Vision-Based Vehicle Speed Measurement System

Embed Size (px)

Citation preview

  • 7/31/2019 Vision-Based Vehicle Speed Measurement System

    1/4

    Journal of Computer Applications (JCA)

    ISSN: 0974-1925, Volume V, Issue 3, 2012

    82

    Abstract - Vision-based vehicle speed measurement

    (VSM) is one of the most convenient methods available in

    intelligent transportation systems. A new algorithm for

    estimating individual vehicle speed based on two

    consecutive images captured from a traffic safety camera

    system. Its principles are first, both images are

    transformed from the image plane to the 3D world

    coordinates based on the calibrated camera parameters.

    Second, the difference of the two transformed images is

    calculated, resulting in the background being eliminated

    and vehicles in the two images are mapped onto one

    image. Finally, a block feature of the vehicle closest to the

    ground is matched to estimate vehicle travel distance and

    speed.

    Index Terms images, vehicle, vehicle speed measurement.

    I. INTRODUCTIONIntelligent transportation systems are becoming more

    important due to their advantages of saving lives, money, and

    time. Acquiring traffic information, such as lane width traffic

    volume (the number of travelling vehicles per time period

    through a position in a lane), traffic density (the number of

    total vehicles in a given area at a given time) and vehicle

    speed, these are the key part of intelligent transportation

    systems, and such information is used to manage and control

    traffic. It focuses on vehicle speed since reducing speed can

    help to reduce accidents.

    Intelligent transportation systems is a new approach to

    estimate vehicle speed using video sequence data recorded

    form roadway surveillance cameras under occlusion. Rather

    than detecting or tracking individual vehicles in each image

    frame as is conventionally done, It transform the sequence of

    frames into a graphical representation of vehicle trajectories

    using time stacks. Where a time stack is constructed from a

    video sequence out of temporally successive slices of the

    same 1D region in the frame. Most vision based trafficmonitoring systems (VBTMS) attempt to detect and track

    discrete vehicles. But such segmentation and tracking is

    hampered by occlusion, changes in lightning conditions and

    shadows. Many Solutions have been developed, Segmenting

    discrete vehicles for detection and tracking in video is

    difficult, measuring the speed of vehicle features is much

    simpler and so this work seeks to bypass the need for

    segmenting each vehicle when measuring traffic speed over

    Manuscript received 8/Sept/2012.

    Manuscript selected 11/Sept/2012.

    Prakash.V,Sri Ramakrishna Institute of Technology , Coimbatore, India

    E-mail: [email protected]

    Devi.P, Assistant Professor, Dept of ECE, Sri Ramakrishna Institute of

    Technology , Coimbatore, India, E-mail: [email protected]

    time and space. Proposed system consists of 5 steps to

    estimate mean speeds:

    (1) Camera calibration from image to world co ordinatetransformation to generate velocity fields,

    (2) Video stabilization to decrease unwanted cameramotion in the video sequence,

    (3) Pre-processing using edge removal apply imageenhancement for background estimation and shadow

    removal,

    (4) Background estimation and shadow removal,(5) Cubic spline interpolation for trajectory

    approximation.

    The overall intention is to implement a vision-based speedsensor considering occlusion caused by overlapping vehicles.

    It is handling the occluders rather than the occluded vehicles.

    For almost two decades these CCTV systems were strictly

    used to extend the view available to human traffic managers.

    The objective behind these video-based traffic surveillance

    systems is to extract important traffic parameters for traffic

    analysis and management such as vehicle speed, vehicle path,

    flow rate and density. Vision-based vehicle detection is

    intended to be a lower cost alternative to many other detector

    technologies, e.g., loop detectors. Loop detectors are

    essentially metal detectors embedded in the pavement, and

    have been the most common detector to record trafficparameters. Depending on deployment these detectors can

    collect information such as vehicle length, speed and flow

    rate. However, loop detectors only detect vehicles directly

    overhead. In contrast, vision-based systems collect

    information over an extended space compared to the

    conventional vehicle detection systems.

