The transmission tests were performed for the rear driving axle of a certain vehicle and it was revealed that there was an abnormal vibration domain at the frequency wear 165 Hz. By the modal analysis with the finite element method and the constraint modal test at the bench, it was found that this frequency is the first order bend mode frequency of the rear axle. An optimally tuned damped dynamic vibration absorber was designed aiming at this resonance frequency. A virtual prototype of the absorber was simulated and a real sample was tested on the bench. The results showed that because the first order constraint mode frequency on the real vehicle is near that at the test bench, and the effective frequency band of the dynamic absorber is wide, the designed absorber will behave well also at the real vehicle.

An active return-to-center control method without using steering angle sensor was proposed, and it was integrated into the electric power steering system control program in the form of software to improve the vehicle return-to-center capability without additional system components. The real vehicle test results showed that the proposed method can effectively improve the vehicle return-to-center capability with a good application perspective.

The shift quality assessment indexes of the dual-clutch transmission(DCT) were defined in detail from the aspect of control. A DCT vehicle model was built using MATLAB/Simulink tools. A shift quality assessment system was designed. Through the co-simulation of the vehicle model and the assessment system, the DCT shift quality control targets, namely the expected control range of each assessment index was defined, providing the reference and guidance for the shift control strategy and algorithm. Taking the highest level of shift quality assessment as the target function, using the multi-target optimization method of the genetic algorithm, the control strategy of the clutch switch timings during the torque phase and the inertia phase was optimized, and the optimal solution sets of the clutch engaging and disengaging speeds were optimized. Based on the co-simulation platform of DCT vehicle model and assessment system, the simulation results on shift quality before and after the optimization were abtained. The vehicle test results showed that both the proposed shift control strategy and control algorithm are correct, they can improve the shift quality of DCT.

Taking a certain hydrostatic drive track vehicle as the research objective, 2 parameters, the climbing margin coefficient and the speed margin coefficient were introduced to characterize its straight driving capability at climbing and maximum speed conditions. The magnitude of the coefficient reflects the capability of the vehicle to overcome the dynamic impact and adapt to different working conditions. The suitable set of coefficients can direct the design of the track vehicle hydrostatic drive system. The research results showed that the straight driving demand of the vehicle was satisfied essentially at the climbing margin coefficient no less than 1.07 and the speed margin coefficient 1.67~1.88.

Key NVH technologies of hybrid electric vehicle (HEV) were studied to treat the more complicated noise and vibration problems. The problems came from the new assemblies which include power motor, power battery, etc., and the new drive modes in HEV. Based on the control theory of vehicle noise and vibration, the structural characteristics of HEV were analyzed and the control methods were discussed from the noise and vibration excitation source, modal distribution, transfer path, evaluation mode and other aspects. At the same time, a new indigenous HEV was presented how to analyze and resolve the NVH problems in its product development process, providing a clear working direction to control noise and vibration performance of hybrid electric vehicle.

An evaluation system was established to evaluate efficiently and accurately the sound quality of the interior noise of B-class vehicle in the steady working conditions based on the radial basis function(RBF) neural network.The subjective evaluation tests were carried out for 30 steady noise signal samples with the grade estimation method.The objective parameters having significant effect on the sound quality were extracted by the correlation analysis.An evaluation model was built to predict the sound quality of the vehicle interior noise using an RBF neural network,its average relative predict error is 4.5%.Based on the predict model, using the modular design concept and the multi-threaded parallel processing technique, a virtual instrument based sound quality evaluation system was designed.The field test results showed that compared with traditional subjective evaluation method,the designed system reduces the test time by 90%,and the quality of the evaluation is better.

According to the source-path-receiver model of vehicle interior airborne noise, the m pseudo-stochastic sequence correlative identification method of transfer function was proposed. Simulation results demonstrated that the method has excellent noise immunity performance. A m sequence generator with adjustable shift clock frequency and optional register series was designed by the complex programmable logic device CPLD as core and external circuit. The field tests indicated that compared with broadband white noise sweep method, the identification method based on m sequence technique has advantages in accuracy and noise suppression. The test error ratio was less than 1/3, and the background noise of test environment could increase by 40 dB(A).

