Method and Sytem for Container Identification

• US 20090180683A1
• Filed: 12/27/2007
• Published: 07/16/2009
• Est. Priority Date: 12/27/2007
• Status: Active Grant

• Current Assignee: Asia Vision Technology Limited
• Original Assignee: Asia Vision Technology Limited
• Inventors: Wong, Wing Kin, Sin, Ka Yu, Lee, Chung Mong
• Granted Patent: US 8,189,919 B2
• Time in Patent Office: Days
• US Class Current: 382/156
• CPC Class Codes: G06F 18/251 of input or preprocessed data
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1918 Fusion techniques, i.e. combining data from various sources, e.g. sensor fusion
• CPC Class Codes: G06V 30/274 Syntactic or semantic context, e.g. balancing
• CPC Class Codes: H04N 5/77 between a recording apparatus and a television camera
• CPC Class Codes: H04N 5/781 on disks or drums

Abstract

A method and system for container identification are disclosed. The method comprises obtaining a plurality of digital images of a character sequence on the container, extracting the character sequences from the images, combining the character sequences into at least one identification code candidate, and selecting one of the candidates as the identification code. The system comprises at least one camera and a computer system that is electrically coupled to the camera, whereby when the computer system receives a trigger signal, said computer system takes a plurality of digital images from the camera, produces character sequences as partial recognition results for the plurality of digital images, and combines the partial recognition results together to produce the identification code.

Claims

1. A method of determining an identification code comprising:

obtaining a plurality of digital images that include said identification code from an object;

extracting a character sequence from each said plurality of digital images, each character in said character sequence having a confidence score;

combining character sequences obtained from different said digital images to form at least one identification code candidate; and

selecting the identification code candidate that satisfies a predetermined criterion as said identification code of said object.

2. The method according to claim 1 wherein said combining step further comprises:

composing at least one sequence pair wherein a first sequence is one of said character sequences, and a second sequence is a character sequence different from said first sequence;

determining at least one longest common subsequence for each said at least one sequence pair by identifying common segments found in both said first sequence and said second sequence and concatenating them together;

extending each of said longest common subsequence to form at least one combined sequence by inserting characters to said longest common subsequence, said characters found in said first and second sequences but not in said longest common subsequence; and

choosing a preferred sequence as said identification code candidate, said preferred sequence having the highest average confidence score among said at least one combined sequence.

3. The method according to claim 2 wherein said determining step further comprises:

generating a table of characters matched from said first sequence and said second sequence; and

extracting said longest common subsequence from said table.

4. The method according to claim 1 further comprising:

depositing said identification code candidates that satisfy a pre-specified check-sum formula to a valid pool;

assigning said identification code candidates that fail to satisfy said check-sum formula to an invalid pool;

and said combining step further comprises;

composing at least one sequence pair wherein a first sequence is one of said character sequences, and a second sequence is a character sequence different from said first sequence;

determining at least one longest common subsequence for each said at least one sequence pair by identifying common segments found in both said first sequence and said second sequence and concatenating them together;

extending each of said longest common subsequence to form at least one combined sequence by inserting characters to said longest common subsequence, said characters found in said first and second sequences but not in said longest common subsequence;

choosing a preferred sequence as said identification code candidate, said preferred sequence having the highest average confidence score among said at least one combined sequence; and

repeating said composing, determining, extending and choosing steps using said identification code candidates in said invalid pool as said character sequences until said valid pool is non-empty or until said repeating step has been performed for a predetermined number of times.

5. The method according to claim 1 wherein said object is a container, and said selecting step selects the code candidate that satisfies a container code check-sum formula and achieves a highest average confidence score as the container code for said container.

6. The method according to claim 5, wherein said obtaining step in claim 1 further comprises:

determining a length of said container being operated on;

moving a camera that is mounted on a mobile support to a pre-defined position according to said length of said container, said mobile support coupled to a motor attachable to a container handler;

capturing said plurality of digital images from said camera when said container is being hoisted or lowered to a pre-defined height relative to said container handler.