    II. BRIEF REVIEW OF VISIONBASED VEHICLE SPEEDMEASUREMENT SYSTEMS

    The block diagram of the proposed algorithm is depicted in

    Fig. 1. To begin with, the two consecutive images taken

    through a perspective view are transformed from their 2Dimage plane to 3D world coordinates using camera

    parameters calculated from lane-marking calibration. As there

    is no height information immediately available from the two

    images, only the X-Y plane in 3D is considered for all

    subsequent estimation. From the two reconstructed images, an

    image differencing is applied so that the two vehicles are

    overlaid on one resultant image. Although this can effectively

    remove the background features as the time difference

    between the two images is small, it also resulted in a large

    amount of noise and holes along the boundaries of the

    vehicles. These are removed by a carefully selected threshold

    and morphological operations to obtain a combined vehicle

    mask. After that, a block matching algorithm is applied, which

    includes specifying the block and matching it within a search

    window. As the 3DX-Yplane is given in physical dimension

    (meters in this case), he distance between the two matched

    Vision-Based Vehicle Speed

    Measurement SystemPrakash.V a,*, Devi.P b,1

  • 7/31/2019 Vision-Based Vehicle Speed Measurement System

    2/4

  • 7/31/2019 Vision-Based Vehicle Speed Measurement System

    3/4

    Journal of Computer Applications (JCA)

    ISSN: 0974-1925, Volume V, Issue 3, 2012

    84

    noted that this is only true for points on the road plane. Other

    points that are not on the road plane do not represent the same

    as height information is not available. Therefore, the problem

    in matching is not finding the most correlated blocks between

    the two vehicle images as, but to make sure that the feature

    points in the block are on the road surface as well. The most

    reliable points in these images for the purpose are the points

    where the wheels touch the ground. Unfortunately, these

    points are extremely difficult to be determined. For the reasonthe subsequent estimation is based on the shadow of the back

    bumper that casts on the ground. Although this feature is

    relatively easy to find, this is also represents a source of error

    due to road colour change, illumination change and shadow.

    Background Componsation and Vehicle Detection

    Separating the vehicle from the road (background) is a major

    issue in detecting a vehicle from a captured image. The

    background image is an image that contains no moving

    objects. As the roads scene changes every second, it is

    impossible to get a consistent background image from a single

    image and create the empty background image with a series of

    images containing the moving objects to create an empty

    background image from a few captured images.

    This algorithm uses the Gaussian mean and the standard

    deviation between pixels from each image to ignore the

    absurd value pixels that represent the moving object.

    Background subtraction computes the inter frame difference

    (IFD) between the background image and the incoming image

    from the camera. The result of the IFD process, i.e., the

    foreground image, involves a black background with floating

    vehicles. Image difference is a widely used method for the

    detection of motion changes. The interframe difference (IFD)

    is calculated by performing a pixel-by-pixel subtraction

    between two images.. In particular, after creating thebackground, displacements can be seen between the incoming

    frames and the generated background.

    Moving Region Extraction

    If a background reference image is available, it can easily

    extract the vehicle masks of each image and estimate the

    vehicle position directly.

    On the premise that no background reference image is

    available, it regard one of the two images as a background

    reference, then perform conventional vehicle mask extraction

    on the other image and finally get an overlaid vehicle mask.

    Morphological ProcessA simple vehicle mask extraction method was adopted, which

    includes three steps: temporal differencing, thresholding and

    morphological operations. First, It overlay one image onto the

    other by taking an absolute difference between the two

    images. This results in the image as depicted in Fig.5(a). For

    colour images and obtain the absolute difference in each

    colour channel first and then average them to a gray level

    image. The image depicted in Fig. 5(a) is then threshold to

    produce a binary image as depicted in Fig.5(b). The threshold,

    T, is a predefined value to accommodate small pixel intensity

    variation. After that, morphological opening is used to clean

    up all the isolated points and morphological closing is used tojoin the disconnected vehicle components together, resulting

    in the extracted vehicle mask as shown in Figures.

