System and method for OCR assisted bar code decoding

• US 5,880,451 A
• Filed: 04/24/1997
• Issued: 03/09/1999
• Est. Priority Date: 04/24/1997
• Status: Expired due to Term

• Current Assignee: United Parcel Service Of America Incorporated (United Parcel Service Incorporated)
• Original Assignee: United Parcel Service Of America Incorporated (United Parcel Service Incorporated)
• Inventors: Lei, Ming, Smith, Christopher E.
• Primary Examiner(s): Gross, Anita Pellman
• Assistant Examiner(s): Dunn, Drew A.
• Application Number:
  US08/840,051
• Time in Patent Office: 684 Days
• Field of Search: 235/462, 235/470, 235/463
• US Class Current: 235/462.1
• CPC Class Codes: G06K 7/14 using light without selection of wavelength, e.g. sensing reflected white light G06K7/10831 - G06K7/1097 take precedence
• CPC Class Codes: G06K 7/1473 error correction
• CPC Class Codes: G06V 10/98 Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns

Abstract

A system for using OCR processing to read human readable characters that correspond to an unsuccessfully decoded code word in a bar code symbol. An imaging system captures an image of a label that includes a bar code symbol and corresponding human readable characters. The system attempts to decode the bar code symbol by decoding each of the constituent bar code characters. If a bar code characters is not successfully decoded, the system locates the associated human readable text and segments the text into individual character images. The unsuccessfully decoded bar code character is mapped to one or more of the alphanumeric character images, which are converted into text characters. The resulting ASCII data is used to create a substitute bar code character in the bar code symbology. After the bar code symbol is decoded using the substitute bar code character, the symbol is validated through check summing or other means.

Claims

1. A method for decoding an image of a label, the label including a bar code symbol and human readable characters adjacent thereto, the bar code symbol comprising bar code characters corresponding to the human readable characters, comprising the steps of:

locating the bar code symbol in the image;

attempting to decode a bar code character;

determining whether the bar code character has been successfully decoded;

in response to the bar code character not being successfully decoded,

identifying the human readable characters in the image;

mapping the unsuccessfully decoded bar code character to at least one of the human readable characters;

converting the at least one human readable character into at least one text character; and

creating a substitute bar code character corresponding to the at least one text character,

whereby the substitute bar code character is used to decode the bar code symbol.

2. A method for decoding an image of a label as recited in claim 1, further comprising the step of validating the bar code symbol after all of the bar code characters have been decoded.

3. A method for decoding an image of a label as recited in claim 2, wherein validating the bar code symbol comprises verifying a checksum provided by the bar code symbol.

4. A method for decoding an image of a label as recited in claim 1, wherein identifying the human readable characters comprises the steps of:

identifying a region containing the human readable characters; and

segmenting the region into individual images corresponding to each of the human readable characters.

5. A method for decoding an image of a label as recited in claim 4, wherein identifying the region comprises the steps of:

locating a dark area beneath the bar code symbol; and

scanning a line midway between the bottom of the bar code symbol and the top of the dark area.

6. A method for decoding an image of a label as recited in claim 4, wherein identifying the region comprises the steps of:

scanning a plurality of lines beneath the bar code symbol, the scanned lines being parallel to the bar code symbol and separated from each other by a predetermined number of pixels;

determining the number of transitions between black and white pixels on each of the scanned lines; and

locating the top and bottom of the human readable characters on the basis of the number of transitions between black and white pixels on each of the scanned lines.

7. A method for decoding an image of a label as recited in claim 1, wherein converting the at least one human readable character into at least one text character comprises applying optical character recognition techniques to the at least one human readable character.

8. A method for decoding an image of a label as recited in claim 1, further comprising the steps of:

providing a confidence factor associated with the substitute bar code character; and

selecting the substitute bar code character on the basis of the confidence factor.