According to the analysis of the skid-steer loader working system operation cycle, the working mechanism of self-level valve was elaborated, and the basic principle of achieving the functions of flow rate distribution ratio was discussed theoretically. Based on the comparison of experimental and simulation data, a reliable simulation model of the self-level system was established. The research proved that scaled-up diameter of the fixed and adjustable orifice is propitious to improve the dynamic characteristics of the system. In addition, decrease of the diameter of unloading valve orifice, or increase of the diameter of orifice properly, can reduce the system gain and improve the system dynamic response performance.

The pressure equilibrium equations in the hydraulic cylinder were derived for the hydraulic rolling shear according to its structure. The pressure impacts in 2 cases, the one uses a symmetric valve to control an unsymmetric cylinder, the other uses an unsymmetric valve to control an unsymmetric cylinder, were compared by solving the 2 chamber pressures before and after the commutation. The hydraulic commutation impacts in the hydraulic cylinders in both cases were simulated by software and it was found that the latter case is characterized by less impact than the former case, which is in good agreement with the results of theoretical calculation. An improved hydraulic system was designed based on the simulation results to improve the operation satety of the equipment. The collected in-situ data of 2 chamber pressures of a prototype shear proved that using an unsymmetric valve to control an unsymmetric cylinder can alleviate the commutation impact effectively.

In order to meet the requirement of micro-feeding for the hydroxyapatite cold spraying process, an ultrasonic powder-feeding system was developed. Theoretical analysis and experimental research on the powder-feeding principle and capability of the system were performed. The results showed that the feeding capacity of the system is related to the voltage and frequency of the driving power, the performance and structural parameters of the piezoelectric ceramics, the length and damping coefficient of the feeding tube. From the experiments, the effects of the factors on the feeding rate were analyzed and discussed. For the hydroxyapatite powder, the minimum feeding rate of the developed system is about 0.002 g/s and the maximum about 0.25 g/s. The powder feeding process of the system responses rapidly, meeting the micro-feeding requirements.

The laser-assisted machining (LAM) was researched to solve the processing problem of zirconia(ZrO₂) ceramics with high strength and high brittleness. A 3-D finite element model was built for the heat transfer, and the simulation of the temperature field was performed using software ANSYS. A machining test system was built and the machining tests were carried out. A Nd:YAG laser was used to heat the workpiece, an infrared thermometer was used to measure the surface temperature, and a piezo-ergometer was used to measure the 3-axis forces on-line. Meanwhile, the chip morphology, the tool wear state and the surface roughness were evaluated after each cutting. The research results showed that compared with the traditional cutting processing, LAM effectively reduces the cutting force and the tool wear, improves the quality of machined surface.

In order to facilitate the design and manufacture of the self-excited pulsed nozzle(SEPN), a structural parameter model was built by analyzing the structure of the nozzzle. An adaptive particle swarm optimization(APSO) algorithm was proposed to solve the model of the nozzle. Taking the maximum of the jet impact force as the target the structural parameters of SEPN were optimized. The performances of the model and the algorithm were validated by experiments, and the accurate boundary constraints of the model were obtained. The results showed that the built model is reasonable, the results calculated by APSO algorithm were in good agreement with experiment ones, solving the problem that the boundary conditions of the design and manufacture of SEPN were difficult to express qualitatively.

In order to process the grinding and polishing welded seam, a visual system was built on a minitype grinding and polishing robot based on the charge coupled device and structured light. The algorithm improvements were introduced to the image preprocessing technology, such as the image enhancement and single pixel treatment of binarization.The linear interpolation algorithm with the distance as weight was used for image correction to actively identify and reduce the target area, decrease the processed date amount effectively. To highlight the image target, the grey distribution maximization algorithm was introduced. The binarization and light-center extraction were integrated to speed the computation. The experiment results showed that the built system is stable, reliable with high precision and speed, and the proposed preprocessing technique and algorithms are effective.

Distributed data acquisition technology, which is widely used in seismic exploration, was introduced in electromagnetic geophysical system. The measured electromagnetic signals possess the characteristics of broadband, large dynamic range and strong interference. Low-noise analog signal conditioning circuit was designed and a high-performance data acquisition system was constructed using 24-bit Δ-Σ Analog-to-Digital Converter (ADC) and Field Programmable Gate Array (FPGA). In this data acquisition system, a microcontroller with peripheral modules was used to achieve workflow control, data storage and transmission. The test results of main technical specifications, and flume model and field experiments show that the system is feasible and accurate, which meets the needs of data acquisition of distributed electromagnetic exploration.

