems are provided. Finally, QCSP with boolean operations on variables is solved using SQCSP algorithm.

In light of the idea of clustering, the model of P2P traffic identification was constructed by Support Vector Data Description (SVDD). First, Principal Component Analysis (PCA) was introduced to reduce dimensions of the data set. Then, the classified model was established by support vector samples, which come from the minimal sphere by the principle of SVDD. Finally, the class of the test samples was determined by the minimal distance between the center of sphere and the sample. The model is simple and suitable for P2P traffic identification environment. It overcomes the shortcoming of the current research based on machine learning in model complexity and data unbalance. Experiment results show the high accuracy and reliability of the proposed model.

According to the various characteristics of light source that can be perceived by the audience, the whole domain of stimulation is divided into multiple parts. Each part is mapped to the perception scales by means of the effect of visual staying phenomenon. Thus, in this paper the Weber-Fechner's law is improved first. Then, a mapping model is proposed on the MOS and loss ratio of the macroblock. The model calculates the minimum distortion and undamaged area of which audience awarded by standing at the best observation position ahead of the monitor. It achieves an evaluation of the full field damage by not only reflecting the effect of impairment, which can visually stay in the eyes, but also indicating the spatial extent and temporal propagation of loss. Experiments show that the proposed method can be deployed in any network nodes for its low computational complexity and achieve better video quality in close touch with the audience.

This paper proposes a simple distributed topology control algorithm named Minimum Transmission Power based Topology Control (MPTC). It computes whether there is a multi-hop path between two direct communication nodes with the energy cost across each hop less than that of the two nodes. MPTC reduces the energy consumption while maintaining the network connectivity without the use of location-based information. A simulation-based comparative analysis is presented. The results show that the proposed MPTC has the advantages such as balanced energy consumption and good robust properties.

In order to fit the dynamic variation of the Media Oriented Systems Transport (MOST) network environment, a connection management strategy is proposed based on MOST network management. This strategy updates the central registry dynamically and supervises the source ports and the destination ports of the streaming data in the MOST network. Moreover, it implements the establishing, releasing and switching manipulations of the multi-synchronous channel connection. With the functional verification using the OptpLyzer, a network analyser for MOST, it is proved that this strategy improves the usage of MOST bandwidth effectively. The experimental platform is a ring network of MOST50.

Focusing on the problem of rendering physically correct soft shadows, a method is proposed which uses occlusion masks to evaluate the visibility of an area light source. Prior to the rendering process, a cube map of ambient occlusion masks is computed. In the rendering process, the cube map is employed to compute the ambient occlusion masks of triangles intersecting with a rectangular pyramid, which is used as the bounding volume for light-source sample rays in the visibility testing. Then the "and" bitwise operations are carried out on these ambient occlusion masks. Finally, based on the results of the bitwise operations, the blocking-distance testing is further utilized to determine which light-source sample rays are not blocked. Compared with the lighting computation method via Monte Carlo integration, the proposed method can not only render physically correct soft shadows, but also avoid a large number of intersection-testing operations for shadow rays. This advantage can enhance the efficiency in solving the rendering equation.

The traditional methods for determining optical flow are not robust under variable illumination, and do not work well at object edges. To overcome these shortcomings, a pixel-filtration based method is proposed. The method defines a new basic equation under variable illumination, and uses an improved local flow technique. First, this technique selects proper pixels twice from the local window; then, it determines the flow using the weighted-least-square method based on the selected pixels. Experiment results under variable illumination and noise demonstrate the validity and robust of the proposed method.

Data sealing can not achieve asynchronous sealing and unsealing operation in application layer multitasking environment. To overcome this problem, we propose an improved sealing approach for trusted computing, in which a new register PCR_R is introduced into TCM. It uses the metrics reversible extending scheme to guarantee the state of platform in application layer unanimous. The improved sealing approach also uses the real-time validation of trusting state provided by Root of Trust for Reporting (RTR) and Stored Measurement Log (SML) to solve the asynchronous sealing and unsealing operation in application layer multitasking environment.

