See/Container is a compact stand-alone system that is used to automatically track and read a shipping container identification number together with the license plate of the carrying Truck. This sophisticated image processing system can be placed at various port locations: at the gates, at the docks, or in any other location that requires fully automatic identification.

The identified number string is displayed on the system’s main display, logged in its disk, and transferred for data collection to an external computer (with RS232 line), or can be used in conjunction with a pre-stored list of authorized or non-authorized container numbers in order to open a gate or activate an alarm (with I/O lines).

Each unit can control 3 or 4 video cameras (there are two basic configurations), and handle them simultaneously. The system uses these multiple cameras in order to look at various sides of the container and to capture and analyze the Truck license plate. The system also displays the images on its main display.

A typical installation, shown below, includes a 3 camera system, reading both sides of the container and the truck license plate.

Figure 3.1 : Container and Truck ID Identification system

3.2.Sample Applications

The system can be part of the following applications (a sample of possible use of the system):

·‏ Port Gate

The system is located at the gates of the port, and records the entry or exit of

the truck together with its container. The system collects the truck’s license plate (using See/Car) and attaches it to the container. Both identifications (in form of a short ASCII string) are transmitted to the gate’s computer, and could be viewed and printed at the gate. Additionally, they are transmitted to the port’s computer and are used for data collection.

·‏ Port Load/Unload

The system can be placed at load and unload areas. The system will record the containers that are moved in that area and store them in the port’s main computer.

·‏ Docks

The system can be located at the dockyard, at the side of the ship. The containers will be identified and transmitted to a remote computer.

·‏ Factory Store

The system will record the load/unload of containers at factory store.

3.3. Overview of Container ID

An example of a container ID shown in figure 3.4. Note that this is one of the several container ID formats that the international standard (ref. #2.2) defines.

Figure 3.3: Example of container ID

The container ID is composed of several fields, including the following fields:

1. the shipping company (e.g., “UXX”)

2. the equipment category (always “U” for freight containers)

3. the serial number of the container (e.g., “423697”).

4. the check digit of the first 3 fields (e.g.,”0”)

5. the container type (e.g.,”SE4310”)

Only the first 3 fields are relevant to the identification of the container, and represent a unique identification number for each shipping container. In the above case, this ID is “UXXU 423687”.

The purpose of the See/Container software is to extract the alpha numeric digits out from the picture (by image processing software), verify its correctness (using the check digit as a final verification test), and transmit the identification string.

3.4.Reduction factor

The See/Container recognition software actually performs a huge reduction in the amount of information. It converts a 512 x 512 pixel picture to a {6+4} digit ASCII string transmission (the first two fields, a total of 10 bytes).

The total reduction factor from the raw image to the extracted string is 1:262,144 ! Thus, the system saves a great deal of transmission and post processing when it merely sends the string rather than the entire picture.

Additionally, the system sends the identified string only when it detects that it faces the container, thus further reducing the need to send multiple images of different positions of the markings. This increases the savings of the transmission and storage of the raw images, an incredibly high number. This is the essence of the system.

Combined with a similar recognition package for the carrying truck license plate (See/Car), the system performs a large reduction and automates the traffic recording process which is usually performed manually, thus increasing the processing speed at the gate.

3.5.System advantages

The system has the following advantages over existing manual recorded sites:

· fully automatic process (no man-in-the-loop)

· increases the processing of the container/truck traffic at the congested gates

· the system data collects the traffic history

· handles simultaneously container & truck identification

· provides live view of the gate traffic

The system has the following advantages over other automated solutions:

· simple configuration (few cameras)

· covers all types of containers (20,40,20/20feet and combinations)

· performs recognition while at motion (the container does not stop)

· simple integration into the existing computer resources at the port

· has a high recognition rate (as redundancy on each container)

· has a user friendly display and operation

· a reliable system

· 24 hour operation

· fast response (output in seconds)

· low cost solution

4.Architecture

This section describes the architecture of the See/Container system which identifies both the Shipping Container markings and the Truck License Plate.

4.1.Overview

The system is based on a stand-alone hardware system that includes the following:

· PC running Windows 95 (to run the system’s software)

· See/Container and See/Car 32-bit DLL (to recognize the marking and license plate)

· 3 or 4 cameras with special illumination (to view the container and truck license plate)

· frame grabber (to capture the images)

· I/O (for sensor and gate input and output)

· Communication (for outputting the result)

The following illustration shows this configuration.

Figure 4.1: See/Container System configuration

The operation of the system is as follows: when the moving truck and the containers that it carries enter the detection zone, the sensors are activated and signal to the program (via the IO card) that the container is present. The program now starts the recognition process: a sequence of images are captured according to the container types. After this, the images are sent to the See/Container and See/Car DLLs for container and license plate identification.

After receiving the results and applying certain logic and validation/verification, the program outputs the combined Container & Truck ID results to the display, saving them in a log file and transmitting them out to the port host computer.

4.2.The Image processing DLLs

The system calls 2 image processing DLLs that analyze the captured images of the container and the carrying truck:

· See/Container DLL - an ISO 6346 Container Marking recognition software

· See/Car DLL - a license plate recognition (LPR) package specifically adapted to the country where the system is installed

These 32-bit Windows DLLs perform the following image processing functions:

·‏ Image Enhancement - improves the quality of the captured image

·‏ ID numbers finding - locates the marking (Container) or truck license plate (Truck) within the image

·‏ Characters detection - locates each of the marking’s characters

·‏ Characters identification - identifies each of the detected characters

·‏ Validation - compares the resulting string to the standard formats, and calculates and verifies check digits (for the container ID)

The program analyzes the results of the recognition across the series of the captured images, and combines them into a final recognition of the Truck and its containers, and displays/files/sends their results.

