一种基于计算机视觉的车辆速度检测方法_英文_魏武
Journal of Southeast Universi ty(English Edition) Dec. 2000 Vol.16 No.2 ISSN1003 7985 A M e tho d of V eh ic le S p eed D e tection Ba sed on C om pu te r V is ion
Wei Wu1 Huang Xinhan1 Wang Min1 Zhang Qisen2
(1Department of Control Science and Engineering,Huazhong University of Science and Technology,Wuhan430074) (2Department of High way and Bridge,Changsha Communications University,Changsha410076)
Abstract: This paper presents a method of vehicle speed detection based on computer vision.It can effective-
ly detect vehicle speed by analyzing vehicle and scene informati on extracted from i mage.It merely needs two
CCD TV cameras to complete road-traffic parameter detection in b-i directional and four-lane high way or urban
road.The experi mental results show that the correct rate of detection is close to92%,and rea-l time perfor-
mance i s satisfactory.
Key words: vehicle speed detection,computer vision,frames differencing,i mage processin g,symbol ana-
lyzing
Road-traffic monitoring using optical sensor involves the collection of data describing the characteris-tics of vehicle and road-traffic para meters.Road-traffic parameters,such as vehicle count,vehicle dens-i ty,vehicle speed,queue situation and queue length are valuable to traffic monitoring and management in-cluding congestion analyzing,incident detec ting,road capacity increasing,traffic data statistic processing and so on.
Currently,detection of road-traffic parameters mainly relies on detectors such as ultrasonic,in-frared,radar,microwave and induc tive-loop detectors.For ultrasonic and microwave detectors,in most case,it is not easy to satisfy precision requirement of parameter detection.In addition,the detection dis-tance is not far(<12m)enough.For infrared detector,because of disturbance of vehicle heat source and environmental noises,the quality of detection is not satisfactorily high.Radar can accura tely measure vehicle speed,but the cost is too high.Inductive-loop detector can effectively detect road-traffic parame-ters,but its installation is more disruptive to road surface than other detectors.During recent decades, researchers pay more attention to the method using CCD TV ca meras[1,2].It is potentially powerful to de-tect road-traffic parameters due to the development of computer vision,image processing,artificial intell-i gence and intelligent transportation systems.
This paper presents a method of vehicle speed detection using computer vision.This method makes full use of vehicle and scene information extracted from image to detect vehicle speed.I t merely needs two CCD TV ca meras to complete vehicle speed detection in b-i directional and four-lane highway or urban https://www.wendangku.net/doc/2b14546125.html,ing this method,the distance which can be detected is more than50m along lane direction and the installation is cheaper than other detec tors.The experimental results show that the correct rate of de-tection is close to92%,and rea-l time performance is impressive.
1 S ystem Overview
I mage processing is an important technique in the proposed method.First,the difference is calculat- Received2000-02-26.
The project supported by Key Science and Technology Developing Plan Project of Hubei Provi nce(No.991P0111).
Born i n1970,male,graduate.
ed between two frames from image -grabber and CCD TV ca meras in different time.Frame differenc -ing [3-5]can detect appearance and move ment of vehicles.Road -traffic parameters can then be deter mined by analyzing the appearance and movement of vehicle.Procedure flo w is shown in Fig. 1.
Fig.1 Procedure flow Prototype frames are 512 256grey -level
images.Image grabber needs to capture 50frames
in one second.Frame differencing is the key pro -
cess during vehicle speed detecting.
Frame differencing can reveal regions of mo -
tion due to the fact that difference e xists between
two frames captured at different times.It can be
described as Eq.(1).A difference image d (i ,
j )is generated by calculating the absolute differ -ence be tween two frames f 1,f 2(captured by CCD
TV camera in different time),and then thresholding the result with T 0.
d(i,j )=0,if |f 1(i,j )-f 2(i,j )| T 01,otherwise
(1)where T 0is a suitable threshold.
In the case of road -traffic monitoring it is usual for f 1to be the incoming frame and f 2to be a reference of background frame.The reference frame is merely an image of scene with no vehicles inside.If the incoming frame contains no vehicles then it will be identical to the reference frame and the difference frame will be zero.However,if the incoming frame does contain vehicles,then difference frame will show up.The function of the threshold T 0is to reduce the effects of noise and changes in scene illumination.A simple threshold is not usually sufficient to overcome the effects in varying environment.It is needed to employ a method of dynamically updating the reference frame so that it adapts to changes in scene illumination.
It is difficult to operate the calculation of frame differencing of 50frames with size 512 256within one second on a PC computer.In order to reduce the cost of computation,two points are very important: Reduce processing data obtained from the optical sensors; Adopt simple algorithms.Frame differencing is concentrated processing on only key regions of the image,relying heavily on particular characteristics in ten lines (each CCD TV camera processes three lines)which are perpendicular to the lanes.The distance between lines is determined according to the lines layout in the image.The distances among lines in roadway are approximately equal.The distances among lines in the image are not equal,but the location of each line in the image can be calculated by geometrical relations and the order of lines in image.The line width in image is about 3pixels.Because difference between images is calculated only in line area,the cost of calculation is relatively low.The distribution of lines in the roadway and corresponding image is shown as Fig.2.Each CCD TV camera is mounted facing vertically down on a roadway,g iving a field of view of about 20m (two cameras about 50m).Four lanes are coded to A ,B,C and D .
