交通运输工程学报

Test methods of airport pavement and subjacent foundation void

ZHANG Yu-hui, ZHANG Xian-min

Abstract: The test results of heavy weight deflectometer(HWD), ground penetrating radar, surface wave and image were compared and analyzed contrastively, and the bottom voids of airport rigid pavement slab, foundation subsidence and void cavity deformation were identified.The spectrum characteristics at different void depths and positions were compared, the effectiveness and superiority of each pavement void test method based on stress wave and electromagnetic wave were studied, and the test accuracy was verified by field core-drilling test.Verification result indicates that the test data of traditional HWD are pavement surface displacements under load, and do not reflect the foundation subsidence and void deformation under pavement slab.The subsidence and void deformation in soil foundation can be reflected by surface wave test data, because the test depth is greater.The bottom void cavity under pavement slab can be determined by analyzing the properties of return wave from pavement slab bottom by using image test method.But the interval sampling is adopted by above three methods, field test is limited by the number of sampling points, which results in that the runway integrity representativeness of test data is insufficient, and the evaluation error from sampling interval can not be overcome.The continuous and large scale test with ground penetrating radar can beconducted, the test coverage is 100%, the coincidence rate of void between radar data and HWD data is 95%, and the test accuracy of void in hole bottom is 87.5% by core-drilling test with 8sample points.However, the quantitative analysis on void with radar is insufficient, surface wave test and image test are proved to be effective supplement and verification to pavement void testing and soil foundation cavity testing with radar.More>

2016, 16(6): 1-11.

FullText HTML (130) PDF (5894KB) (2919)

Cyclic shear dynamic properties of geotextile-sandy soil interfaces

WANG Jun, LIU Fei-yu, WANG Pan, GENG Xue-yu

Abstract: A series of cyclic direct shear tests of soil-reinforcement interfaces were performed by using a large-scale direct shear device.Woven geotextile and nonwoven geotextile were used as reinforcement materials.Chinese ISO standard sand was used as soil mass.When the vertical stresses are 30, 60 and 90kPa respectively, the cyclic shear displacement amplitudes are 1, 3and5 mm respectively, and the sandy soil densities are 22%, 52% and 75% respectively, theirinfluences on the cyclic shear properties of soil-reinforcement interfaces were studied, and the development laws of peak shear stresses and the relationships of shear stresses and shear displacements in the processes of cyclic shear tests on two kinds of geotextile-sandy soil interfaces were analyzed.Study result indicates that the cyclic shear softening phenomena appear on woven/nonwoven geotextile-sand interfaces, and the softening laws are different.When the vertical stress increases from 30 kPa to 90 kPa, the peak shear stress of woven geotextile-sandy soil interface increases by 72.9%, and the peak shear stress of nonwoven geotextile-sand interface increases by 167.5%, so the influence of vertical stress on the cyclic shear properties of geotextile-sandy soil interface is obvious.When the shear displacement amplitudes are 1, 3and5 mm respectively, the peak shear stresses of woven geotextile-sandy soil interface are 25.9, 27.9and 29.8kPa respectively, and the peak shear stresses of nonwoven geotextile-sandy soil interface are 21.8, 23.8and 22.6kPa respectively, which shows that the peak shear stress of woven geotextile-sandy soil interface increases with the increase of shear displacement amplitude, while the peak shear stress of nonwoven geotextile-sandy soil interface firstly increases and then decreases.Under the three sandy soil densities, the differences among the peak shear stresses of woven geotextile-sandy soil interface do not exceed 2kPa, and the differences among the peak shear stresses of nonwoven geotextile-sandy soil interface do not exceed 3kPa, which shows that the sandy soil density has no significant influence on the cyclic shear properties of woven/nonwoven geotextile-sandy soil interfaces.More>

2016, 16(6): 12-20.