A volumetric reconstruction method with multiple photo-consistency constraints is proposed in the multi-view construction environment. First, the visibility determination based on the distance between the voxel and the viewpoint is introduced to calculate the photo-consistency precisely. Then, multiple photo-consistency constraints are applied to eliminate those voxels which do not belong to the shape of the reconstructed object. The maximization of the photo-consistency constraint and the maximization of the neighbor photo-consistency constraint can be jointly used to reconstruct the effective shape in order to solve the "hole" problem induced by the unsuitable threshold of color-consistency. Experiment results demonstrate the effectiveness of the method in guaranteeing the completeness of the reconstructed shape by comparing with conventional methods.

The Discrete Element Method (DEM) analysis model of a corn thresher was established by using its CAD model. The analysis model of corn ears was also established with particle agglomerate method. The contact force in threshing process was calculated using the contact and connection mechanics model. On this basis, a new method and software were established to analyze the corn threshing process. The feasibility of the new method and software was validated.

In order to implement the path following control of wheeled mobile robot with non-holonomic constraint, the tracking error equations are established according to the virtual target method. Based on Lyapunov stability theorem, the backstepping method is designed using feedback gain technique. By tuning of the controller's feedback gain, the nonlinear terms in the error dynamic model can be compensated without direct use of nonlinear cancellation. This can also avoid the existence of high-order derivative of the predefined virtual control variable which is different from the traditional backstepping technique. Simulation results show that the parameters of the controller are easy to be adjusted, and can make wheeled mobile robot to track the desired arbitrary path precisely.

An omnidirectional walking mechanism which synthesizes legged robot and steering-wheel was proposed, the specific motion features and steering performance were introduced. To achieve a good result of motion control, equivalent manipulator model of the steering mechanism was built, and the dynamics were developed by means of Lagrangian dynamic approach. The motion planning and experiments on the prototype validate that the omnidirectional robot with a steering wheel can be manipulated simply, flexibly and accurately.

In order to achieve wideband spectrum sensing under Alpha stable noise, generalized power spectrum density is proposed, based on which the distributed compressive wideband spectrum sensing is put forward. This method can effectively restrain the Alpha stable noise. First, the received wideband signal is processed by fractional lower order, then, the autocorrelation vectors are calculated. Combined with the compressed sensing theory, the fractional generalized power spectrum density is derived by Fourier transform. Simulation results show that the method performs well in the estimation of the wideband power spectrum density, and can achieve wideband spectrum sensing under Alpha stable noise.

A novel dynamic spectrum access technology based on multi-features model clustering is presented to improve transmission quality for different users in the cognitive network. After the multi-features models of the channels are extracted, these channels suitable for the cognitive users are clustered into different spectrum poolings using a multi-objective clustering algorithm. Those channels in the white pooling are prior to be accessed for cognitive users. Multiple feature combinations and dynamic spectrum simulations are carried out to verify the efficiency of the proposed method. Analysis and simulation results show that the channels can be classified efficiently according to their performance. The access efficiency can be improved by the proposed method.

The periodic and odd-periodic quadriphase mismatched filtering arrays are proposed in this paper. Existing conditions of periodic and odd-periodic quadriphase arrays and two construction methods, folding and periodic multiplication, are presented. With such construction, as the quadriphase array size increases, the loss, with respect to the periodic or odd-periodic correlation, converges to zero. The periodic and odd-periodic quadriphase arrays can be applied in an actual system, and the influence of the sidelobes on the system can be overcome.

Large amount of data and less efficiency of stereo matching algorithm are far from meeting the needs of real-time transmission. A Fast stereo matching algorithm used in stereo video coding based on video plus depth is proposed. It utilizes motion information generated in the reference view coding as compensation in depth coding to reform disparity estimation in the stereo matching process. The information will be reorganized as the disparity estimation of Frame P for the subsequent iteration. Full use of the existing coding information reduces the complexity of disparity map generated in Frame P. Experimental results demonstrate that the proposed method can double the speed of generating the disparity map.