In most case,frame differencing can successfully detect appearance and movement of vehicles.Some time,the detection of vehicle appearance and movement is error due to unfit threshold and effects of shadows and wet road.
We set virtual trip -wire near the line i (for example,if line i is located at row k ,k 1,then the virtual trip -w ire locates row k +4in the image).This is useful to reduce the error detection.The state of each trip -w ire is determined by looking at the cross -section of the image under the trip -w ire.The cross -section is located at the v irtual trip -wire in which vehicle may be appeared.The cross -section is first low -pass filtered to reduce the effects of noise and remove high frequency texture
118Wei Wu,Huang Xinhan,Wang Min,and Zhang Qisen
Fig.2 Distr ibution of lines in ro adway and in image
information which is often associated with the road surface.The median deviation of the cross -section is then calculated as Eq.(2)
md (j )=1N N i=1|x (i,j )-x (j )|, x (j )=1N N
i=1
x (i,j )(2)where x (i,j ),i =1,2, ,N are the filtered values of N pixels under the cross -section;N is the width (represented with pixels)under the cross -section located in trip -wire.Fig.3shows the cross -section of the image,the cross -section after low -pass filtering and median deviation for both a broken and a complete trip -wire.In figure,horizontal axis represents the pixel poison under the cross -section,and vertical axis represents corresponding the value of the median deviation.It can be seen that when a vehicle is crossing the trip -wire the median deviation is considerably higher due to the greater amount of contrast associated with a vehicle compared to that of the road.The median deviation is therefore a suitable cue for deciding the appearance and movement of vehicle together with frame differencing.
The vehicle symbol which describes the vehicle state (appearance or no appearance)can be noted according to the situation of the frame differencing on the lines in different times and also the trip -wire.For example,we consider the range [a,b]of line 1in a different image (|b -a | 30with respect to width of vehicle is less than 3m),if the values d (i,j )of all pixels are 1,while the situation of trip -wire is broken and line 1 lane A ,then vehicle appears in the line 1of lane A ,and also we mark a symbol + in the corresponding place,otherwise symbol * .Another symbol is time symbol,which is marked in the prototype image and corresponding difference image according to capturing time.For ex ample,15:52:10is time symbol,where,15is minute,52is second,and 10is the tenth 20ms.2 Vehicle Speed Detection
2 1 Definition
In order to describe clearly,we give the following definition.
ST ={T m :T s :T ms }:a set of time symbol,T m ,T s {1,2,3, ,59,60},T ms {1,2,3, ,49,50};
Lan ={A,B ,C,D }:a set of lanes;
lin ={1,2,3,4,5,6,7,8,9,10}:a set of lines;
M (p ,i,t),p Lan ,i lin ,t ST:the symbol is + in line i of lane p at time t ;M *(p ,i,t),p Lan ,i lin ,t ST:the symbol is * in line i of lane p at time t ;
119
A Method of Vehicle Speed Detection Based on Computer Vision
(a)M edian deviation =55(vehicle on trip -w ire) (b)M edian deviati on =4(no vehicle on trip wire)Fig.3 Determining state of trip -wir e from median deviation of low -pass filtered cross -scetion of image
N (p ,i,t t +T),p Lan ,i lin :the number of vehicle passed line i and lane p from time t to t +T ;
SP (p ,i j ,t t +T ),p lan ,i lin:the average speed of vehicle passing lane p between line i and line j from time t to t +T .
Fig.4 T he first case for calculating vehicle speed 2.2 Vehicle speed detection
The average speed of vehicle,SP (p ,i
i+1,t t+T)in the area between line
i and line i+1in lane p from time t to t+
T ,can be calculated by the marked symbols
described in the last section.For example,
for lane C ,the average speed of vehicle can
be calculated using three methods with
respect to three cases.The first case is
shown in Fig.4.Seven stages represent
different areas in the same lane,in which vehicle appear.
The first case is that there is only one vehicle within the area between line i and line i+1,and the length of vehicle is less than 10m.In fact,in most case,it is impossible that more than two vehicles are located in the same lane in area between line i and line i +1,because the distance between line i and line i +1is 10m.However,the length of two vehicles add the headway in most case is more than 10m.During the period of a vehicle X passing the area between line i and line i +1(including line i and line i +1),the symbols in line i and line i +1satisfy Eq.(3)and (4). M *(C,i,t) M (C ,i,t +k ) M *(C,i,t +q )=TURE (3)
120Wei Wu,Huang Xinhan,Wang Min,and Zhang Qisen
M *(C,i +1,t +q +v ) M (C,i +1,t +q +w +j )
M *(C,i +1,t +q +w +l)=TRUE (4)where k =1,2, ,q -1;v =1,2, ,w ;j =1,2, ,l -1.The average speed of vehicle X in area between line i and line i +1can be calculated as Eq.(5).
SP (C ,i i +1,t +1 t +q +w +1)=10/(q +w )(5)The second case is that there is only one vehicle within the area between line i and line i +1,and the length of vehicle is more than 10m.In this case,symbols in i and line i +1satisfy Eqs.(6)and (7).