‏

4.3.Camera & Sensors Configuration

The cameras (with their built-in illumination) are installed in a typical configuration as follows: Truck license plate (#1), side cameras (#2,#3) and an optional back side camera (#4).

The sensors are configured as: Truck sensor (a magnetic sensor on the lane), front and rear sensors (optical sensors mounted at the sides of the lane).

The following illustration shows this arrangement:

Figure 4.3: Cameras & Sensor configuration

4.4.Display

The main display is a Window with a predefined arrangement that is displayed on the screen and operates like a console. The user does not need to select any sub-windows (only in installation mode) and thus can see what the system is doing, its status and the recognition results and history.

The system main display is a Windows-95 Window that is composed of several sub-Windows:

· System Menu - contains several options that allow the user to select

· Images display (#1 through #4) - shows the camera images

· History Log window - shows the recent history of identifications

· Identifications Window - shows a graphical presentation of the recognitions and the type of containers

This is shown in the following illustration:

Figure 4.4: System Display

A display example is shown in the following illustration. In this example the rear side image of the container is seen in the main window, the history of containers/trucks are shown in the bottom list, and the system status is shown in the right side.

4.5.‏ Inter-application Messages

See/Container is designed to share the container/truck identifications with other processes. This can be done either by external communication (RS232) or by application-to-application messages. The latter method is implemented by DDE messages that are sent after each identification cycle.

One or more Client applications can intercept the DDE messages. The client process receives the messages and uses the information for various purposes. For example, if the client process is a database interface, it will insert each message data into the database.

5.Detailed Functional Description

This section describes the functions of the container identification system which identifies both the Shipping Container markings and the Truck License Plate.

Note that Hi-Tech Solutions may change these functions on future system releases, and additional functions may be tailored for specific customer requirements.

5.1.General Requirements

A. Intended system use: stand-alone Container and Truck ID identification system.

B. Single Lane - supports one container lane only

C. Container Types - All standard cases: 40 feet, 20 feet front, 20 feet rear, double

20 feet.

D. Moving containers - the system will handle the recognition while the container

is moving, from 5KM/H up to 40KM/H.

5.2.GUI (Graphical User Interface)

A. Camera View Display- 4 cameras video display (see: image handling)

B. Event Log - list of identifications (one row for each car/front/rear container)

C. Sensors status - on/off status

D. System status - state: installation/Idle/capturing/identifying

E. Graphic result display - show picture of car/container with current Ids

1. Display Symbols of Containers/Cars

2. Show recognition results

F. Options setting -

1. TAB Sheets - using tabs sheets to select different catagories

2. Security - have 2 levels of users to prevent simple operator ability to make

changes

G. Welcome Page - upon entry

H. About Dialog - upon request

5.3.Operation Logic

A. Identification

1. See/Container DLL- extracts container marking string

2. See/Car DLL - analyzes the image and returns a recognized plate number

B. Recognition bank - looks on the recent history of all valid or partial identifications in order to maximize the output based on the multiple recognitions

C. Merge Container & Car Ids - logic that takes all identifications and ties them together

5.4.Image Handling

A. Display camera

1. Installation mode Display - (see: Installation mode)

2. Dynamic Display - display the container or car image after the sensor is

activated. The image that is displayed is the one that was identified on

each active camera.

B. Configuration

1. Configure # of captures per application

2. Configure See/Car or See/Container per camera

5.5.Installation Mode

A. Live Video - select one of the cameras

B. Measure Test Area - displays ROI size in # of pixels to determine best zoom (for calibration).

C. Image capture - (see “FILE” section); used for off-line debug and development.

5.6.Sensor Control

A. Sensor types : Car, Front, Rear

B. Display Status of sensor

C. Sensor Configuration

1. NC/NO (normally-connected or normally-open state)

2. Pin # on Parallel connector

3. Active Rise or Fall

5.7.File

A. LOG File

1. View - view the daily log file; option to change and add notes manually

2. Print - print the daily log file

3. Export - save the daily log file (method TBD)

4. Search - search the daily log file for certain car or container ID

B. Images File Save (for development)

1. Start/Stop Record - allow the user to start or stop the image file save

2. Modes:

a) Full image on all active cameras - for debug

b) identified cameras only - for debug

C. Configuration - Save after changing

5.8.Communication

A. Printer - print daily log file

B. RS232

1. Set prefix string

2. Send Car/Front/Rear string

3. Set suffix string

C. DDE Messages - send recognition results to other applications.

5.9.Authorized Cars/Containers List

A. View

B. Add/Delete

C. Print

See/Lane application is also available in a simulated version - which is available on our web-site.

Visit our Home page http://www.I-Cube.co.za and download recognition demo players, see sample applications and get updated information. Qualified customers also receive on-line updates in the support page.

Visit our Home page http://www.I-Cube.co.za and download:

demo applications

See/Lane simulation program

Latest tools and releases (in our on-line support page)

We also recommend to install pcAnywhere with TCPIP connection. This software package will allow us to view the on-line operation, fine-tune the parameters, download images and update the software. See more details in the web site support page.