FullText HTML (134) PDF (1264KB) (2750)

Water migration rule of loess subgrade with sand interlayers

YAN Zhang-gen, ZOU Qun, XU Yu, WAN Qi, SHI Yu-ling, MA Gang-feng

Abstract: Based on the loess subgrade of Shitian Expressway and the principle of soil water potential, laying sand interlayer in the loess subgrade was proposed to reduce the development of collapsibility water immersion.In the indoor model test, the sand interlayers were lain in the middle and bottom of loess subgrade, the seepage of top water, the infiltration of slope water and the capillarity of underground water for loess subgrade were simulated, the variation laws of water content in loess subgrade, their influence on the strength and stability of loess subgrade were analyzed, and their effect of water resistance and leakage resistance for sand interlayers was proved.Test result shows that the migration of water is rapider in initial period and slower in the middle and later periods; under the effect of ground water, the water contents of subgrade with a sand interlayer at the bottom of subgrade are 24%-27% after 32 days, but the water contentsreach to 60%in middle and lower parts(0.5mbelow the top surface)of pure loess subgrade, and the final influence depth of water reaches 1.2m;under the seepage of top water, the water contents of all depths of pure loess subgrade are more than 60% after 12 days, while the water contents of loess subgrade with a sand interlayer in the middle are less than 40% at 15 cm under the interlayer(0.8mbelow the surface of subgrade), and the moisture content is less than 30%at 25 cm under the interlayer.Obviously, the sand interlayers lay in middle and bottom of compacted loess subgrade can cut off the capillary water and reduce water infiltration, which can decrease the inside water content and increase the whole stability and strength for loess subgrade.More>

2016, 16(6): 21-29.

FullText HTML (132) PDF (1051KB) (2492)

Risk probability assessment of seismic damage for embankment based on Monte Carlo method

YIN Chao, WANG Xiao-yuan, LIU Fei-fei, TIAN Wei, A MELANI

Abstract: The embankment of Lianyungang-Horgos Expressway at K1125+470 was taken as research object, the seismic damage, embankment structure form and ground motion input were determined, the seismic vulnerability curves were drawn by combining incremental dynamic analysis and probabilistic seismic demand analysis, and the seismic vulnerability of embankment was evaluated.Embankment's seismic risk probability was defined as the convolution of seismic hazard and embankment seismic vulnerability, and its calculation method was put forward.The probability distribution model of seismic intensity and the relationship between seismic intensity and PGA were studied, and the probability distribution function of PGA was propssed in the next 50 years.The risk probabilities of embankment seismic were evaluated by using the Monte Carlo method for embankments with and without retaining wall, and the positive effect of retaining wall on improving embankment's seismic performance was verified.Analysis result shows that when PGA is 0.6g, the probability exceeding severe damage of embankment without retaining wall is 65.910%;When PGA is 0.8g, the probability is 99.995%, so the seismic vulnerability of embankment without retaining wall is high.The risk probability of exceeding severe damage for embankment without retaining wall is 29.07% and the risk probability of main integrity and minor damage is 31.97% in the next 50 years.In the next 50 years, the risk probability exceeding severe damage for embankment with retaining wall is 7.9% lower than the value without retaining wall, and the risk probability of main integrity and minor damage is 12.14%higher than the value without retaining wall, which indicates that retaining wall can significantly reduce the seismic damage risk of embankment.The acceptable risk probability exceeding severe damage in the next 50 years is defined as 40%in order for new embankments' aseismic design and existing embankments' aseismic reinforcement.More>

2016, 16(6): 30-38.

FullText HTML (228) PDF (669KB) (2455)

Security calculation method of steel sheet pile cofferdam based on incremental method

DAI Liang-jun, ZHU Da-yong

Abstract: The existing calculation theories of steel sheet pile were analyzed, and the soil pressure hydrostatic pressure, flowing water pressure, wave pressure and supporting pressure acting on steel sheet pile cofferdam were calculated.The construction process of steel sheet pile cofferdam was analyzed, the steel sheet pile was taken as nonlinear elastic foundation beam, the nonlinear differential equations of elastic foundation beam were established based on the incremental method, the horizontal resistance coefficient of foundation soil, the bending deformation coefficient of steel sheet pile and the increments of internal and external loads of steel sheet pile cofferdam were calculated, and the forces and deformations of steel sheet pile under all conditions were studied.Based on the steel sheet pile cofferdam of Yuxi River Bridge, the internal force, deformation, strength, overall stability, anti-overturning safety, anti-uplift safety of steel sheet pile cofferdam were checked.Calculation result shows that the maximum displacement is 38.12 mm and less than the specification value of 80mm; the maximum allowable stress of steel sheet pile is 168.60 MPa and less than the specification value of 218.52MPa; the maximum allowable stress of pipe pile is112.29 MPa and less than the specification value of 174.07 MPa; the overall stability safety factor of steel sheet pile cofferdam is 2.8and greater than the specification value of 1.3;the antioverturning safety factors under 4cases are 3.4, 2.8, 2.9, 1.9respectively and greater than the specification value of 1.2;the anti-uplift safety coefficients based on Prandtl theory and Terzaghi theory are 3.1, 3.5respectively and greater than the specification value of 1.6.Obviously, the safety calculation results of steel sheet pile cofferdam based on the incremental method meet the requirements of the relevant specification, so the calculation method is reasonable and reliable.More>

