ASD

ASD PRODUCT 

ASD - a sophisticated image processing system that tracks cars’ plates, reads and identifies their numbers. It is intended for high speed lanes, up to 200 KM/H.  The Violation Enforcement System (VES) Solution will be provided to accommodate the need for consistent, real–time License Plate Recognition (LPR) for vehicles passing through the enforcement lanes, including:

o BUS / HOV lane Enforcement

o Average Speed - calculate the average speed of the vehicles, based on the time of journey.

o Flow estimation – the average speed and time of journey can be used for road conjunction analysis that can assist traffic planning.

o On-line traffic report – the roadside information can be reported on web sites in order to supply live reports from the freeway.

o Monitoring – the recognition information may be used for various security applications.

SeeWay is a state-of-the-art vision based recognition system for high speed Roadside installations. The system can integrate multiple lanes and multiple cameras per lane into a sophisticated vision-based License Plate Recognition (LPR) system that identifies and tracks number plates on vehicles travelling at high speeds. The system is used for various applications, including Toll Roads, monitoring, average speed and car flow. The application is supported by a full set of optical and hardware subsystems as well as software applications and utilities.

System description:

The proposed system consists of 2 enforcement lanes, with a single camera monitoring the entrance and exit of the each lane. All vehicles passing the camera will be recorded in terms of the time, lane, direction, license plate (if present), if allowed to use the lane (white list) and an alarm if the vehicle is wanted (black list).

System operation:

The system will capture all vehicles entering and exiting the enforcement lanes, storing the vehicle image, license plate if present, date, time, lane and image. The software will allow vehicles to be enrolled into a white list, linking this information to the license plate. If a vehicle is detected which is not allowed to use the lane, this will be recorded.
If the vehicle is in the black list, an alarm will be generated.

The operation of the system is as follows:   When the vehicle detector signals the start / end of vehicle (high / low going signal), it will trigger the application (via a IO card) telling it to take a series of pictures. The application could capture the images from two or more cameras that are installed on that lane (for example front & rear).

The application also controls the IR illumination level via the IO card – switching from off (the normal idle state) through low, medium and high illumination states. This set is predefined in the application’s parameter settings, and could be defined for different time of day (such as night and day capture sets).

Capture an additional image using an additional colour camera that will be positioned to cover the entire width of the lane. This camera will supply a view of the lane which can be used for a manual review – it can be used to verify the presence of a plate and the colour of the vehicle’s body.

The application analyses each of the captured images and extracts the plate string from each image. After collecting and comparing the results from all captures, the application determines the final result and selects the best image. It then outputs its 2 constituents: the JPEG image and OCR Data.

Video flow: After the cameras have acquired the images data, they will be transferred to a corresponding iPort video to IP unit. Each iPort unit corresponds one-to-one with each camera. The iPort units convert the camera-link interface to data that will be transmitted on the VES Gigabit, a closed Gigabit LAN whose sole function is to transfer image data from the cameras to the VES servers, and to send and receive commands to/from the lane controller. All cameras are connected via a Gigabit switch (8 ports), so the VES server can actually interface with each of the iPort units to all the lane cameras. This is in case one of the VES servers is down, and the other VES servers can be commanded to handle the other cameras as well.  When the VES Server receives the image data, it will convert the data to arrays of images. It then proceeds to perform image enhancement and recognition. It will select the best result out of the captured images.

Specification of the camera: Digital 2/3" monochrome progressive scan CCD camera 1392 (h) x 1040 (v) 6.45 μm square pixels

Extended IR sensitivity

10 bit video output as Camera Link

Full 1380 (h) x 1030 (v) frame readout in 1/24 second

Partial scan to 1/8 and vertical binning

Edge pre-select (EPS) and pulse width control (PWC) trigger modes

Analog video output for controlling auto – iris lenses

Shutter speeds from 1/24 to 1/10,000 second in 10 steps

Trigger and timing signals as LVDS or via Camera Link

Setup by switches or serial control (short ASCII commands)

Windows 98/NT/2000/XP setup software

The camera interface box inside the enclosure converts the digital image to network packets, and connects on a Gigabit network via a communication cable. It pumps the uncompressed video images at real time stream. This connection is also used to control the camera settings.

The IP engine interfaces the cameras and transfers the images and control over the Gigabit link. Its main features:

High-performance, ultra-efficient connectivity between camera link cameras and Gigabit Ethernet Links or LAN acquires images from a wide range of cameras delivers long-distance bi-directional links over inexpensive CAT-5 copper LAN cable serial RS232 control over camera

Inputs : 2 TTL inputs, 1 LVDS, 1 optically isolated input

Outputs : 2 TTL outputs, 1 optically isolated output

Triggers:  The sensors that trigger the system can be either a fast traffic loop detector or a laser detector. The triggers are dry-contact type, and each lane camera requires one trigger.

Illumination units:  IR solid state strobe units connected to the TBL and  get the sync strobe from the camera. The strobes have 4 levels of illumination (off, low, medium and high) that are controlled by a pair of control units from the TBL.

