-
1.发明专利
公开(公告)号：CA2087502A1
公开(公告)日：1993-10-15
申请号：CA2087502
申请日：1993-01-18
Applicant: IBM
Inventor： CHEFALAS THOMAS E , TAPPERT CHARLES C
Abstract: A method for use in a handwriting recognition system, and apparatus for accomplishing the method, that generates a User set of prototype symbols. The method includes the steps of (a) inputting one or more strokes for representing a predetermined symbol; and (b) comparing characteristics of the inputted strokes to stroke characteristics associated with one or more predetermined Starter prototype symbols. If the stroke characteristics match to within a predetermined threshold, the method further includes a step of (c) averaging the characteristics of the inputted stroke or stroke with the characteristics of a corresponding one or ones of the strokes of the matching one of the Starter prototype symbols to form a User prototype symbol. Further steps include (d) storing the User prototype symbol for subsequent use; and (e) deleting one or more Starter prototype symbols of the same label as the input symbol. If the stroke characteristics do not match, the method further includes the steps of (f) storing the characteristics of the one or more inputted strokes as the User prototype symbol for subsequent use; and (g) deleting the one or more Starter prototype symbols. A counter is maintained and deletes the Starter set only after a predetermined number of instances of the symbol have been entered by the user.
-
2.发明专利
公开(公告)号：CA2087502C
公开(公告)日：1996-05-14
申请号：CA2087502
申请日：1993-01-18
Applicant: IBM
Inventor： CHEFALAS THOMAS E , TAPPERT CHARLES C
Abstract: A method for use in a handwriting recognition system, and apparatus for accomplishing the method, that generates a User set of prototype symbols. The method includes the steps of (a) inputting one or more strokes for representing a predetermined symbol; and (b) comparing characteristics of the inputted strokes to stroke characteristics associated with one or more predetermined Starter prototype symbols. If the stroke characteristics match to within a predetermined threshold, the method further includes a step of (c) averaging the characteristics of the inputted stroke or stroke with the characteristics of a corresponding one or ones of the strokes of the matching one of the Starter prototype symbols to form a User prototype symbol. Further steps include (d) storing the User prototype symbol for subsequent use; and (e) deleting one or more Starter prototype symbols of the same label as the input symbol. If the stroke characteristics do not match, the method further includes the steps of (f) storing the characteristics of the one or more inputted strokes as the User prototype symbol for subsequent use; and (g) deleting the one or more Starter prototype symbols. A counter is maintained and deletes the Starter set only after a predetermined number of instances of the symbol have been entered by the user.
-
公开(公告)号：CA1223366A
公开(公告)日：1987-06-23
申请号：CA481844
申请日：1985-05-17
Applicant: IBM
Inventor： FOX ABIJAH S , GREANIAS EVON C , KIM JOONKI , TAPPERT CHARLES C
Abstract: A SYSTEM FOR AUTOMATIC ADJUSTMENT AND EDITING OF HANDWRITTEN TEXT IMAGES A system of information formatting and editing using handwritten input data on an all points addressable tablet. Input words are segmented then reformatted in a consistent manner of spacing and horizontal alignments to provide a legible output while still maintaining the handwritten format. Editing functions such as delete, add and move are provided in an interactive manner.
-
公开(公告)号：CA1222055A
公开(公告)日：1987-05-19
申请号：CA478624
申请日：1985-04-09
Applicant: IBM
Inventor： TAPPERT CHARLES C
Abstract: RECOGNITION SYSTEM FOR RUN-ON HANDWRITING A method of processing a word with the segmentation and recognition steps combined into an overall scheme. This is accomplished by a three step procedure. First, potential or trial segmentation points are derived. This is done in a manner so as to ensure that essentially all true segmentation points are present but also includes extra or not true segmentation points. Second, all combinations of the segments that could reasonably be a character are sent to a character recognizer to obtain ranked choices and corresponding scores. Finally, the recognition results are sorted and combined so that the character sequences having the best cumulative scores are obtained as the best word choices. For a particular word choice there is a corresponding character segmentation, simply the segment combinations that resulted in the chosen characters. With this recognition scheme the initial character segmentation is not final and need not be highly accurate, but is subject to a lesser constraint of containing the true segmentation points.
-
-
-
Search Results
4
records
(took 0.015 seconds)
