Method and device for recognizing distribution data on postal packets

• US 6,665,422 B1
• Filed: 07/16/1999
• Issued: 12/16/2003
• Est. Priority Date: 11/12/1996
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Keil, Göran, Seidel, Gert
• Primary Examiner(s): Mehta, Bhavesh
• Assistant Examiner(s): DESIRE, GREGORY M
• Application Number:
  US09/297,975
• Time in Patent Office: 1,614 Days
• Field of Search: 382/101, 382/321, 209/583, 209/584, 209/3.1, 209/900
• US Class Current: 382/101
• CPC Class Codes: B07C 3/20 Arrangements for facilitating the visual reading of addresses, e.g. display arrangements coding stations
• CPC Class Codes: G06V 10/987 with the intervention of an operator
• CPC Class Codes: Y10S 209/90 Sorting flat-type mail

Abstract

The present invention pertains to a method and device for recognizing distribution data on postal packets, small packages or journals during the automatic mail processing, especially to systems by which the automatic handwritten adress recognition process is supplemented and improved by video coding. The distribution data are written on various backgrounds, including pattern backgrounds. After a first unsuccessful optical character recognition attempt, the complete surface of the postal packet is displayed with a grid on a videocoding-screen. Once a picture segment has been selected which contains a larger part of the distribution data, the operator submits an image portion containing said picture segment and the adjacent areas to a new optical character assessment, while making sure that all the data area is included in the image portion. If the second optical character recognition attempt does not succeed, the image portion is reoriented as many times as required for the final video coding to appear on the screen.

Claims

1. A method of recognizing distribution information on mail items, in which an image of a surface of each mail item containing the distribution information is obtained and evaluated by means of automatic OCR recognition methods and video coding, the method comprising:

a) performing a first automatic search of a region of interest (ROI), the ROI being a region of the mail item containing the distribution information, the search including determining an orientation of the distribution information and performing a first automatic OCR recognition;

b) displaying an entire image of the surface containing the distribution information on a video-coder monitor if the first automatic OCR recognition yields an ambiguous result;

c) dividing the image on the video-coder monitor into image segments with a grid;

d) manually selecting the image segment from the displayed image that contains a portion of the ROI via a single keystroke;

e) performing a second automatic search of a new region containing the distribution information, where the new region is designated by the manually selected image segment and includes marginal regions of segments adjacent to the manually selected image segment such that the new region encompasses the distribution information; and

f) performing a second automatic OCR evaluation of the new region containing the distribution information.

2. The method according to claim 1, further comprising, representing only the new region on an enlarged scale on the video-coding monitor such that the distribution information is oriented with a correct orientation and position based on information from the second OCR evaluation, and no video coding is performed, if the new region containing the distribution information is unambiguously evaluated with the second automatic OCR evaluation.

3. The method according to claim 1, wherein the single keystroke is effected in a video-coding station.

4. The method according to claim 1, further comprising visually enhancing the region containing the distribution information, which was ascertained in the first automatic search, on the monitor.

5. The method according to claim 1, further comprising:

determining if the results of the second automatic OCR evaluation are ambiguous; and

performing a third automatic OCR evaluation, which employs the results of the video coding if the results of the second OCR evaluation are ambiguous.

6. The method according to claim 5, further comprising:

determining if the results of the third automatic OCR evaluation are ambiguous; and

performing a fourth automatic OCR evaluation, which employs the results of the third automatic OCR evaluation if the results of the third automatic OCR evaluation are ambiguous.

7. A device for recognizing distribution information on mail items, comprising

means for obtaining images of mail items;

means for automatic OCR evaluation of distribution information contained in the images of the mail items, including means for performing a search of distribution information;

at least one video-coding station that includes a monitor and an input device for video coding by manually selecting an image segment; and

an image controller receiving the images of mail items, generating a grid and a superposed display with the image of the mail item containing the distribution information on the monitor of the video-coding station, generating an enlarged image section, comprising the selected image segment and border regions to all adjacent segments, and

if the evaluation is ambiguous, representing the enlarged image section on the monitor of the video-coding station with the distribution information corrected in terms of orientation and position and transferring the enlarged image section and the distribution information that was ambiguously evaluated by the video coding to the means for automatic OCR evaluation for a second evaluation based on the enlarged image section.