Data recognition equipment and method using partial pattern recognition

• US 5,588,074 A
• Filed: 10/04/1993
• Issued: 12/24/1996
• Est. Priority Date: 04/06/1989
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Sugiyama, Mitsumasa
• Primary Examiner(s): Couso, Jose L.
• Assistant Examiner(s): Bella, Matthew C.
• Application Number:
  US08/131,913
• Time in Patent Office: 1,177 Days
• Field of Search: 382/13, 382/9, 382/30, 382/23, 382/21, 382/39, 382/31, 382/38, 382/33, 382/24, 382/37, 382/190, 382/177, 382/209, 382/229, 382/218, 382/200, 382/217
• US Class Current: 382/209
• CPC Class Codes: G06V 30/242 Division of the character sequences into groups prior to recognition; Selection of dictionaries

Abstract

A data recognition equipment includes an input device for inputting pattern data, and a selection device which recognizes a partial pattern data in the input pattern data and selects a candidate pattern for pattern recognition of the input pattern data. The equipment also includes two comparing devices. The first one compares the selected candidate with the input pattern data at a part other than the partial pattern data. The second one compares the selected candidate and the input pattern data at a part of the partial pattern based on the results of the comparison made by the first comparing device. The equipment also includes a determining device which determines the results of recognition among the selected candidates based on the results of the comparisons by the first and second comparing devices.

Claims

1. Data recognition equipment comprising:

dictionary means for storing partial pattern data and pattern data;

inputting means for inputting segmented pattern data;

recognition means for recognizing partial pattern data among input pattern data from said inputting means by comparison with the partial pattern data stored in said dictionary means;

selection means for selecting a candidate pattern having the recognized partial pattern from said dictionary means;

first comparing means for comparing the selected candidate pattern with the input pattern data for each part other than the recognized partial pattern data and for outputting a distance obtained from the comparison in correspondence to the selected candidate pattern;

second comparing means for comparing the output distance from said first comparing means with a reference value; and

third comparing means for comparing the selected candidate pattern with the input pattern data at a part of the recognized partial pattern if it is determined that the distance is smaller than the reference value by said second comparing means; and

output means for outputting character information representing the candidate pattern compared with the input pattern data by said third comparing means.

2. Data recognition equipment according to claim 1, wherein, when it is determined that the distance is smaller than the reference value, the distance is used as a new reference value in a succeeding comparison by said second comparing means.

3. Data recognition equipment according to claim 1, wherein, when it is determined that the distance is larger than the reference value, a new candidate is selected.

4. Data recognition equipment according to claim 1, wherein said selection means recognizes a partial pattern data on the basis of position data and the number of strokes.

5. Data recognition equipment according to claim 1, wherein said inputting means comprises a tablet.

6. Data recognition equipment according to claim 1, wherein said output means comprises a display device.

7. A data recognition method comprising the steps of:

storing partial pattern data and pattern data;

inputting segmented pattern data;

recognizing a partial pattern data among the input pattern data by comparison with the partial pattern data stored in a dictionary;

selecting a candidate pattern having the recognized partial pattern from said dictionary;

a first comparing step for comparing the selected candidate pattern with the input pattern data at each part other than the recognized partial pattern data and outputting a distance obtained from the comparison in correspondence to the selected candidate pattern;

a second comparing step for comparing the output distance with a reference value; and

a third comparing step for comparing the selected pattern with the input pattern data at a part of the recognized partial pattern if it is determined that the degree of difference is smaller than the reference value in said second comparing step; and

outputting character information representing the candidate pattern compared with the input pattern data by said third comparing step.

8. A data recognition method according to claim 7, wherein, when it is determined that the distance is smaller than the reference value, the distance is used as a new reference value in a succeeding performance of said second comparing step.

9. A data recognition method according to claim 7, wherein, when it is determined that the distance is larger than the reference value, a new candidate is selected.

10. A data recognition method according to claim 7, wherein, in said selecting step, the partial pattern data is recognized on the basis of position data and the number of strokes.

11. A storage medium for storing a program executed in a data recognition apparatus, the program comprising the steps of:

storing partial pattern data and pattern data;

inputting segmented pattern data;

recognizing partial pattern data among the input pattern data by comparison with the partial pattern data stored in a dictionary;

selecting a candidate pattern having the recognized partial pattern from the dictionary;

a first comparing step of comparing the selected candidate pattern with the input pattern data at each part other than the recognized partial pattern data and outputting a distance obtained from the comparison in correspondence to the selected candidate pattern;

a second comparing step of comparing the output distance with a reference value;

a third comparing step of comparing the selected candidate pattern with the input pattern data at a part of the recognized partial pattern if it is determined that the distance is smaller than the reference value in said second comparing step; and

outputting character information representing the candidate pattern compared with the input pattern data in said third comparing step.