One of the main purposes and advantages of image based recognition systems is the creation of the image files.  Due to the high resolution, the images to be kept in a jpg format, with a quality level that is defined in the settings. The user may select one of the 5 levels (lowest quality to highest quality).  The other parameters include the local images directories path, and the option to save the images in a daily sub-folder, weekly or single directory.  The naming of the images files is composed of an optional camera prefix and a random number that is assigned to the image. Thus, all images are identified in a unique identification for each lane. This unique number may be used for creating a transaction number.  Solution designed to share the vehicle identifications with other processes. This can be done by application-to application messages.

Message structure:  Each message contains the results of the recognition process, and one message is sent after each trigger.  The fields include:

• Data and time

• Lane number

• Plate string

• Image file path

• Confidence

• Match result & name of driver (in case of a database)

• Other fields

Remote sites: The standard Gigabit network installation on basis of a CAT6 cable assumes that the camera output is limited to 100M.  However, for longer distances a fibre connection may be used, with a fibre converter for each camera. Alternatively, the switch can be connected near the camera and connected with CAT6 to each camera, while the output of the switch will be converted to fibre, and reconverted from fibre to a GB switch near the computer.

Redundancy: There is a need to guarantee an absolute up time, i.e  the systems always work, even in case of malfunction or required service.  The system is designed to work in an automatic redundancy mode, where the 2nd server automatically takes over the functions of the other down server.  The dual servers can be set to monitor each other through network messaging and revert to degraded mode if there is a fault in one of the servers. To support this mode, both servers should be connected to the same cameras.

Utilities provided should include:

• a watchdog utility that periodically checks the aliveness of the application. In case the application does not respond, the application will attempt to rerun the application. If this fails, the utility resets the PC. In any such case the event is written to the Windows event log.

• Monitor tool resides on a central server, and monitors the state of each system – by checking the Windows event log. It will alert external systems in case of a fatal error. It can also show soft errors

(warnings) status, and display a set of graphs of past recognition results, which is a very important diagnostic tool.

• Cleaner – This utility cleans the old images directories after a specified time has elapsed, and also cleans local diagnostics files. Thus, the system will not grow endlessly in size, a common source of problem in other Windows based systems.

APPENDIX: ADDITIONAL ASD INFO:

APPENDIX 1: “ASD (PDF)” – A PowerPoint introducing license plate recognition, ASD system design, showing the proposed ASD site layout, how the ASD system works and some example images of the ASD results from the N3 demo.

APPENDIX 2: “See Way Manual (PDF)” – This document provides a technical overview on SeeWay, a state-of-the-art vision based recognition system for high speed Roadside installations. The application is supported by a full set of optical and hardware subsystems as well as software applications and utilities.

APPENDIX 3: “Overview (PDF)” – Contains a brochure of the LPR solutions available from ASD.

APPENDIX 4: “SeeLane (PDF)” – Contains a brochure of the See Lane software on which See Way is based.

APPENDIX 5: “See Car DLL (PDF)” – Contains a brochure of the See Car DLL software on which See Way is based.

APPENDIX 6: “See Data (PDF)” – Contains technical information on the See Data a software service application that connects a cluster of recognition systems (such as See Lane or See Way) together by a network.

APPENDIX 7: “VES (PDF)” – technical information on the See Way install, operation and design.

APPENDIX 8: “See Road Technical (PDF)” – Contains technical details of the See Road software on which See Way is based

APPENDIX 9: “See Lane Manual (PDF)” – Contains technical details of the See Lane software on which See Way is based

APPENDIX 10: “ASD N3 RESULTS Sat Jan 06th (PDF)” – Contains the typical daily results from the N3 for a single day

APPENDIX 11: “ASD N3 RESULTS SUN JAN 7th (PDF)” – Contains the typical daily results from the N3 for a single day

APPENDIX 12: “ASD - LPR Solution for 1 site 2 lanes (XLSX)” – Full equipment and software list, capital, rental or cost per transaction amounts

APPENDIX 13: “pd130dsw.(PDF)” – Contains details of the PD130 Single Channel vehicle detector

AVAILABLE FROM THE I-CUBE WEB SITE (www.I-Cube.co.za) ARE:

“SEE WAY DEMO” – Use the see way demo to see how the LPR software works over multiple lanes and watch as alarms are generated when the vehicle exceeds the set average speed above which alarms will be generated.

“PLAYER” – Use the PLAYER.EXE to watch the cars from the N3 & M4 being recognised using the LPR DLL

“SEE LANE” – Use the SEE LANE DEMO to see how the software works, enrol allowed cars, alarm on WANTED cars, see the log of all the vehicles and see a history of the vehicles.

If you would like to know more about this service please SMS 082 562 8225 / or E-Mail ASD at I-Cube dot co dot za 

If you would like to SUBSCRIBE to this service please call 031 764 3077 or Fax  0866539659 E-Mail ASD at I-Cube dot co dot za 

I-Cube.   All rights reserved.  Revised: January 13, 2008 .