    (a) (b)Figure 5. (a) Binary image after thresholding.

    (b) Vehicle mask after morphological operations.

    In general, the case as depicted in Fig. represents a typical

    scenario. By computing the connected components in the

    binary image, it has two blobs which represent the masks of

    two individual vehicles. From the distance of the bottom edge

    of the two blobs, displacement of the vehicle in two

    consecutive frames can be roughly estimated, which could be

    refined by block matching. However, for vehicles that are

    travelling at low speed, or have long length or the two

    consecutive images are taken at a shorter time.

    Block Matching

    In this Section, the block matching algorithm (BMA)

    estimates motion between two images or two video frames

    using "blocks" of pixels. First, a source block is defined in

    terms of a horizontal base line, a fixed height and a variable

    width according to the size of the mask. The blob shown in

    Fig.6 is part of the lower blob from the masking image, in

    which a horizontal base line is drawn across the lowest part of

    the blob. From the base line, an upper boundary of k pixelabove the line and a lower boundary also of k pixels below the

    line are defined. As the upper boundary line intersects the

    vehicle mask, the left and right boundaries are defined by the

    width of the mask intersected by the upper boundary line. This

    defines a source block of 2kw in size.

    Figure 6. Definition of Source Block.

    With these boundaries, it has a block defined in the first

    reconstructed image as the source block. The search is

    performed in the second reconstructed image. The mean

    absolute difference (MAD) measurement between the source

    block and candidate blocks is employed as a criterion in

    proposed matching algorithm and roughly estimated

    displacement of the vehicle. The underlying supposition

    behind motion estimation is that the patterns corresponding to

    objects and background in a frame of video sequence movewithin the frame to form corresponding objects on the

    subsequent frame. The idea behind block matching is to

    divide the current frame into a matrix of macro blocks that

  • 7/31/2019 Vision-Based Vehicle Speed Measurement System

    4/4

    Vision-Based Vehicle Speed Measurement System

    85

    are then compared with corresponding block and its adjacent

    neighbours in the previous frame to create a vector that

    stipulates the movement of a macro block from one location

    to another in the previous frame. This movement calculated

    for all the macro blocks comprising a frame, constitutes the

    motion estimated in the current frame. The search area for a

    good macro block match is constrained up to p pixels on all

    fours sides of the corresponding macro block in previous

    frame.

    Figure 7. Travel Distance Estimation

    Velocity Measurement

    Once a block is matched, travelled distance in pixels, PD, of

    the vehicle between the two images is represented by the

    distance between the source block and the matched block.

    Vehicle speed is the distance travelled normalized to physical

    distance in m, and divided by the time taken to traverse that

    distance, which is given by,

    Where, PD - Travelled distance in pixels,R - Resolution of reconstructed images,

    t - Time interval between the two images,

    v - Estimated speed

    III. EXPERIMENTAL RESULTS

    Figure 8. Speed of the vehicles

    In this experiment, we evaluate the accuracy of the proposed

    method in estimating individual vehicle speed from real

    traffic images at day-time as well as night-time, by comparing

    the estimated value with the speed measured by a Doppler

    speed radar system. All the test images are taken by a

    surveillance camera at 0.5 seconds apart, with the size of

    12801024 and 24 bit color depth mode. We first calculate

    the reconstructed images as described in Section 2.2 with

    resolution of 10 pixels/meter, and then threshold the 256 gray

    level frame difference using a threshold of 30 to generate the

    vehicle mask. k, sxand syin block matching period were set to

    10 pixels, 20 pixels and 10 pixels respectively. For the 31 sets

    of day-time images, the results estimated by the proposed

    method and reference speed.

