Method and device for handwritten letter recognition

• US 4,685,142 A
• Filed: 04/10/1985
• Issued: 08/04/1987
• Est. Priority Date: 04/12/1984
• Status: Expired due to Term

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Nakamura, Yoshikatsu, Hidai, Yutaka, Kurosawa, Yoshiaki, Ooi, Katsunori
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US06/721,906
• Time in Patent Office: 846 Days
• Field of Search: 382/13, 382/24, 382/56, 382/59, 382/33, 178/18-20
• US Class Current: 382/185
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching

Abstract

A handwritten letter recognition method in accordance with the present invention detects, for each subset, input strokes that coincide with the standard strokes, according to the order of the standard strokes of a standard pattern, and carries out in turn the coincidence processing of strokes by removing the coincident input strokes from the input pattern. The recognition processing for the input patterns is carried out by extracting the standard patterns possessing standard strokes that are coincident for all of the strokes of the input pattern. Therefore, it becomes possible to recognize the input letter pattern easily and efficiently without regard to the order of the strokes input, and even for an input letter pattern with large number of strokes. Moreover, the coincidence detection processing of the strokes can be executed easily and accurately even for the case when the number of strokes that form a letter patterns is large or the case of complicated pattern form, since it detects input strokes that will coincide next, by removing the input strokes that are detected as coincident from the input pattern, to extract a standard pattern all of whose strokes coincide for each and every subset.

Claims

1. A handwritten letter recognition method for recognizing a handwritten letter which is inputted by writing, comprising the steps of:

(a) storing stroke information and stroke order information of a candidate letter pattern;

(b) detecting stroke information of a written letter;

(c) reading the stroke information and stroke order information of the candidate letter pattern, and detecting in turn the stroke information of the written letter which corresponds to the stroke information of the candidate letter pattern;

(d) rearranging the stroke information of the written letter to be in an order which corresponds to the stroke information order of the candidate letter pattern; and

(e) processing the written letter information to identify the written letter by comparing the stroke information of the written letter to stroke information of letters having said candidate letters pattern in the order indicated by said stroke information order.

2. A handwritten letter recognition method as claimed in claim 1, in which, the stroke information and stroke pattern information for the written letter are classified and and stored for each one of a plurality of subsets, in said step (c) the stroke information for the candidate letter pattern is read in turn for each one of the subsets, and the stroke information for the written letter which corresponds to the stroke information of the candidate letter pattern is detected in turn, and, in said step (d) of said written letter which correspond to strokes of said candidate letter pattern are rearranged in the order of said stroke order information.

3. A handwritten letter recognition method as claimed in claim 2, in which the subsets are determined by partitioning the standard pattern of the recognition object letter in accordance with the formal characteristics of the standard pattern.

4. A handwritten letter recognition method as claimed in claim 1, in which the stroke information of the candidate letter pattern includes information on the direction and information on the position of each stroke.

5. A handwritten letter recognition method as claimed in claim 1, in which, in said step (a), the stroke information comprises the direction of each written stroke and the position coordinates of the starting point, midpoint, and terminal point of each written stroke.

6. A handwritten letter recognition method as claimed in claim 1, further where step (c) comprises:

memorizing the order of the stroke information of strokes of the written letter which corresponds to strokes of said candidate letter pattern.

7. A handwritten letter recognition device for recognizing a handwritten letter which is inputted by writing, comprising:

means for detecting stroke information of a written letter pattern;

a standard pattern dictionary which stores stroke information and stroke order information of a plurality of candidate letter patterns;

means for arranging the stroke information of the written letter to correspond to the order of the stroke order information of one of the candidate letter patterns; and

means for recognizing said written letter by comparing the stroke information of said written letter to stroke information of letters having said one of said candidate letter pattern in the order of said stroke order information of said candidate letter pattern.