Character recognition apparatus and method with low-resolution storage for character extraction

• US 5,361,309 A
• Filed: 12/23/1993
• Issued: 11/01/1994
• Est. Priority Date: 09/07/1989
• Status: Expired due to Term

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Sugiyama, Mitsumasa
• Primary Examiner(s): Moore, David K.
• Assistant Examiner(s): Johns, Andrew W.
• Application Number:
  US08/173,861
• Time in Patent Office: 313 Days
• Field of Search: 382/9, 382/63, 382/38, 382/48, 382/18, 382/62, 382/47, 382/56
• US Class Current: 382/177
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches

Abstract

A character recognizer includes a memory for storing low resolution data of a group of characters to be recognized in an image and a memory for storing high resolution data of the group of characters. Wide range image data for the characters stored in the high resolution memory is selected on the basis of narrower range image data for corresponding characters stored in the low resolution memory. Character recognition is performed according to the selected character image data.

Claims

1. A character recognition apparatus, comprising:

first memory means for storing low-resolution image data expanding already stored compressed image data;

second memory means for storing high-resolution data expanding the already stored compressed image data;

extraction means for extracting an area in which one character exists from the data in said first memory means by detecting a frequency of dot-information of the image data in said first memory means representing black dots;

selecting means for converting coordinate data of the extracted area in the first memory means to coordinate data in the second memory means and selecting from said second memory means the high-resolution data corresponding to the converted coordinate data; and

recognition means for recognizing the selected high-resolution data of an image of the selected area in said second memory means corresponding to the area extracted by said extracting means.

2. A character recognition apparatus as claimed in claim 1, wherein said recognition means uses data of a wider range than the corresponding area of said image data stored in said first memory means.

3. A character recognition apparatus according to claim 1, wherein said extracting by said extraction means is performed by counting dot-information of the image data in said first memory in horizontal and vertical directions.

4. A character recognition apparatus according to claim 1, wherein said selecting means selects a larger amended area from the extracted area.

5. A character recognition apparatus according to claim 4, wherein the selected area is amended within a value between 0 and a ratio of the low-resolution to the high-resolution.

6. A character recognition apparatus according to claim 1, further comprising converting means for converting a coordinate value indicating the extracted area to a coordinate value indicating the selected area according to the ratio of the low-resolution to the high-resolution.

7. A character recognition method, comprising the steps of:

storing low-resolution data expanding already stored compressed image data in a first memory;

storing high-resolution data expanding the already stored compressed image data in a second memory;

extracting an area in which one character exists from the low-resolution data by detecting a frequency of dot-information of the image data in the first memory representing black dots;

converting the coordinate data of the extracted area in the first memory to coordinate data in the second memory and selecting from the second memory the high-resolution data corresponding to the converted coordinate data; and

recognizing the high-resolution data of an image of the selected area in the second memory.

8. A character recognition method as claimed in claim 7, wherein the step of recognizing image data corresponding to characters comprises selecting image data of a wider range than the corresponding character area of said image data stored in said first memory.

9. A character recognition method according to claim 7, wherein said extracting is performed by counting dot-information of the image data in said first memory in horizontal and vertical directions.

10. A character recognition method according to claim 7, wherein said selecting step selects a larger amended area from the extracted area.

11. A character recognition method according to claim 10, wherein the selected area is amended within a value between 0 and a ratio of the low-resolution to the high-resolution.

12. A character recognition method according to claim 7, further comprising a step for converting a coordinate value indicating the extracted area to a coordinate value indicating the selected area according to the ratio of the low-resolution to the high-resolution.