2016, 16(6): 39-47.

FullText HTML (126) PDF (834KB) (3623)

Coupling stability of couplers between locomotive and vehicle

ZOU Rui-ming, MA Wei-hua, LUO Shi-hui

Abstract: In order to figure out the coupling stability of couplers between locomotive and vehicle, apolygonal contact model was proposed to simulate the contact and friction of coupling surfaces between two couplers, namely, the surface of coupler knuckles was modeled by polygon way, so the contact area was determined and discretized.The contact force was solved by using the elastic foundation model, and the corresponding friction force was determined by the normal contact force and relative tangential velocity.The reliability of the presented modeling method was verified by coupler connection contour for locomotives and rolling stock(TB/T 2950—2006), interference analysis and finite element calculation.A dynamics model of heavy haul train was built to reflect the dynamics performance of middle locomotive and the dynamic behaviors of its coupler and buffer device.The coupling stability of couplers between locomotive and vehicle was analyzed when train runs down a long down-gradient with slope of 1.2% and negotiates a curve with 300 mradius under different conditions.Analysis result indicates that the coupler forces have major effect on the relative motion between the coupling surfaces of couplers, and normal contact force is generated between coupling surfaces of couplers due to stable coupler force.Ifthere exist relative motion or the motion tendency between coupling surfaces of couplers, the relative motion of coupler is prevented by the generated tangential friction force.When coupler force is fluctuant or smaller, couplers are effortless in a state of free clearance, therefore, greater relative motion occurs because generating constraints are not sufficient.The maximum vertical relative displacement of the moment reaches to approximately 149.5 mm.About 4.5° of maximum relative angle is produced in horizontal plane to compensate the deflection behavior of the couplers when negotiates a curve with 300 mradius.More>

2016, 16(6): 48-54.

FullText HTML (109) PDF (1432KB) (3609)

Fault-tolerant compensation control of PMSM applied under open-circuit faulty condition

BAI Hong-fen, ZHU Jing-wei, QIN Jun-feng

Abstract: Based on the fault-tolerant compensation control strategy of the unchangeable principle of magnetomotive force(MMF), the mathematical model of the MMF under healthy condition was studied in the permanent magnet synchronous motor(PMSM)applied in ship electric propulsion system, and the mathematical model of the MMF under one-phase open-circuit faulty condition was established. When one-phase open-circuit fault occurred, the MMF could be remained unchanged by changing the size and angle of currents in two other healthy phases.And the vector control based on space vector pulse width modulation(SVPWM)was still applicable in the motor control.The whole simulation model of ship electrical propulsion system was set upand simulated in MATLAB/Simulink combining the system model of ship electrical propulsion system, and the parameters of torque, speed and current were analyzed.Simulation result shows that the ripple of output torque in the propeller model, and the motor's torque, current and speed ripples are 0.85%, 5.90%, 6.25% and 2.01% respectively, which shows that the motor has good operating performance under healthy condition. Under one-phase open-circuit faulty condition, the failure response of the propeller without fault-tolerant compensation control strategy is intense, the ripple of output torque in the propeller is 22.2%, the output torque, current and speed of the propulsion motor have large fluctuations that are about 82%, 50% and33% respectively, and the flux linkage track motor's stator is unstably elliptical.After using the vector fault-tolerant compensation control strategy based on the unchangeable principle of MMF, the output torque of the propeller will return to the stable value and the ripple is 1.02%.The ripples of output torque, current and speed of the motor are significantly decline that are about6.20%, 6.78% and 2.22%respectively.The ripples limits are close to the values under healthy condition, and the flux linkage track of motor's stator returns to be the same circle under healthy condition.So the propulsion motor can run normally under healthy and open-circuit faulty conditions when the fault-tolerant compensation control strategy is adopted, and the propeller load and the propulsion motor performance can match well all the time.More>

2016, 16(6): 55-62.