M *(C,i ,t ) M (C,i,t +k) M *(C,i ,t +q) M *(C,i +1,t +u)=TR UE (6) M (C,i +1,t +u +1) M (C,i +1,t +u +j )
M *(C,i +1,t +u +l)=TRUE (7)where k =1,2, ,q -1;j =1,2, ,l -1;k
SP (C ,i i +1,t +1 t +q +u +1)=10/(q +u)(8)In the third case,there are more than one vehicle within area between line i and line i +1,and the total length of two vehicles and headway is less than 10m.In this case,symbols in line i and line i +1satisfy Eqs.(9),(10),(11)and (12).
M *(C,i,t) M (C ,i,t +k ) M *(C,i,t +q )=TURE (9) M *(C,i +1,t +q +v ) M (C,i +1,t +q +w +j )
M *(C,i +1,t +q +w +l)=TRUE (10) M *(C,i +1,t +q +w +l +1) M *(C,i +1,t +q +w +l +g )
M (C,i +1,t +q +w +l +f )=TRUE (11) M (C,i +1,t +q +w +l +f +d)
M *(C,i +1,t +q +w +l +f +n)=TRUE (12)where j =1,2, ,l-1;g =2,3, ,f -1;d =1,2, ,n -1.The average speed of vehicle X in area between line i and line i +1can be calculated as Eq.(13).
SP (C ,i i +1,t +1 t +q +w +l +f )=10 50/(q +w +l +f -1)(13)We choose one of Eqs.(5),(8)and (13)to calculate the average speed of vehicle according to vehicle length L .Because we do not know L in prior,we cannot directly decide by L .We present a method which can effectively decide correct expression without requiring L to be accurately known in prior by the following steps.
Step 1 If the time of vehicle passing line i is more than the time of vehicle passing the area between the line i and line i +1(not including line i and line i +1),then L is longer than 10m,otherwise,L is shorter than 10m.The second case can be determined by expression(14).
Step 2 First,analyze Eqs.(9)-(12).If two vehicles have passed line i+1,and within the same time no vehicle has passed line i ,that is to say that if Eqs.(15)and (16)are satisfied,then it belongs to the third case,other w ise it belongs to the first case.
u q ,h +u >q (14) N (C,i,t +q +w t +q +w +l +f +n)=1(15) N (C,i +1,t +q +w t +q +w +l +f +n)=2(16)In the same way,by analyzing symbols,we can obtain vehicle speed on the other lane.Also,other road -traffic parameters (traffic flow,vehicle density,queue situation and queue length,etc.)can be obtained by analyzing symbols.Details will be presented in other papers.
3 Experiment and Conclusion This method is used in the I MSUR (intelligent monitoring system in urban road)system in 121
A Method of Vehicle Speed Detection Based on Computer Vision
122Wei Wu,Huang Xinhan,Wang Min,and Zhang Qisen
Wuhan City of Hubei province,P.R.China.The experimental results show that the correct rate of vehicle speed is close to92%.Using this method to detect vehicle speed,rea-l time performance is satisfactory.Errors of detection in most case come from error difference image generated by varying illumination condition.
References
1 P.G.Michalopoulos,Vehicle detection through image processing,the autoscope sys tem,IEEE Tr ans.on Vehicu lar
Technology,vol.40,no.1,pp.21-29,1991
2 B.Carlson,Vision makes traffic control intelligent,Advanced Imaging,vol.12,no.2,pp.54-56,1997
3 W.Wei,X.H.Huang,M.Wang,and W.Li,Method of road-traffic parameter detection using optical sensor,In:
SPIE:Proceeding of I n ternational Con f.on Sensors and Control Techniques,Wuhan,China,pp.469-472,Jun.2000 4 A.T.Ali,J.Bulas-Cruz,and E.L.Dageless,Vision based road traffic data collection,In:Proc.ISATA26-th Int.
Con f.,pp.609-616,Sept.1993
5 P.Briquet,Video processing applied to road and urban traffic moni torin g,In:IEE.6-th Int.Con f.,Road Tra ffic
Monitoring,London,pp.153-157,April1992
一种基于计算机视觉的车辆速度检测方法
魏 武1 黄心汉1 王 敏1 张起森2
(1华中理工大学控制科学与工程系,武汉430074)
(2长沙交通学院路桥工程系,长沙410076)
摘 要 提出了一种基于计算机视觉的车辆速度检测方法,通过对图像中提取的车辆和场景信息
分析来有效地检测车辆速度,只需要2台CCD TV摄像机可完成双向四车道高速公路或城市道路
上车辆的速度检测.实验结果表明:车辆速度检测的正确率达92%,并具有满意的实时特性.