This paper proposes a fast algorithm of ECG denoising and QRS wave identification based on wavelet lifting. On the basis of wavelet lifting, the weighted threshold shrinkage method is introduced to ensure not to lose useful ECG information and improve the denoising effect. Using the intermediate result of the denoising and reconstruction together with the simple finite difference method, the proposed algorithm uses the first derivative of the smooth function to process the signals by the lifting wavelet transform. Such process can avoid the secondary operation of the lifting wavelet transform, thus significantly reducing the complexity of operation meanwhile maintaining the identification precision. Experimental results demonstrate that the proposed algorithm can achieve relatively higher SNR and lower MSE. In addition, the accuracy rate of QRS wave identification is above 99.5%. Moreover, this algorithm can be realized on the hardware platform of FPGA, which is convenient for ECG monitoring equipment integration.

Because Photoplethysmography (PPG) signal is disturbed by breath, position and movement, it is necessary to detect and remove disturbed signal before computing the user's vital sign. Based on Euclidean distance, pulse wave transit time and coefficient of the pulse waveform, this paper presents a parameter, called associate factor of pulse wave, which is used to detect abnormal pulse waveform. The PPG time serious is divided into pieces using R wave of Electrocardiogram (ECG) signal; the time line of each piece of PPG signal is normalized in order to inhibit sensitive issue of time line in Euclidean distance calculation. The accuracy and feasibility of the proposed algorithm in identifying abnormal waveform are verified by analyzing measured pulse wave signal.

This paper introduces an image based system which is use to identify and count bacteria in food. The image feature of bacteria is analyzed, and a method to extract the feature is proposed. The method can identify bacteria from the image taken from camera that is fixed on the eyepiece of the microscope. It is also immune to noise that comes from the high-path of the microscope and the measurement of the camera. The proposed method is tested and result shows good performance.

Conventional Super-Resolution (SR) methods have some limitations. First, most of existing SR algorithms can not cope with local motions and hence not suitable for most video sequences. Second, the blurring operator is assumed to be known in advance and be a constant for all the low-resolution images. Finally, SR noise is assumed to be either Gaussian or Laplacian. To overcome these limitations, a general local cost function is proposed that consists of weighted L1-and L2-norms considering the SR noise model. In this function, the weights are generated according to the error of registration and noise distribution, and the inaccurately registered parts of the image are penalized. Both the super-resolved images and blurring operators are estimated at the same time. Both objective and subjective evaluations demonstrate the effectiveness of the proposed algorithm.

Generating the configuration contexts is one of the key challenges in the design of task compiler for reconfigurable media processors. In order to reduce configuration cost, a novel configuration contexts generation method is proposed. The sub-graph searching and graph isomorphism algorithms are introduced to obtain graph attributes, and the constraints of generating congruent configuration template are add to promote the configuring performance. The proposed configuration contexts generation method was used in the design of task compiler for reconfigurable media processor. Experimental results show that the proposed method can generate the configuration contexts effectively and the configuring performance is significantly higher than that of traditional greedy generation method.

To improve the measurement accuracy of beam quality M² obtained by artificial setting threshold method based on experience or floating threshold method,a double gate adaptive threshold method was presented. Experimental results indicate that beam quality M² factor relative error of proposed method is the lowest among three methods,and its standard deviation (STDEV) value is less than 0.1. The STDEV value of beam width in x direction and y direction are 0.0935 and 0.0962, respectively, which are less than 1%,satisfying the measurement requirement of beam quality M².

On the basis of single factor test, the orthogonal rotary test was employed to optimize the process of sialic acid isolation from hen eggshell membrane. Four factors: temperature, time, pH value and liquid-to-material ratio, were investigated in order to achieve best experiment results. The order of the importance of the factors was temperature, time, pH value and liquid-to-material ratio. The optimal values of the factors were: temperature of 83℃, time of 2.1 h, pH value 1.4 and liquid-to-material ratio of 101 mL/g. Under this condition the yield of sialic acid was 85%. This procedure is steady and efficient. The experiment results were consistent with the model prediction.