    IV. CONCLUSION AND FUTURE DIRECTIONIn conclusion, a novel method for estimating individual

    vehicle speed using two consecutive images from a traffic

    surveillance camera has been proposed. Compared with speed

    from radar, the averaged estimation errors for day-time cases

    is 3.27%, while for night-time cases is 8.51%. Future work

    will be focused on the block matching accuracy and the

    robustness of the algorithm against large changes in ambient

    illumination conditions.

    REFERENCES

    [1] C. Sun and S. G. Ritchie, "Individual vehicle speed estimation usingsingle loop inductive waveforms," Journal ofTransportation

    Engineering-Asce, vol. 125, pp. 531-538, 1999.

    [2] Y. H. Wang and N. L. Nihan, "Freeway traffic speed estimation withsingle-loop outputs," Advanced Traffic ManagementSystems and

    Automated Highway Systems 2000, pp. 120-126, 2000.

    [3] D. L. Woods, B. P. Cronin, and R. A. Hamm, "Speed Measurementwith Inductance Loop Speed Traps.," Texas Transportation Institute.,Research Report FHWA/TX-95/1392-8, 1994.

    [4] K. W. Dickinson and R. C. Waterfall, "Video Image Processing forMonitoring Road Traffic," in Proceeding of IEEInternational

    Conference on Road Traffic Data Collection, pp. 5-7, 1984.

    [5] R. Ashworth, D. G. Darkin, K. W. Dickinson, M. G. Hartley, C. L.Wan, and R. C. Waterfall, "Application of Video Image Processing for

    Traffic Control Systems," in Proceeding ofSecond International

    Conference on Road Traffic Control, vol. 14-18 London, UK, 1985.

    [6] S. Takaba, M. Sakauchi, T. Kaneko, Hwang Byong Won, and T.Sekine, "Measurement of traffic flow using real time processing of

    moving pictures," in Proceeding of 32nd IEEEVehicular Technology

    Conference, vol. 32, pp. 488-494 San Diego, California, USA, 1982.

    [7] Z. Houkes, "Measurement of speed and time headway of motor vehiclewith video camera and computer," in Proceeding of 6 thIFAC/IFIP

    Conference on Digital Computer Applications toProcess Control, pp.

    231-237, 1980.[8] D. J. Dailey and L. Li, "An algorithm to estimate vehicle speed using

    uncalibrated cameras," in Proceeding of IEEE/IEEJ/JSAIInternational

    Conference on Intelligent TransportationSystems, pp. 441-446, 1999.

    [9] F. W. Cathey and D. J. Dailey, "One-parameter camera calibration fortraffic management cameras," in Proceeding of7th International IEEE

    Conference on IntelligentTransportation Systems, pp. 865-869, 2004.

    [10] F. W. Cathey and D. J. Dailey, "A novel technique to dynamicallymeasure vehicle speed using uncalibrated roadway cameras," in

    Proceeding of IEEE Intelligent VehiclesSymposium, pp. 777-782,

    2005.

    BIOGRAPHY

    V. Prakash received the BE degree in Electronics and

    Communication Engineering from Sri RamakrishnaInstitute of Technology, Anna University Coimbatore

    in 2012. He is an active member of IETE. He had

    presented national conferences and international

    conferences in various fields He had done the project

    in the area of Digital Image Processing. His area of

    interest is Vehicular technology, Digital Electronics, Digital Image

    Processing.

    P. Devi received the BE degree in Electronics and

    Communication Engineering from VLB Janakiammal

    College of Engineering and Technology, Anna University

    Chennai in 2009 and the ME degree in Communication

    Systems from Sri Krishna College of Engineering and

    Technology, Anna University Coimbatore in 2011.

    Currently working in Sri Ramakrishna Institute of

    Technology as an Assistant Professor in ECE Department.She has one year teaching experience. She has presented the papers in the

    national conference. Under her guidance, the final year students are doing

    their project in various fields. Her field of interest is Digital Image

    Processing.