关键词 车辆速度检测,计算机视觉,图像帧差,图像处理,标号分析
中图法分类号 U491 14
中英文文献翻译-电动助力转向系统
附录A 外文文献 Electric Power Steering system 1.History In automobile development course, Steering system experienced four stages of development: from the initial mechanical Steering system (for your DNS setting Steering, abbreviation ) development for Hydraulic Steering system (Hydraulic Power Steering, abbreviation HPS), then again appeared electronically controlled Hydraulic Steering system (Electro Hydraulic Power Steering, abbreviation EHPS) and Electric Power Steering system (Steering, room Power as EPS). Assemble mechanical steering system of car parking and low-speed driving, when the driver's steering control burden too heavy, in order to solve this problem, the American GM in the 1950s took the lead in the car hydraulic steering system. But, hydraulic steering system can't juggle vehicles to speed portability and high speed, so the steering stability Koyo in Japan in 1983, with the company introduced the application of speed sensing function of hydraulic steering system. This new type of steering system can provide speed increased with the decreasing steering, but complicated structure, cost is higher, and cannot overcome hydraulic system itself has many shortcomings, is a cross between a hydraulic steering and electric power steering the transition between the products. In 1988, Japan Suzuki company first in small cars equipped with Cervo Koyo company development on the steering column, power type electric power steering system; In 1990, Japan Honda NSX in sports car company adopted self-developed rack power type electric power steering system, henceforth unveils the electric power steering in cars applications history 2.Working principle Electric power steering system are as follows: first, the working principle, torque sensor measured on steering wheel drivers on the manipulation of the moment, the wheel speed sensors detect the vehicle driving speed, then present the two signals to ECU; According to the built-in control strategy: ECU, calculates the ideal target booster torque, into current instructions to motor; Then, the power generated by the torque motor slowdown institutions amplification on steering system in mechanical manipulation of the moment, and the driver together to overcome resistance torque, realize to the vehicle steering. 3. Working process
人工智能与计算机视觉
过去几年，全球的互联网公司包括谷歌、微软、Facebook以及中国的百度、阿里巴巴都在加强人工智能领域的投资，设立自己的人工智能研究院。vivo是第一家设立专攻人工智能方向研究院的中国手机公司。此举是vivo内部已经确立的一份3-5年的中长期发展的战略规划，未来对人工智能的发展研究是必然趋势，vivo公司创始人兼CEO沈炜曾表示“人工智能和5G的结合将会是5G时代手机发展的趋势”。 今年我们看到vivo在产品上不少创新，比如AI拍照、商用屏下指纹技术等等，这些都是基于生物特征(biometrics)的鉴别技术，除此之外还有对人脸、虹膜、指纹、声音等特征上的识别，这些大多涉及到视觉信息，正是体现了计算机视觉的应用性，那什么是计算机视觉呢? 计算机视觉技术的概念 正像其它学科一样，一个大量人员研究了多年的学科，却很难给出一个严格的定义，模式识别如此，目前火热的人工智能如此，计算机视觉亦如此。与计算机视觉密切相关的概念有视觉感知(visual perception),视觉认知(visual cognition),图像和视频理解( image and video understanding)。这些概念有一些共性之处，也有本质不同。 从广义上说，计算机视觉就是“赋予机器自然视觉能力”的学科。自然视觉能力，就是指生物视觉系统体现的视觉能力。一则生物自然视觉无法严格定义，在加上这种广义视觉定义又“包罗万象”，同时也不太符合40多年来计算机视觉的研究状况，所以这种“广义计算机视觉定义”，虽无可挑剔，但也缺乏实质性内容，不过是一种“循环式游戏定义”而已。 实际上，计算机视觉本质上就是研究视觉感知问题。视觉感知，根据维科百基(Wikipedia)的定义, 是指对“环境表达和理解中，对视觉信息的组织、识别和解释的过程”。根据这种定
机械类汽车转向系统外文文献及翻译
1 Introduction The key task for the automobile industry and its suppliers in future lies in speedily developing and implementing ecologically sound and economically justifiable mobility systems. Light metals such as aluminum and magnesium along with glass and carbon fiber reinforced materials, ceramics and composites have opened up the potential for considerable weight reduction and for "green" vehicle concepts which can be realized economically. Aluminum in particular can provide the impetus for new designs for the next millennium. Decades ago, the use of aluminum in auto construction was seen as an "experiment"; Today it is a vital factor in reducing weight and thus lowering fuel consumption. The average passenger car today contains 60 to 70 kg of aluminum, and current developments point to a doubling of this amount in the next few years. Motor vehicles both now and in future must meet requirements for: greater performance, greater safety, comfort, low pollution. Lightweight construction is not just about reducing weight; it is a question of -striking the right balance between reduced weight and structural efficiency. In vehicle construction this normally means making the best use of the generally very tight space available for individual components so as to allow weight to be minimized while still meeting all stiffness, strength, natural frequency or acoustical requirements. To achieve this, stresses must be distributed throughout the structure as evenly as possible. Modern numerical analysis methods such as FEA allow a very detailed analysis of system behavior, provide cost-efficient support for the complex process of optimization and thus make a huge contribution to advances in lightweight construction. Packaging, safety considerations, reproducibility and price place restrictions on the degree of weight reduction achievable. The broad range of expertise available to Krupp Presta AG allows the company to analyze customer specifications for steering systems and provide appropriate solutions. 2 Requirements to be met by steering systems The steering is an important part of the feel of a car. The steering system should make driving an enjoyable experience with no unpleasant vibration from the road surface while guaranteeing the required hand- sing. It is also important that high safety requirements be met, both under normal conditions and in crash situations. The key criteria for the steering system are thus as follows:rolling friction, torsional stiffness /strength, Damping, temperature, corrosion, durability / fatigue, weight. Crash kinematics and energy absorption steering column requirements:natural frequency / stiffness, mass, damping, space, strength (crash, misuse), ergonomics, handling, acoustics, crash kinematics and energy absorption. Other basic conditions:interfaces with adjacent components, installation, joining techniques, price. 3 Materials material light weighting can be achieved by using either stronger or lighter material. When stiffness or natural frequency are Important sizing criteria, low density materials with a high modulus of elasticity by quired. Non-exotic materials must be selected which are readily recyclable, low in price and display good durability.Further requirements are set by the manufacturing and joining processes. Steel, aluminum, magnesium and a variety of plastics are the materials of choice for steering systems.