A transmitting device based on the magnetic-signal was designed in order to get the manual-signals in hybrid-source mag-netotellurics detection. The physical structure and the equivalent circuit method model of the transmitting antenna were discussed. The equivalent circuit model was built, and the magnetic densities in different directions were obtained. Combined with the existing experimental conditions, the rationality and effectiveness of designed device were verified. The experiments proved that the transmitting device can provide effective manual-signal with frequency in the range of 1 kHz~80 kHz which makes up the shortage of the natural-signal, and expands the exploration depth of the instrument.

The finite element model of one-third composite cylindrical shell was established using ABAQUS, and its compressive properties were analyzed. Results show that the shell failed in buckling, which is in agreement with testing results, thus validating the finite element model. Based on this model, the material, ply sequence and size etc of the cap were analyzed to study their effects on the compressive properties of the shell. This provides the basis for design improvement of the cap to fulfill engineering requirements. Computing results show that deflecting the ply angle of the cap by 25? can improve the buckling strength by 6.8%. The caps with the same composite material of the reinforced frame were batter than those made of metal. Under the condition that the weight of the components is not critical, aluminum alloy 7075 can be used instead of CFRP, which can improve the buckling strength and meet the design requirement. This research provides the reference for the design and analysis of composite cylindrical shell.

Taking the typical porous sound-absorbing material polyurethane as the base material, 3 bionic models were designed based on the multi-level organization structure and morphological characteristics of skin and coverts of the asprey, including the uni-bionic models with trapezoidal ribbed surface or backing cavity, and the dual-bionic coupled model with both features. Refering to the national standard GBT-18696.2-2002 for the measure method of sound-absorbing coefficient by transfer function method in the impedance tube, the sound-absorbing performance(SAP) of the porous polyurethane with different structures were analyzed by the finite element method. The results showed that all the bionic models demonstrate good SAP, in particular, the trapezoidal ribbed surface improves effectively the SAP of material in the intermediate and high frequency bands, the backing cavity improves the low frequency SAP, and the bionic coupled model improves SAP in the whole frequency range.

An algorithm was established for springback compensation and modification in multi-point forming (MPF) by combination of theoretical analysis, numerical simulation and forming experiment. A calculation formula for the curvature before springback was derived based on the elasticpower hardening model of the material. Using the cubic B-sphine fitting method, a springback compensation method to correct the die surface was proposed. On the basis of calculation results of die surface information, numerical simulations and forming experiments were done for MPF process. The results showed that the compensation algorithm can effectively control error caused by springback for single-curvature surface in MPF.

A nanosecond-long laser pulse beam external reshaping optical path was designed using a series of multiple optical ring-cavities. In gas laser spontaneous Raman scattering experiments, it avoids effectively the strong laser induced gas dissociation, optical quertz window components damage and the ignition of the flammable gas, so improves substantially the signal-to-noise ratio of the weak signal generated by the Raman effect. The results of numerical study and optimization design showed that the delay time of the cavity 1 equals to the full-width-half-maximum(FWHM) of the original laser beam, and the ratios between the delay times of the subsequent cavities should be the nultiples of 2, the splitting ratio of the beam splitter is 30%~50%. Using 2 optical ring-cavities, the original laser beam with pulse width(FWHM) 6.5 ns, energy 400 mJ was widened to 35 ns(FWHM) with a peak power reduction of 20% to less than 0.02 GW. The concentration measurements were performed in the gas mixture pool based on the quantification by the laser Raman spectrum.This technique can be further used in the quantitative measurement of gas mixture concentration and temperature field of the optical engine.

A vision frequency detection algorithm was proposed based on the eccentric rate vector of the edge information to deal with the feature characterization as well as classification and identification of the hybrid traffic. Utilizing the edge information extracted by the spatial-temporal context and the image highlights, an eccentric rate vector was built to characterize the features of the hybrid traffic. An extreme learning machine was used to establish a fast learning mechanism to achieve the rapid classification and identification, solving the problem that the support vector machine is difficult to achieve the real-time detection. The experiment results showed that the edge eccentric rate vector features of the hybrid traffic are distinct, the identification accuracy is above 93% while the calculation is rapid enough to meet the requirement of real-time detection.