FullText HTML (122) PDF (683KB) (3566)

Multi-objective optimization inspection decision-making method based on delay-time model

LU Xiao-hua, ZUO Hong-fu, BAI Fang

Abstract: In accordance with the preventive check plans, the defects detected in the course of inspection, and the repairing and renewing records in a certain cycles for the HPTACC system of a type of aeroengine of an airline freight fleet, the feasibilities for regarding the delayed time for the defects detected and the inspection and repairing costs as safety and economy optimization objectives were analyzed respectively.Under the inspection and repairing strategy of defects detected at the moments of preventive inspection, the probability expressions of expected number of defects and the delayed time for defects detected based on the delay-time model at any inspection momentwere deduced.Under the inspection and repairing strategy of defects degrading into failure and then being found timely and renewed at once, the probability expression of expected number of failure occurring based on delay-time model in every inspection interval was deduced.Based on the probability expressions under 2inspection and repairing strategies, the likelihood function for the system in a given life cycle was build.The double optimization objective functions including inspection and repairing costs and expected delayed time for detected defects were formed.A Pareto optimal solution set of double objective functions were derived by using the improved nondominated sorting genetic algorithm.According to the deciders' objective preference options and the empirical estimates of inspection and repairing costs and the delayed time for detected defects corresponding to their boundary values, the objective preference functions of inspection and repairing costs and the delayed time for detected defects in certain life cycles were determined respectively.The preference interval for every value in the Pareto optimal solution set was determined by using the objective preference function.Based on the collected inspection and repairing data and the proposed methods, an example was analyzed, in which the objective preferences of the delayed time of detected defects and the inspection and repairing costs for deciders were general and good respectively.Analysis result shows that the optimal inspection intervals are 67, 70 or 77landing and take-off cycles, which could provide detailed and more accurate decision-making reference of multi-objective relative optimization for deciders.More>

2016, 16(6): 63-71.

FullText HTML (116) PDF (682KB) (3679)

Optimal design of driving motor structural parameters for electric vehicle

WANG Jun-nian, LIU Jian, CHU Liang, WANG Qing-nian, WU Jian

Abstract: In the process of parameters matching study of driving motor for electric vehicle, an optimal design method for structural parameters of driving motor was proposed.Based on the basic given motor parameters, the influences of axial length, rotor outer diameter, winding turns, wire diameter, pole arc factor, and permanent magnet thickness on motor efficiency were analyzed.The mapping relationships between the main ontology structural parameters and the efficiency characteristic of the motor were established.The procedures of the preliminary design and the optimal design for the motor ontology structural parameters were proposed.Based on the optimized motor efficiency characteristic, the economic performance of whole vehicle was simulated and verified by using the forward-facing simulation vehicle model under 4typical working conditions.Simulation result shows that in the aspect of output characteristic, compared with the initial motor, the torque ripple of optimized motor reduces obviously, the value in constant torque area reduces to 14%, the value in constant power area reduces to no more than40%, and the top efficiency increases to 94%.In the aspect of whole vehicle economy performance, the energy consumptions per kilometer of optimized motor for whole vehicle reduceby about 7.1%, 6.7%, 4.1%, and 2.9% under NEDC、UDDS、JC08、1015working conditions, and the average value is 5.2%.Under the premise of meeting the driving requirements, the operating point distribution in higher efficiency range is improved, and the average efficiency in the area with higher rotational speed and lower torque rises significantly by using the proposed optimal design method.The design method can better improve the driving efficiency of motor, and can supply the theoretical supervision for the optimal design of driving motor ontology structural parameters according to the requirement of vehicle performance.More>

2016, 16(6): 72-81.