计算机视觉技术
目录 1立体视觉 (1) 1.1计算机视觉技术 (1) 2立体视觉技术 (3) 2.1双目立体视觉技术 (3) 致谢 (8) 附录： (9)
立体视觉 我的毕业论文排版样文 1立体视觉 1.1计算机视觉技术 计算机视觉既是工程领域也是科学领域中的一个富有挑战性的重要研究领域。计算机视觉是一门综合性的学科，它已经吸引了来自各个学科的研究者参加到对它的研究之中，其中包括计算机科学和工程、信号处理、物理学、应用数学和统计学、神经生理学和认知科学等[18]。 视觉是各个应用领域，如制造业、检验、文档分析、医疗诊断和军事等领域中各种智能自主系统中不可分割的一部分。由于它的重要性，一些先进国家，例如美国把对计算机视觉的研究列为对经济和科学有广泛影响的科学和工程中的重大基本问题，即所谓的重大挑战。“计算机视觉的挑战是要为计算机和机器人开发具有与人类水平相当的视觉能力。机器视觉需要图像信号，纹理和颜色建模，几何处理和推理，以及物体建模。一个有能力的视觉系统应该把所有这些处理都紧密地集成在一起[19]。”作为一门学科，计算机视觉开始于60 年代初，但在计算机视觉的基本研究中的许多重要进展是在80 年代取得的。现在计算机视觉已成为一门不同于人工智能、图象处理、模式识别等相关领域的成熟学科[20]。 不少学科的研究目标与计算机视觉相近。这些学科包括图像处理、图像识别、景物分析、图像理解等。由于历史发展或领域本身的特点这些学科互有差别，但又有某种程度的相互重叠。为了清晰起见，把这些与计算机视觉有关的学科从研究目标和方法角度加以归纳[21]。 （1）图像处理 图像处理技术把输入图像转换成具有所希望特性的另一幅图像。例如，可通过处理使输出图像有较高的信噪比，或通过增强处理突出图像的细节，以便于操作员的检验。在计算机视觉研究中经常利用图像处理技术进行预处理和特征抽取。 （2）图像识别 图像识别技术根据从图像抽取的统计特性或结构信息，把图像分成预定的类别。在计算机视觉中图像识别技术经常用于对图像中的某些部分(例如分割区域)的识别和分类。 第 1 页（共9页）
智能汽车中英文对照外文翻译文献
智能汽车中英文对照外文翻译文献 (文档含英文原文和中文翻译) 翻译： 基于智能汽车的智能控制研究 摘要：本文使用一个叫做“智能汽车”的平台进行智能控制研究，该小车采用飞思卡尔半导体公司制造的MC9S12DG128芯片作为主要的控制单元，同时介绍了最小的智能控制系统的设计和实现智能车的自我追踪驾驶使用路径识别算法。智能控制智能车的研究包括:提取路径信息,自我跟踪算法实现和方向和速度控制。下文介绍了系统中不同模块的各自实现功能，最重要部分是智能车的过程智能控制：开环控制和闭环控制的应用程序包括增量式PID控制算法和鲁棒控制算法。最后一步是
基于智能控制系统的智能测试。 关键词：MC9S12DG128；智能控制;开环控制；PID；鲁棒； 1.背景介绍 随着控制理论的提高以及信息技术的快速发展,智能控制在我们的社会中发挥着越来越重要的作用。由于嵌入式设备有小尺寸、低功耗、功能强大等优点，相信在这个领域将会有一个相对广泛的应用,如汽车电子、航空航天、智能家居。如果这些技术一起工作，它将会蔓延到其他领域。为了研究嵌入式智能控制技术，“智能汽车”被选为研究平台,并把MC9S12DG128芯片作为主控单元。通过智能控制,智能汽车可以自主移动,同时跟踪的路径。 首先,本文给读者一个总体介绍智能车辆系统的[2、3]。然后,根据智能车辆的智能控制:提取路径信息,自我跟踪算法实现中,舵机的方向和速度的控制。它提供包括了上述四个方面的细节的智能车系统信息。此外,本文强调了智能车的控制过程应用程序包括开环控制、闭环增量PID算法和鲁棒算法。 2.智能车系统的总体设计 该系统采用MC9S12DG128[4]作为主芯片，以及一个CCD传感器作为交通信息收集的传感器。速度传感器是基于无线电型光电管的原理开发。路径可以CCD传感器后绘制收集的数据，并且系统计算出相应的处理。在同时，用由电动马达速度测试模块测量的智能汽车的当前速度进行响应的系统。最后，路径识别系统利用所述路径信息和当前的速度，以使智能汽车在不同的道路条件的最高速度运行。图1示出了智能车辆系统的框图。
人机交互中的计算机视觉技术.