Assuming the vehicle speed follows a truncated normal distribution ranging from a minimum speed to a free flow speed, the dispersion characteristics of queuing platoon as it moves down from the upstream intersection were analyzed from the aspect of traffic flow density. The piecewise calculation formulas of the traffic flow density function of the platoon at temporal-spatial coordinates were derived. The calculation formulas were derived for the numbers of vehicles of the platoon having passed and not passed a downstream section, and the traffic flow distribution modes at upstream and downstream sections were obtained. A platoon density dispession model was established. The dispession models at the front and rear of the platoon were also derived to facilitate the coordination in a traffic signal control system. Taking the numerical calculation about the coordinate control of signals in the successive intersections as an example, the proposed model was validated.

A model was built for the gap acceptance using the analysis method of logistic model to describe the driver behavior of gap acceptance and rejection in the upstream transition area to study the gap acceptance of merging in the expressway work zone.Merge probability of driver accepted gap was studied in gap acceptance model and 6 independent variables were selected from merging. It was found that 2 independent variables, the back relative headway, vehicle speed of being overtaken were significantly related to merge probability based on actual data and mixed step-by-step selection method. The proposed logistic regression model was tested on the fittness and predictive accuracy and the encouraging result was achieved. A critical gap calculation formula was derived from the gap acceptance model. The conclusion provided technical support for the design of upstream transition length, traffic organization and management of the work zone.

For the dynamic monitoring and early-warning of abnormal vehicle loading in the internet of things (IOT), an accurate detection device was developed based on displacement sensor with overloaded protection. Based on dynamic suspension characteristics, a data processing algorithm for measured signals of suspension deformation by the displacement sensor was compiled with the empirical mode decomposition to solve dynamic loading status of moving vehicle. A test platform was built upon FAW Jiefang heavy truck and UNO2170 industrial computer of Advantech combined with GPS and GPRS. After repeated tests, this system can accurately detect the overload status of moving vehicle with errors are less than 5% of total weight of vehicle, providing the technical support for the development of IOT.

The affecting factors on the transfer behavior were analyzed from the aspect of the traveler based on the theory of planned behavior. The main factors affecting the transfer decision-making were extracted by the method of factor analysis. Based on this, the sensitivity of each factor on the selection from different ways was defined using the mode selection method. Taking the Liaoyang city integrated passenger transportation hub as an example to analysis and the results showed that the application of the theory of planned behavior to solve the passenger transfer behavior reflects the effect of the individual difference of the traveler on the transfer mode selection, facilitates the accurate prediction of the transfer volumn among the various transportation modes in the hub, provides a scientific basis for the planning and construction of the integrated passenger transportation hub.

In order to grasp real-timely the stability of the soil slope, an evaluation analysis method was proposed based on the information of slope displacement monitor. The strength reduction method was used to describe the stability of the slope. A functional relationship between the strength reduction factor and the slope displacement was established based on the neural network algorithm using the data from the finite element numerical simulation as the train sample. By means of the function the strength reduction factor can be calculated from the monitored slope displacement to evaluate the slope stability. The effectiveness and practicality of the proposed method was verified with a group of numerical simulation data.

The conventional Bacterial Foraging Optimizer (BFO) is easy to get stuck in local minima in solving high dimensional optimization problems. This paper proposes an adaptive Bacterial Swarm Optimizer (BSO) with time-varying acceleration coefficients to solve the local-optimal problems of BFO. The proposed method, termed as ABSO-TVAC, is applied to optimize Combinatorial Optimization Problems (COPs). When dealing with COPs, according to the cell-to-cell signaling in E. coli swarm, a conversion rule between continuous space and discrete one is designed. All arithmetic operators in the velocity and position updating rules used in the ABSO-TVAC are described in a manner of set. Simulation shows that the proposed algorithm has the capability to avoid the premature problem, and it is superior to Ant Colony Optimization (ACO) and comparable to set-based particle swarm optimization.

A diagnosis method of hybrid systems based on distributed automaton was proposed, and a distributed automaton diagnosis system was designed based on hybrid band graph. Under the dynamic change of the system status, this method can generate a distributed temporal casual graph and distributed automaton according to the corresponding temporal casual graph. Corresponding diagnosis system can improve the solving efficiency by diagnosis on the distributed automaton. Distributed real-time diagnosis test on the three Tank Water Control System benchmark problems was carried out. Results show that the proposed method can deal with diagnosis problems of hybrid system by lower complexity with higher efficiency.