FullText HTML (116) PDF (1739KB) (3821)

Intelligent assessment for collaborative simulation training in ship engine room

DUAN Zun-lei, REN Guang, ZHANG Jun-dong, CAO Hui

Abstract: The assessment for simulation training in ship engine room was studied based on the theory of man-machine-environment system engineering.The system model called role-missionresource for collaborative training was constructed, and the intelligent assessment methods based on expert system and machine learning were proposed to improve the three dimensional engine room collaborative training system.The scene knowledge base, mission knowledge base and evaluation index system of simulation training in ship engine room were established.The assessment rules were extracted according to the expert experience and the evaluation criteria.The data were evaluated by using multiple fuzzy comprehensive evaluation and the weights were optimized by genetic algorithm.The structures of machine learning network were constructed appropriately according to the practical assessment problem.The computing steps were simplified by using the advantage of self-learning of BP neural network and deep learning algorithms.The normalized evaluation index was used as the input of evaluation model, and the evaluation resultswere used as the target data.The features of mass sample data were transformed by using sparse autocoder.Sample features were learned deeply and used for classification evaluation, thus the better evaluation model was achieved after repeated training.A comparative analysis of the intelligent assessment methods based on expert system and machine learning was conducted.Analysis result indicates that after the optimization of genetic algorithm, the assessment result error is smaller apparently.The average absolute error is 0.761 point, the average relative error is 0.983%, the mean square error is 0.938 point, the maximum error is 2.263 point, and the minimum error is 0.248 point.For the evaluation of simpler missions, the maximum absolute error of assessment result based on machine learning is 3.521 point, and the minimum absolute error is 0.304 point.The assessment errors of all indexes in the better assessment network based on deep learning are less than 1point.More>

2016, 16(6): 82-90.

FullText HTML (139) PDF (566KB) (3560)

Extraction method of main routes in South China Sea based on spatial cluster analysis

WANG Jia-sheng, LIU Yong-xue, YANG Kun, LI Man-chun, SUN Chao

Abstract: An extraction method of main routes in South China Sea based on spatial cluster analysis was proposed.The VOSClim ship data in 2001-2014 were used to obtain the extraction and classification result of main routes in South China Sea through route generation, route cluster and cluster centerline extraction.Extraction result was compared with the routes of American Central Intelligence Agency.The distribution characteristics of main routes in South China Sea were analyzed.Analysis result shows that 21 main routes in South China Sea are extracted.There are 5, 6and 10 routes for classes one, two and three respectively.Among 21 main routes in South China Sea, up to 14 routes go through the surrounding waters of Nansha Islands.Among 5routes of class one in South China Sea, 1, 2and 2routes are distributed at the northern, western and southern South China Sea respectively.The first five routes sorted by significance index from high to low are Singapore-Hong Kong, Hong Kong-Taiwan Strait, Singapore-Balabac Strait, Singapore-Bashi Strait and Singapore-Kaohsiung routes successively.The direction of main routes in South China Sea is mainly southwest-northeast.Singapore andHong Kong are two most important transportation hubs in South China Sea.More>

2016, 16(6): 91-98.

FullText HTML (113) PDF (2510KB) (2657)

Influence of atmospheric stability on formation of artificial anoxic area over sea surface

JIN Liang-an, LIU Wen-peng, GAO Zhan-sheng, ZHENG Zhi-lin

Abstract: Aiming at the formation of specified anoxic area by gas release in water, MATLAB and Gaussian plume model were used to simulate and analyze the change of gas concentrations at different locations under six atmospheric stability classes(A-F), and the corresponding isoconcentration curve and surface were drawn.The specific influence law of atmospheric stability on anoxic area formation was given. Analysis result shows that the geometric height of continuous point source is 0, the rising height is approximately 0, and the effective source height is approximately 0.The corresponding atmospheric stability classes of gas concentrations from low to high at the same downwind distance and the same sea surface location are A-F successively.The corresponding atmospheric stability classes of areas covered by isoconcentration curves and regions covered by isoconcentration surfaces are A-F successively.With the increase of atmospheric stability, the residence time and concentration of gas increase.So the tendency of diffusion close to sea surface is obvious, which is beneficial to the formation and maintenance of anoxic area with larger effective radius.More>

2016, 16(6): 99-106.