人机交互中的计算机视觉技术 基于视觉的接口概念 计算机视觉是一门试图通过图像处理或视频处理而使计算机具备“ 看” 的能力的计算学科。通过理解图像形成的几何和辐射线测定, 接受器(相机的属性和物理世界的属性, 就有可能 (至少在某些情况下从图像中推断出关于事物的有用信息, 例如一块织物的颜色、一圈染了色的痕迹的宽度、火星上一个移动机器人面前的障碍物的大小、监防系统中一张人脸的身份、海底植物的类型或者是 MRI 扫描图中的肿瘤位置。计算机视觉研究的就是如何能健壮、有效地完成这类的任务。最初计算机视觉被看作是人工智能的一个子方向, 现在已成为一个活跃的研究领域并长达 40年了。 基于视觉的接口任务 至今,计算机视觉技术应用到人机交互中已取得了显著的成功,并在其它领域中也显示其前景。人脸检测和人脸识别获得了最多的关注, 也取得了最多的进展。第一批用于人脸识别的计算机程序出现在 60年代末和 70年代初,但直到 90年代初,计算机运算才足够快,以支持这些实时任务。人脸识别的问题产生了许多基于特征位置、人脸形状、人脸纹理以及它们间组合的计算模型, 包括主成分分析、线性判别式分析、 Gabor 小波网络和 .Active Appearance Model(AAM . 许多公司,例如Identix,Viisage Technology和 Cognitec System,正在为出入、安全和监防等应用开发和出售人脸识别技术。这些系统已经被部署到公共场所, 例如机场、城市广场以及私人的出入受限的环境。要想对人脸识别研究有一个全面的认识,见。 基于视觉的接口技术进展 尽管在一些个别应用中取得了成功,但纵使在几十年的研究之后,计算机视觉还没有在商业上被广泛使用。几种趋势似乎表明了这种情形即将会发生改变。硬件界的摩尔定律的发展, 相机技术的进步, 数码视频安装的快速增长以及软件工具的可获取性(例如 intel 的 OpenCV libraray使视觉系统能够变得小巧、灵
转向器英文文献翻译
Electronic power steering system What it is： Electrically powered steering uses an electric motor to drive either the power steering hydraulic pump or the steering linkage directly. The power steering function is therefore independent of engine speed, resulting in significant energy savings. How it works : Conventional power steering systems use an engine accessory belt to drive the pump, providing pressurized fluid that operates a piston in the power steering gear or actuator to assist the driver. In electro-hydraulic steering, one electrically powered steering concept uses a high efficiency pump driven by an electric motor. Pump speed is regulated by an electric controller to vary pump pressure and flow, providing steering efforts tailored for different driving situations. The pump can be run at low speed or shut off to provide
汽车悬架系统中英文对照外文翻译文献
汽车悬架系统中英文对照外文翻译文献(文档含英文原文和中文翻译) 汽车悬架 现代汽车中的悬架系统有两种，一种是从动悬架，另一种是主动悬架。 从动悬架即传统式的悬架，是由弹簧、减振器（减振筒）、导向机构等组成，它的功能是减弱路面传给车身的冲击力，衰减由冲击力而引起的承载系统的振动。其中弹簧主要起减缓冲击力的作用，减振器的主要作用是衰减振动。由于这种悬架是由外力驱动而起作用的，所以称为从动悬架。 而主动悬架的控制环节中安装了能够产生抽动的装置，采用一种以力抑力的方式来抑制路面对 车身的冲击力及车身的倾斜力。由于这种悬架能够自行产生作用力，因此称为主动悬架。 主动悬架是近十几年发展起来的，由电脑控制的一种新型悬架，具备三个条件： （1）具有能够产生作用力的动力源；
（2）执行元件能够传递这种作用力并能连续工作； （3）具有多种传感器并将有关数据集中到微电脑进行运算并决定控制方式。因此，主动悬架汇集了力学和电子学的技术知识，是一种比较复杂的高技术装置。 例如装置了主动悬架的法国雪铁龙桑蒂雅，该车悬架系统的中枢是一个微电脑，悬架上有5 种 传感器，分别向微电脑传送车速、前轮制动压力、踏动油门踏板的速度、车身垂直方向的振幅及频率、转向盘角度及转向速度等数据。电脑不断接收这些数据并与预先设定的临界值进行比较，选择相应的悬架状态。同时，微电脑独立控制每一只车轮上的执行元件，通过控制减振器内油压的变化产生抽动，从而能在任何时候、任何车轮上产生符合要求的悬架运动。因此，桑蒂雅桥车备有多种驾驶模式选择，驾车者只要扳动位于副仪表板上的“正常”或“运动”按钮，轿车就会自动设置在最佳的悬架状态，以求最好的舒适性能。 另外，主动悬架具有控制车身运动的功能。当汽车制动或拐弯时的惯性引起弹簧变形时，主动悬架会产生一个与惯力相对抗的力，减少车身位置的变化。例如德国奔驰2000 款CL 型跑车，当车辆拐弯时悬架传感器会立即检测出车身的倾斜和横向加速度，电脑根据传感器的信息，与预先设定的临界值进行比较计算，立即确定在什么位置上将多大的负载加到悬架上，使车身的倾斜减到最小。 汽车主动悬架—液压和空气式 从控制力的角度划分，悬架可分为被动悬架，半主动悬架和主动悬架。目前，大多数汽车的悬架系统装有弹簧和减振器，悬架系统内无能源供给装置，其弹性和阻尼不能随外部工况变化，因此称这种悬架是被动悬架。 主动悬架有作为直接力发生器的动作器，可以根据输入与输出进行最优的反馈控制，使悬架有最好的减震特性，以提高汽车的平顺性和操纵稳定性。它由弹性元件C和一个力发生器Fe组成。 半主动悬架可看作由可变特性的弹簧和减振器组成的悬架系统，虽然它不能随外界的输入进行最优的控制和调节，但它可按存储在计算机的各种条件下最优弹簧和减振器的优化参数指令来调节弹簧的刚度和减振器的阻尼状态。它由弹性元件 C 和一个一个阻尼系数能在较大范围内调节的阻尼器组成。
汽车电动助力转向系统的研究大学毕业论文外文文献翻译及原文
毕业设计（论文）外文文献翻译 文献、资料中文题目：汽车电动助力转向系统的研究 文献、资料英文题目：The auto electric power steering system research 文献、资料来源： 文献、资料发表（出版）日期： 院（部）： 专业： 班级： 姓名： 学号： 指导教师： 翻译日期：2017.02.14
