GB1166542A - Improvements in or relating to Character Recognition Systems - Google Patents
Improvements in or relating to Character Recognition SystemsInfo
- Publication number
- GB1166542A GB1166542A GB6873/67A GB687367A GB1166542A GB 1166542 A GB1166542 A GB 1166542A GB 6873/67 A GB6873/67 A GB 6873/67A GB 687367 A GB687367 A GB 687367A GB 1166542 A GB1166542 A GB 1166542A
- Authority
- GB
- United Kingdom
- Prior art keywords
- points
- character
- centre
- segments
- nearest
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/16—Image preprocessing
- G06V30/168—Smoothing or thinning of the pattern; Skeletonisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Abstract
1,166,542. Character recognition. MULLARD Ltd. 29 March, 1968 [13 Feb., 1967], No. 6873/67. Heading G4R. A character thinning system applies two intersecting raster scans to a character to find points lying substantially at the centre of character strokes or parts. A character, stored as a digitized matrix, is scanned in a horizontal and in a vertical raster, the raster pitches being comparable to stroke width so that scanned rows and columns of the matrix are separated by a number of unscanned rows and columns. During each scan, the co-ordinates of the centre point of each series of four or more consecutive black (character) points are noted (an isolated white point being treated as black). Any two centre-points, one from the vertical raster and one from the horizontal, which are too close together are next replaced by a new centre point midway between them. All surviving centre points are now joined by straight line segments by starting at one centre point, finding the two remaining centre points nearest to it and the two nearest to each of the latter (excluding the last two already used) and constructing all segments between these points which are " compatible " with the character i.e. do not pass through more than two consecutive white points. The straightest two-segment line found is extended by successive application of the " two nearest points " technique and segment construction, priority being given to branches which are closest to the original direction. When such a branch ends, other branches are extended in a similar way until all compatible segments between centre points have been constructed to form a line-thinned character (consisting of the segments) which is used for recognition and is also displayed together with the original character. The above algorithm is executed by suitably programming a computer. Line ends, junctions and loops detected during the thinning are retained for use in the recognition. As a modification, the character may be undigitized, the centre points being obtained as minima, but the centre points are then processed digitally as above.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB6873/67A GB1166542A (en) | 1967-02-13 | 1967-02-13 | Improvements in or relating to Character Recognition Systems |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB6873/67A GB1166542A (en) | 1967-02-13 | 1967-02-13 | Improvements in or relating to Character Recognition Systems |
Publications (1)
Publication Number | Publication Date |
---|---|
GB1166542A true GB1166542A (en) | 1969-10-08 |
Family
ID=9822342
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB6873/67A Expired GB1166542A (en) | 1967-02-13 | 1967-02-13 | Improvements in or relating to Character Recognition Systems |
Country Status (1)
Country | Link |
---|---|
GB (1) | GB1166542A (en) |
-
1967
- 1967-02-13 GB GB6873/67A patent/GB1166542A/en not_active Expired
Similar Documents
Publication | Publication Date | Title |
---|---|---|
GB1171627A (en) | Improvements in or relating to Character Recognition Machines | |
EP0005972B1 (en) | Infra-red target detection arrangements | |
GB1328677A (en) | Display apparatus | |
US6965703B1 (en) | Optical scanner and software for correcting distorted images | |
GB1045929A (en) | Data display system | |
KR950009855B1 (en) | Data compression apparatus and method | |
GB1442269A (en) | Character recognition systems | |
GB1232329A (en) | ||
GB1058720A (en) | Apparatus for character recognition | |
GB1166542A (en) | Improvements in or relating to Character Recognition Systems | |
ES8106816A1 (en) | Apparatus and method for providing electron beam patterns using expanded beam array. | |
GB1303181A (en) | ||
GB1174340A (en) | Character Recognition Apparatus. | |
GB1311030A (en) | Apparatus for recognising graphic symbols | |
GB1121023A (en) | Character recognition apparatus | |
GB1442544A (en) | Apparatus for use in character generation | |
GB2021905A (en) | Display systems and control systems therefor | |
US3490002A (en) | Complete scanning cycle detector for character recognition systems | |
JPS615383A (en) | Character pattern separating device | |
GB2019154A (en) | Improvements in or relating to display systems | |
JPH022189B2 (en) | ||
Borgefors et al. | Multiresolution representation of shape in binary images | |
JPS62210596A (en) | Split pattern matching method | |
GB1402008A (en) | Television image-display system | |
SU100598A1 (en) | TV image decomposition method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PS | Patent sealed [section 19, patents act 1949] | ||
PLNP | Patent lapsed through nonpayment of renewal fees |