9. A system for reading a data from an object, the object including a bar code symbol and human readable characters, the bar code symbol comprising bar code characters corresponding to the human readable characters, comprising:

an imaging system including a camera for capturing an image of the package, the image including the bar code symbol and the human readable characters; and

a label decoding system for processing the image;

the label decoding system being programmed to;

locate the bar code symbol in the image;

attempt to decode each character in the bar code symbol;

determine that one of the bar code characters failed decoding;

locate the human readable characters in the image;

map the failed bar code character to a human readable character;

convert the mapped human readable character into a text character;

create a substitute bar code character corresponding to the text character; and

use the substitute bar code character to decode the bar code symbol.

10. A system for reading a data from an object as recited in claim 9, wherein the label decoding system validates the substitute bar code symbol by verifying a checksum provided by the bar code symbol.

11. A system for reading a data from an object as recited in claim 9, wherein the label decoding system locates the human readable characters by performing the steps of:

identifying a region containing the human readable characters; and

segmenting the region into individual images corresponding to each of the human readable characters.

12. A system for reading a data from an object as recited in claim 11, wherein the label decoding system identifies the region by performing the steps of:

locating a dark area beneath the bar code symbol; and

scanning a line midway between the bottom of the bar code symbol and the top of the dark area.

13. A system for reading a data from an object as recited in claim 11, wherein the label decoding system identifies the region by performing the steps of:

scanning a plurality of lines beneath the bar code symbol, the scanned lines being parallel to the bar code symbol and separated from each other by a predetermined number of pixels;

determining the number of transitions between black and white pixels on each of the scanned lines; and

locating the top and bottom of the human readable characters on the basis of the number of transitions on each of the scanned lines.

14. A system for reading a data from an object as recited in claim 9, wherein the label decoding system converts the at least one human readable character into at least one text character by applying optical character recognition techniques to the at least one human readable character.

15. A system for reading a data from an object as recited in claim 9, wherein the label decoding system is further programmed to:

provide a confidence factor associated with the substitute bar code character; and

select the substitute bar code character on the basis of the confidence factor.

16. A computer-readable medium on which is stored a computer program for decoding an image of a label, the label including a bar code symbol and human readable characters adjacent thereto, the bar code symbol comprising bar code characters corresponding to the human readable characters, the computer program comprising instructions which, when executed by a computer, perform the steps of:

locating the bar code symbol in the image;

attempting to decode the bar code symbol by decoding each character in the bar code;

determining that a bar code character has failed decoding;

identifying the human readable characters in the image;

mapping the failed bar code character to at least one of the human readable characters;

converting the at least one human readable character into at least one text character;

creating a substitute bar code character corresponding to the at least one text character;

using the substitute bar code character to decode the bar code symbol; and

validating the bar code symbol.

17. A computer-readable medium as recited in claim 16, wherein validating the bar code symbol comprises verifying a checksum provided by the bar code symbol.

18. A computer-readable medium as recited in claim 16, wherein identifying the human readable characters comprises the steps of:

identifying a region containing the human readable characters; and

segmenting the region into individual images corresponding to each of the human readable characters.

19. A computer-readable medium as recited in claim 18, wherein identifying the region comprises the steps of:

locating a dark area beneath the bar code symbol; and

scanning a line midway between the bottom of the bar code symbol and the top of the dark area.

20. A computer-readable medium as recited in claim 18, wherein identifying the region comprises the steps of:

scanning a plurality of lines beneath the bar code symbol, the scanned lines being parallel to the bar code symbol and separated from each other by a predetermined number of pixels;

determining the number of transitions between black and white pixels on each of the scanned lines; and

locating the top and bottom of the human readable characters on the basis of the number of transitions on each of the scanned lines.

21. A computer-readable medium as recited in claim 16, wherein converting the at least one human readable character into at least one text character comprises applying optical character recognition techniques to the at least one human readable character.

22. A computer-readable medium as recited in claim 16, further comprising the steps of:

providing a confidence factor associated with the substitute bar code character; and

selecting the substitute bar code character on the basis of the confidence factor.