英文原文 The auto electric power steering system research Along with automobile electronic technology swift and violent development, the people also day by day enhance to the motor turning handling quality request. The motor turning system hanged, the hydraulic pressure boost from the traditional machinery changes (Hydraulic Power Steering, is called HPS), the electrically controlled hydraulic pressure boost changes (Electronic Hydraulic Power Steering, is called EHPS), develops the electrically operated boost steering system (Electronic Power Steering, is called EPS), finally also will transit to the line controls the steering system (Steer By Wire, will be called SBW). The machinery steering system is refers by pilot's physical strength achievement changes the energy, in which all power transmission all is mechanical, the automobile changes the movement is operates the steering wheel by the pilot, transmits through the diverter and a series of members changes the wheel to realize. The mechanical steering system by changes the control mechanism, the diverter and major part changes the gearing 3 to be composed. Usually may divide into according to the mechanical diverter form: The gear rack type, follows round the world -like, the worm bearing adjuster hoop type, the worm bearing adjuster refers sells the type. Is the gear rack type and follows using the broadest two kinds round the world -like (uses in needing time big steering force).In follows round the world -like in the diverter, the input changes the circle and the output steering arm pivot angle is proportional; In the gear rack type diverter, the input changes the turn and the output rack displacement is proportional. Follows round the world -like the diverter because is the rolling friction form, thus the transmission efficiency is very high, the ease of operation also the service life are long, moreover bearing capacity, therefore widely applies on the truck. The gear rack type diverter with follows round the world -like compares, the most major characteristic is the rigidity is big, the structure compact weight is light, also the cost is low. Because this way passes on easily by the wheel the reacting force to the steering wheel, therefore has to the pavement behavior response keen merit, but simultaneously also easy to have phenomena and so on goon and oscillation, also its load bearing efficiency relative weak, therefore mainly applies on the compact car and the pickup truck, at present the majority of low end passenger vehicle uses is the gear rack type machinery steering system. Along with the vehicles carrying capacity increase as well as the people to the vehicles handling quality request enhancement, the simple mechanical type steering system were already unable to meet the needs, the power steering system arise at the historic moment, it could rotate the steering wheel while the pilot to provide the boost, the power steering system divides into the
中英文文献翻译-汽车转向系统
Auto steering system Abstract: With the rapid development of electronic technology, people on the vehicle steering control performance of more and more high. Wire control steering system will be the trend of the development of automobile steering system. Normally according to mechanical steering gear form can be divided into: super-modulus gear type, circulation ball type, worm wheel type, worm gear and worm refers to pin type. The increase of vehicle weight and people to the improvement of vehicle handling performance requirements, simple mechanical steering system can not meet needs, power steering system came into being, it can in the pilot turned the steering wheel and at the same time provide power, power steering