To reduce interference of the shared LLC on a chip Multi-Processor (CMP) architecture, we propose a Replacement policy based Reusable Data Predicting mechanism (RRDP) to select possible reused data from replaced data according to the memory access frequency characteristics of application. When some data will be replaced in LLC Cache, we filter out predicted reusable data by checking their history reference information, and store them into a dedicated on-chip memory. Our simulation results show that the proposed mechanism increases the IPC of programs by 2.9% on average and reduces the harmful replacements by 22.69%, so it can effectively reduce Cache thrashing.

Due to its semi-structured characteristic XML document implies rich term semantics in the structure information. It is inappropriate to measure XML term weight by general Term Frequency-Inverse Document Frequency (TF-IDF) approach. This is because that, when measuring the XML term weight, this approach only considers the term frequency and document frequency but ignores the term semantics implied in the structure information. A novel term weight measuring approach is proposed to overcome the above shortcoming and improve the performance. This new approach takes the factors affecting term semantics into account, such as the paths which contain terms, term frequency in a certain path, frequency of document which contains a certain path, depth of path, etc. Experimental results on several datasets show that, compared with TF-IDF and rule based approaches, the proposed approach can improve the recall, the precision and F₁-measure in the classification of XML documents.

This paper improves the subkey extendable algorithm of International Data Encryption algorithm (IDEA). This algorithm employs the pseudo-random sequence to implement the randomness of the subkey. So the attacker can not analyze the position of the initial key in subkeys, and the attacker can not ensure the position of the weak key. In the process of subkey generation using pseudo-random sequence, the linear probing rescattering is imported to avoid the occurrence that different frequencies of initial key may cause new weak key. Testable attacking experiment and analysis of security and efficiency show that such improvement can enhance the safety and efficiency.

A Spare Tensor Principal Component Analysis (STPCA) for feature extraction based on Multi-linear Principal Component Analysis (MPCA) was proposed. This STPCA can obtain spare projection matrix by transforming the eigen-decomposition problem into linear regression problem. With the help of the spare projection matrix the effect of occlusion can be reduced in the process of feature extraction. Experiments on Georgia tech and AR face database show that the proposed method outperforms the MPCA method in terms of accuracy and the robustness to occlusion.

The objectives of this paper are how to select the most effective DCT coefficients as recognition features and how to extract face features from locality and globality. A face recognition method is proposed based on Discrimination Power Analysis (DPA) and LDA-LPP algorithm. In this method, first, DCT transformation of face image is conducted. Then according to DPA, DCT coefficients are selected. Finally, integrating LDA, which can describe raw data from global situation, and LPP, which can maintain local adjacent relationship, a fusion recognition algorithm based on LDA-LPP is developed. The new algorithm can maintain both the local characteristics and global characteristics of the data. Besides, it rebuilds and lays out the raw data from the global situation. Experiments on ORL and Yale face databases show that the new algorithm can not only select the effective DCT coefficients but also improve the precision and robustness of the recognition process.

As Local Image Fitting (LIF) model is sensitive to the location of initial curve and can be easily trapped into local minimums, an active contour model based on image region information is proposed. This model uses global and local image information to segment images. Its energy function consists of four terms: local term, area term, length term and penalty term. By incorporating the local image information into the proposed model, the images with intensity inhomogeneity can be efficiently segmented. The introduction of a global indicating function in the area term can speed up the convergence of the proposed model and avoid being trapped into local minima. As the constrained level-set function of the penalty term approaches the signed distance function, the proposed model does not need re-initialization, thus the segmentation time is reduced. In addition, to segment the interested region of an image, narrow-band realization method is given for the proposed model. Experiment results show that, the proposed model is insensitive to the initial contour; its convergent speed is fast; it can accurately segment images with uneven gray intensity, and can segment the interested region of an image.