FullText HTML (95) PDF (2124KB) (2284)

Attenuation law of retroreflection coefficient for highway traffic sign retroreflective sheeting

CHEN Yan-yan, JIANG Ming

Abstract: The retroreflection coefficients of 230 traffic sign retroreflective sheetings set in Beijing test site of Ministry of Transport were observed continuously for thirteen years.The prediction models of attenuation values for the retroreflection coefficients of traffic sign retroreflective sheetings with different colors and types were built.The attenuation law of retroreflection coefficient with set time was analyzed.Analysis result shows that the prediction models of attenuation values for the retroreflection coefficient of white typesⅠ-Ⅲretroreflective sheetings are cubic function, logarithmic function and cubic function models respectively, and the determination coefficients are 0.581, 0.732 and 0.559 respectively.The variation amplitudes of attenuation rate for retroreflection coefficient of red types Ⅰ-Ⅲ retroreflective sheetings are 16.0%, 28.0% and 42.0% respectively, and the variation amplitudes of attenuation rates for retroreflection coefficients of retroreflective sheetings with other colors are less than 21%.So, the quadratic function and cubic function models of attenuation values for retroreflection coefficients have better fitting property, there are significant differences among attenuation laws of retroreflection coefficients for retroreflective sheetings with different colors and types, and theattenuation slows down with increase of set time.More>

2016, 16(6): 107-113.

FullText HTML (147) PDF (1086KB) (2513)

Microscopic simulation parameter calibration of CMEM model for temporary maintenance zone

GAO Tian-zhi, CHEN Kuan-min

Abstract: In order to ensure the accuracy of automobile exhaust emission calculation, the parameters of microscopic simulation model for temporary maintenance zone were calibrated.The temporary maintenance zone located on the Xuwei Highway in Henan Province was taken as example, the traffic data were measured through field investigation.The VISSIM traffic simulation model was built.According to the measured data, the macroscopic parameters of traffic volume and traffic composition were calibrated, the microscopic parameters such as desired speed, desired acceleration were calibrated by using the values of feature points.Four parameters of car-following model including headway time, following variation, threshold for entering following status and vibration acceleration were calibrated by using orthogonal test.According to the valid simulation results, the numerical relationship between the desired speed and the running speed was determined.The emission of automobile exhaust was calculated by using the CMEM model with valid simulated data, and the calculation formulae of emissions were put forward based on average speeds.Analysis result shows the simulated speeds only calibrated by using macroscopic parameters are apparently different from the measured speeds, and the averagerelative errors of speeds for passenger car and truck are 11.36% and 35.12%.Combined with microscopic parameters calibration, the average relative error between measured and simulated speeds is lower than 3%.The desired speeds are 1.270 and 1.165 times of running speeds for passenger car and truck.After calibration, the relative error of simulated and measured emissions is less than 7%, and the calibration effect of model is remarkable.More>

2016, 16(6): 114-124.

FullText HTML (100) PDF (1310KB) (2362)

Joint decision making of pricing and seat inventory control based on discrete time

GAO Jin-min, LE Mei-long, QU Lin-chi

Abstract: In order to maximize the revenue of airline company, passenger booking and refunding tickets were regarded as two independent random processes.Poisson distribution was used to simulate the process of passenger booking tickets in the whole pre-sale period. Negative exponential distribution was used to simulate the process of passenger refunding tickets.The joint decision making model of air ticket pricing and seat inventory control based on discrete time was built from the perspective of using price to control demand.The solving idea and method of dynamic programming were used to solve the model.The air ticket being sold with which price in appropriate time period was determined.The number limit of ticket sales in each time period was set.The model was verified by an example.Analysis result shows that 80% discount, 70% discount, 60% discount, 50% discount, 40% discount, 30% discount, 20% discount, 10% discount and full price air tickets should be sold 60.00, 56.58, 52.87, 48.83, 44.38, 39.44, 33.88, 27.51 and 20.09 din advance respectively.Booking limits should be 3, 43, 79, 111, 141, 171, 199, 227 and 290 respectively. The joint decision making model has validity and practicability.It can be used as the reference for the airline company to set price and control seat inventory reasonably.More>

2016, 16(6): 125-131.