system is divided into hydraulic steering system and electric steering system 2 kinds. Hydraulic steering system which is the most widely used steering system. Hydraulic steering system is the basis of the mechanical system increases the system, including hydraulic pump, v-shaped belt wheel, oil injection equipment, power devices and control valves. With the help of the automobile engine driving hydraulic pump, air compressor and generator etc, in order to hydraulic, pneumatic or electric power increases the power of the front wheel steering the pilot control to make drivers can light the flexibility to manipulate car turned, reduce labor intensity, improve the driving safety. Electric power steering system is now the development direction of car steering system, its working principle is: EPS system to the ECU from steering wheel torque sensor and speed sensor signal analysis, control motor to produce the proper power torque, assist drivers finish to the operation. Key words: Wire control steering system Hydraulic steering system Electric power steering system sensor Along with automobile electronic technology swift and violent development, the people also day by day enhance to the motor turning handling quality request. The motor turning system changed, the hydraulic pressure boost from the traditional machinery changes (Hydraulic Power Steering, is called HPS), the electrically controlled hydraulic
交通运输-外文翻译-外文文献-英文文献-北京运输系统
附件1：外文资料翻译译文 北京运输系统 摘要：运输系统提供了与城市郊区显着不同的服务，这通常会导致有关用户选择交通方式的不同假设。本文所提到的仿真模型，提出了评估影响运输服务的政策。运输方式被认为是公共交通工具，包括轻型轨道交通（轻轨）和公共汽车，加上私人轿车。在三步旅行者行为仿真模型中，使用了广义运输成本的概念。它提出各类交通以及对居民的出行选择和郊区住宅社区形态量化的建议，并使用了源自于中国北京一个典型走廊的数据。 仿真结果表明：票价降低，增加公共交通的综合能力，以及对私家车的惩罚是很有必要的，以此来提高系统效率和郊区的吸引力，特别是对那些低收入的人；没有公路定价，将鼓励中等收入居民转向私人轿车，同时,高收入者可能因为道路拥挤离开郊区；然而公共交通的改善可以吸引更多短距离和中距离的旅客，但轿车用户对于旅行距离并不敏感。 关键词：广义成本，公共交通，拥挤定价，运输服务，北京 从上世纪90年代，中国大中城市已经历郊区化，而且这些城市的空间结构逐步形成，在很大程度上依赖于交通运输的进步。由于人口爆炸和改造中心城市，许多郊区城镇发展为住宅区，这些新发展城镇的大部分居民受雇于中心城市或附近的工业区。例如，80%多回龙关郊区社区的居民在北京或发达地区中心城市工作，几乎有一半的北京社区居民受雇于朝阳区中央商务区。这些领域在交通系统规划与提供经营效率方面对交通政策制定者和城市规划设计师提出了新的挑战。 在许多情况下，边沿城镇通过高速公路和城市轨道连接中心城市或工业园区。相比传统的城市，郊区展览更稳定的运输使用模式，更加依赖公共交通，私人汽车，较少使用机动模式（自行车，步行）。工作距离比任何单一因素对人们的交通方式
汽车转向系统车辆外文文献翻译、中英文翻译、外文翻译
汽车转向系统 随着汽车电子技术的迅猛发展，人们对汽车转向操纵性能的要求也日益提高。汽车转向系统已从传统机械转向、液压助力转向(Hydraulic Power Steering ,简称HPS) 、电控液压助力转向( Electric Hydraulic PowerSteering , 简称EHPS) ，发展到电动助力转向系统(Electric Power Steering ,简称EPS) ，最终还将过渡到线控转向系统(Steer By Wire ，简称SBW)。 机械转向系统是指以驾驶员的体力作为转向能源，其中所有传力件都是机械的，汽车的转向运动是由驾驶员操纵方向盘，通过转向器和一系列的杆件传递到转向车轮而实现的。机械转向系由转向操纵机构、转向器和转向传动机械3大部分组成。 通常根据机械式转向器形式可以分为:齿轮齿条式、循环球式、蜗杆滚轮式、蜗杆指销式。应用最广的两种是齿轮齿条式和循环球式(用于需要较大的转向力时) 。在循环球式转向器中，输入转向圈与输出的转向摇臂摆角是成正比的;在齿轮齿条式转向器中，输入转向圈数与输出的齿条位移是成正比的。循环球式转向器由于是滚动摩擦形式，因而正传动效率很高，操作方便且使用寿命长，而且承载能力强，故广泛应用于载货汽车上。齿轮齿条式转向器与循环球式相比，最大特点是刚性大，结构紧凑重量轻，且成本低。由于这种方式容易由车轮将反作用力传至转向盘，所以具有对路面状态反应灵敏的优点，但同时也容易产生打手和摆振等现象，且其承载效率相对较弱，故主要应用于小汽车及轻型货车上，目前大部分低端轿车采用的就是齿轮齿条式机械转向系统。 随着车辆载重的增加以及人们对车辆操纵性能要求的提高，简单的机械式转向系统已经无法满足需要，动力转向系统应运而生，它能在驾驶员转动方向盘的同时提供助力，动力转向系统分为液压转向系统和电动转向系统2 种。其中液压转向系统是目前使用最为广泛的转向系统。 液压转向系统在机械系统的基础上增加了液压系统，包括液压泵、V 形带轮、油管、供油装置、助力装置和控制阀。它借助于汽车发动机的动力驱动液压泵、空气压缩机和发电机等，以液力、气力或电力增大驾驶员操纵前轮转向的力量，使驾驶员可以轻便灵活地操纵汽车转向，减轻了劳动强度，提高了行驶安全性。 液压助力转向系统从发明到现在已经有了大约半个世纪的历史，可以说是一种较为完善的系统，由于其工作可靠、技术成熟至今仍被广泛应用。它由液压泵作为动力源，经油管道控制阀向动力液压缸供油，通过活塞杆带动转向机构动作，可通过改变缸径及油压的大小来改变助力的大小，由此达到转向助力的作用。传统液压式动力转向系统一般按液流的形式可以分为:常流式和常压式2 种类型，也可根据控制阀形式分为转阀式和滑阀式。 随着液压动力转向系统在汽车上的日益普及，人们对操作时的轻便性和路感的要求也日益提高，然而液压动力转向系统却存在许多的缺点：①由于其本身的结构决定了其