A novel method of fingerprint texture area segmentation is proposed. First, fingerprint image is normalized. Then, intensity difference between the ridge line and the vertical direction is computed as the basis to implement region segmentation. Finally, smoothing and open operation are used on the segmented template which can extract the texture area. Results of comparison with other five algorithms show that the proposed algorithm is fast and effective. It can remove the border of the fingerprint scanner, the background noise and other non-texture regions very well. The match experiment, which adds the segmentation method to the match algorithm, shows that this algorithm can also effectively improve the performance of Automated Fingerprint Identification System (AFIS).

All triangles are rotated in the plane coordinate system so that their longest edges are set horizontally. Each triangular bounding box is placed into atlas space in descending order of its height and slides along the scan lines. The triangular texture is located on the atlas through crash test of its bounding box in the label matrix, thus the texture coordinates of the triangle are gained and its texture is packed into the atlas. The best scaling factor of triangles is tested by bisection method, so that all the triangular bounding boxes just fill the atlas space. Theoretical analysis and experimental results show that the algorithm is simple and stable, and possesses the advantage of storing textures without distortion. Compared with other similar algorithms, the capacity rate of the proposed algorithm can be greatly improved, meanwhile there is no obvious increase in its running time.

Traditional instant message systems have the shortcomings of low security and inefficiency in LANs or enterprises. To overcome the shortcomings, a novel instant message system with an improved transmission protocol was proposed. The management mode of network structure co-operation of three servers was employed in this system. In addition, a unique form of binary tree P2P lookup structure was proposed in the model of multi-client meeting. Furthermore, a novel protocol based on XML, which is different from any other protocols, was developed in the new system. The new protocol can improve the transmission security. Experimental results show that, compared with other two instant message systems, the proposed system is more steady and efficient in term of location lookup rate.

In order to keep edge character and avoid manual pre-definition of the parameters of existing bilateral filtering, a steering kernel bilateral filtering is proposed, which set the parameters of steering kernel adaptively. The steering kernel is designed based on the structure tensor and object scale. The direction angle and the proportion of axial length are set by structure tensor, and the length of short axis is obtained by calculating the object scale. Hence, the projection of the steering kernel is a large-scale circle in smooth regions; it is a slender oval in edge regions, and it changes to small-scale circle in corner regions. Simulation results show that the bilateral filtering based on the steering kernel can filter out the noise effectively and keep more edge characters. The algorithm is superior to traditional bilateral filter, gray-natural logarithm bilateral and the modified bilateral filters in both subjective and objective evaluations.

The robust stabilization problem of a class of continuous-time uncertain Markov jump linear systems with partial unknown transition probabilities was investigated. In contrast with existing literature, in this study, a new system is propose, in which not all elements of the transition probabilities were assumed to be known. By fully considering the properties of the relationship between the transition probabilities, the sufficient conditions for robust exponential stabilization of the underlying systems were derived via linear matrix inequality formulation. Moreover, the proposed concept of partial unknown transition probabilities does not require any acknowledge of the unknown elements, thus, the results obtained by the system have a wide range of applications. Numerical examples illustrate the feasibility of the proposed system.

In order to control the 6-DOF installing-calibrating robot accurately, a method to determine the reverse kinematics solutions of the robot was proposed. The coordinate systems of joints were set up and DH parameters were obtained using DH method. Then forward kinematic solution was derived. The inverse kinematics equations of the robot were established based on analytical method. A solution closest to the current robot for a number of inverse kinematics solutions was chosen through the rules of regional dividing of the end-points in X-Y plane. The inverse kinematics results can be used for precise location of the end-effector and the motion planning, and providing theoretical basis for the robot trajectory planning and online real-time calculation.

Combining equal strength cantilever with Fiber Bragg Grating (FBG), a target-type FBG flow velocity sensor is developed to monitor the pipeline and the fracture flow velocity on-line. The force produced by flow works on the special shape target, then the level and cantilever structure transform the force into the wavelength shift of the FBG and the dual FBG structure is adopted to eliminate temperature effect. In order to determine the target force characteristics in the fracture and the equivalent pipeline diameter under equal force, finite element simulation is carried out using FLUENT6.3. Simulation results show that the velocity sensitivity of the sensor is 8.71?10^-4 s²/m² in the fracture at the flow section of 5 mm?2000 mm. The pipeline calibration experiments indicate that, in the measuring range of 0 m/s~1.20 m/s, the maximum error of the sensor is 0.02 m/s. The sensor is used in the drainage experiment of simple fracture water model. The results indicate the velocity increases when the inlet valve is enlarged. After the valve is turned on, the velocity in the rear drainage pipeline decreases. The trend of velocity change at the same section is in agreement with each other.