FullText HTML (69) PDF (419KB) (2432)

Model of real-time pedestrian detection under vehicle environment based on CS-SD

GUO Ai-ying, XU Mei-hua, RAN Feng, WANG Qi

Abstract: In order to solve the real-time problem in the advanced driver assistant system, a model of pedestrian detection based on the calibration of side-of-pavement line and saliency texture detection (CS-SD) and the location histogram of oriented gradient (L-HOG) was proposed.The CS-SD algorithm was used instead of exhaustive search to quickly mark pedestrian area in the image.The L-HOG was used to quickly extract pedestrian feature, and additive kernel support vector machine (AK-SVM) was used to efficiently classify detected objects.Analysis result shows that when 500 images including 832 pedestrians on personal computer are detected, the model detects 720 pedestrians correctly, the detection rate is 86.5%, the error rate is 4.1%, and the detection time is 39 ms.When 48 400 images including 988 pedestrians on vehicle pedestrian detection system based on BF609 are detected, the model detects 861 pedestrians correctly, misses 127 pedestrians and detects 13 pedestrians in error.The detection speed is 20 fps.Underthe premise of not reducing the detection rate, the proposed pedestrian detection model can reach satisfying detection speed and can be used in vehicle equipment of real-time pedestrian detection.More>

2016, 16(6): 132-139.

FullText HTML (103) PDF (3421KB) (2750)

Classification method of running environment features for unmanned vehicle

KANG Jun-min, ZHAO Xiang-mo, XU Zhi-gang

Abstract: In order to improve the barrier classification ability of mobile 2D LiDAR in urban environment, the creating accuracy of environmental map, and the safety and accuracy of autonomic behavior decision-making for unmanned vehicle, a classification method of environmental features based on machine learning was proposed.The data from 2D LiDAR were divided into independent data segments, and each data segment contains one environmental barrier.In 2D Gaussian probability density space of data segments, the elliptical axial lengths of contour lines, the log likelihood values and the maximum density were taken as the elements of sample data of artificial neural network, and the data segments were classified by the artificial neural network.The classification validity was estimated according to the weights of artificial neural network's output data to retain the effective environmental features, and the features were extracted from the classified data.Computational result shows that in the same test scenario, when the judging condition of classification validity is relaxed, under which the classification stability interval is [0.55, 1], the classification transition interval is [0.45, 0.55), and the classification invalid interval is [0, 0.45), 98 environmental features are extracted, the maximum standard deviation of classified extraction results for the multiple observation data of one environmental feature is 30.7 mm, and the average standard deviation for all features is 5.1 mm; when the judging condition of classification validity is strict, under which the classification stability interval is [0.65, 1], the classification transition interval is [0.35, 0.65), and the classification invalid interval is [0, 0.35), 93 environmental features are extracted, the maximum standard deviation of classified extraction results for multiple observation data of one environmental feature is 22.0 mm, and the average standard deviation for all features is 4.2 mm.Therefore, the proposed classification method has higher noise tolerance ability and classification accuracy.More>

2016, 16(6): 140-148.

FullText HTML (106) PDF (972KB) (2473)

Infrared traffic image's enhancement algorithm combining dark channel prior and Gamma correction

GU Ming, ZHENG Lin-tao, LIU Zhong-hua

Abstract: In order to enhance the visual quality of infrared traffic image collected by the intelligent traffic monitoring equipment effectively, the image defogging method of visible light was introduced into traffic infrared image enhancement processing, and a new infrared traffic image's enhancement algorithm combining dark channel prior and Gamma correction was proposed.First, the original degraded infrared traffic image was processed by dark channel prior algorithm to obtain initially enhanced image.Then, the brightness of initially enhanced image was adjusted by Gamma correction algorithm.The image enhancement effects of the new algorithm and other common infrared image enhancement algorithms were compared.Test result shows that the information entropies of two original infrared traffic images are respectively 4.71 and 5.07 and respectively increase to 6.45 and 5.92 after being processed by the new algorithm.The standard deviations of gray scale for two original infrared traffic images are respectively 6.90 and 19.14 and respectively increase to 31.17 and 32.35 after being processed by the new algorithm.The information entropy computational value of new algorithm is more than the values of other algorithms.So the enhancement effect of the proposed algorithm is better than the enhancement effects of other common infrared image enhancement algorithms, and it can significantly improve the visual effect of infrared traffic image and lay good foundation for following processing and analysis of image.More>

2016, 16(6): 149-158.

FullText HTML (123) PDF (3683KB) (2398)

User Center

Journal Info

Collection of Source JournalsMore>

2016 Vol. 16, No. 6

Search Type

Search Journal

LinksMore>