7. The method according to claim 5, wherein said obtaining step in claim 1 further comprises:

determining a length of said container being operated on;

choosing one of a plurality of cameras that are attached to a container handler according to said length of said container;

capturing said plurality of digital images from said chosen camera when said container is being hoisted or lowered to a pre-defined height relative to said container handler.

8. The method according to claim 1 wherein said object is a vehicle and said identification code is a vehicle license plate number;

said selecting step determining the code candidate that achieves a highest average confidence score as the vehicle license plate number of said vehicle.

9. The method according to claim 1 wherein said object is a vehicle and said identification code is a vehicle license plate number;

said selecting step determining the code candidate that achieves a highest average confidence score and satisfies a pre-specified check-sum formula as the vehicle license plate number of said vehicle.

10. The method according to claim 1 wherein said extracting step further comprises:

locating a segment of digital image that contains said identification code; and

recognizing characters in said segment; and

assigning confidence score for each recognized character; and

combine said recognized character together as said character sequence.

11. The method according to claim 10 wherein said recognizing step uses neural network to perform character recognition.

12. A method of constructing a complete character sequence from a plurality of partial character sequences comprising:

composing at least one sequence pair by extracting a first sequence from said plurality of partial character sequences and a second sequence obtained from said plurality of partial character sequences which are different from said first sequence;

determining at least one longest common subsequence for each said at least one sequence pair, said at least one longest common subsequence comprises at least one segment of character strings found in both said first sequence and said second sequence;

extending each of said longest common subsequence to at least one combined sequence by inserting to said longest common subsequence characters found in said first and second partial sequences but not in said longest common subsequence;

selecting among said at least one combined sequence a sequence that satisfies a pre-determined criterion as said complete character sequence.

13. A system for determining identification code, comprising:

at least one camera; and

a computer system that is electrically coupled to said at least one camera;

whereby when said computer system receives a trigger signal, said computer system takes a plurality of digital images from said at least one camera, produces partial recognition results for said plurality of digital images, and combines said partial recognition results together to produce said identification code.

14. A system according to claim 13 wherein said computer system further comprises:

a recognition module that produces said recognition result; and

an alignment module that aligns said partial recognition results together, said alignment module implementing the longest common subsequence (LCS) algorithm as a basis for alignment.

15. A system for determining container code comprising:

at least one camera mounted on a container handler, said container handler hoisting or lowering a container;

a computer system that is electrically coupled to said at least one camera; and

electronic circuitry electrically coupled to said computer system;

whereby when said cargo container is hoisted or lowered to a specific height, said computer system receives a trigger signal from said electronic circuitry to take a plurality of digital images from said at least one camera, extracts container recognition results from said plurality of digital images, and combines said results together to produce said container code.

16. The system according to claim 15 wherein said at least one camera is mounted on a fixed extension arm;

said fixed extension arm being attached to said container handler and said computer system enables one of said at least one camera to take said plurality of digital images according to the length of said container being hoisted or lowered.

17. The system according to claim 15 further comprising:

a motor attached to said container handler;

a mobile support coupled to said motor, wherein said at least one camera is mounted on said mobile support so that said computer system commands said motor to move said at least one camera to a predefined position according to the length of said container being hoisted or lowered.

18. The system according to claim 17, wherein said mobile support is a translational mobile support longitudinally traversing along the same direction of the side wall of said container being operated on;

one end of said translational mobile support being coupled to said motor and the other end to said at least one camera;

said motor providing a translational movement to said translational mobile support to position said camera assembly to a first position if said container being hoisted is a 20-foot container, and to a second position if said container is longer than 20 feet.

19. The system according to claim 17, wherein said mobile support comprises a rotary support, said camera assembly being coupled to said rotary support at a suitable vertical angle for capturing said container code;

and said motor causing said rotary support to position said camera assembly to a first horizontal angle if said container being hoisted is a 20-foot container, and to a second horizontal angle if said container is longer than 20 feet.

20. The system according to claim 15 wherein said electronic circuitry is a programmable logic controller (PLC).

21. The system according to claim 15, wherein said computer system further comprises a communication module, said communication module transmitting recognized results of said container code to a remote system.