First, a coarse registration model was built. Second, after image edge extraction, Hough transform was employed to extract matched line. Virtual corner dataset was constructed on the basis of the matched lines, description of the dataset was performed, the initial homonymous points are extracted according to the constraint geometry and mutual information. Finally, RANSAC algorithm was used to obtain the final results of the matching, a great amount of high precision control point pairs are used to construct TIN for tiny facet primitive rectification. Experiment results demonstrate that the presented method can extract more matching points than that SIFT matching method can. The Root of Mean Square Error of the matching points obtained by this method is lower than 1 pixel.

As the correlation-based estimation algorithm has poor performance when CP length is short, a novel Matched Filter based algorithm is proposed. The signal recognition is realized by estimating the inter-carrier spacing because that this parameter varies in different Orthogonal Frequency Division Multiplexing (OFDM) based standards. First, the signal model is presented. Then the proposed algorithm is characterized in details. The effects of both frequency and time offsets on the algorithm are analyzed. To resolve the high complexity in parameter searching process, particle swarm optimization algorithm is introduced for simplification. Simulation results show that the matched filter algorithm has higher detection rate and better performance under multi-path channels compared with the correlation-based estimation algorithm.

In receiving and processing of wide bandwidth signals, the existing interference mitigation algorithms based on Nyquist theory are confined to the signal bandwidth and device level. To overcome this problem, an algorithm of adaptive interference mitigation in compressed domain is proposed. First, the compressed sensing technology is utilized to acquire the compressed measurements of wide bandwidth mixtures at a sampling rate that is far lower than Nyquist sampling rate. Then, based on minimal output energy criteria and space projection technology, the projected values are used to detect and mitigate the interference signals. Furthermore, the closed-form solution and the LMS realization process of this algorithm are deduced. Theoretical analysis and simulation results show that this algorithm reduces the requirement of the A/D and DSP devices; it does not need prior information about the location of the expected signal and interference signals in the wide spectrum bandwidth and has no restriction on the number of the interference signals; so this algorithm has better practicability.

To overcome the problems in muti-exposure workpiece images, such as Low Dynamic Range (LDR), more reflective surfaces and excessive saturation, we propose a method of Retinex for enhancing the LDR image order, which extends the processing of Retinex on single image to image sequence and synthesizes the information from each image in the sequence. Introducing the illumination model, the proposed method solves the drawback that the traditional algorithm did not consider the influences of the illumination condition and the exposure time sufficiently. In addition, for protecting the edge, we decompose the approximate and detail components by the Nonsubsampled Contourlet Transform (NSCT), which has the band-pass character with multi-direction. Meanwhile, we separate the illumination and reflection components from the low approximate sub-image and select the optimal component combination. The results show that our proposed method not only removes the super-saturation of some reflective panels, but also promotes the visual dynamic range availably.

To take advantage of the characteristics of KECA for hyperspectral remote sensing image classification, an approach of sample set selection and C-means classification is proposed. The sample selection is based on convex geometry concepts and C-means classification uses spectral angles as distance metrics in feature space. Experiment results of HYDICE hyperspectral data confirm that the proposed approach can improve classification accuracy effectively.

The standard gradient-based edge detectors react to all luminance changes, irrespective of whether they are due to the contours of the objects represented in a scene or due natural textures like grass, foliage, water, and so forth. To improve their performance in detection of object contours and region boundaries in natural scenes, an algorithm called switching surround suppression was proposed which can be easily incorporated into them. Different from previous methods that deploy surround suppression, our method only exerts suppression on texture edges while leaving contours unaffected, which further improves the contour detection performance of standard gradient-based edge detectors. The proposed method was evaluated by the natural images and the standard reference contour edge images. The results showed that the proposed method is better than the standard gradient-based edge detectors as well as other methods that deploy surround suppression.

In this study, the ginseng β-AS gene RNA interference (RNAi) plant expression vector is built through the Recombinant PCR reaction and restriction enzyme digestion connection. Then,by heating, the β-AS gene RNAi is transformed into the Agrobacterium A4 to get β-AS RNAi plant expression vector engineered bacteria. This lays the foundation for further induced transformation of ginseng explants.