JPS59127174A - Read-out method of font code like kanji or the like - Google Patents

Abstract

PURPOSE:To read out successively the corner base point and the font code from a 4-side edge pattern of a font through detection of an image. CONSTITUTION:A character square 10 constituted of (24X24) dots is formed with a font constituting part 20 and four side edges 30. For instance, the corner base point is decided by four square points with a 4-side edge pattern showing a KANJI ''NAKA (center)'' which is prescribed by 35 sections and 70 points with JISC6226. The corner base point is obtained through detection of image. Then ''35'' is read as the section information from the upper and left sides, and at the same time ''70'' is read out as the point information from the lower and right sides. Then a font ''NAKA'' is read out of a font ROM with correspondence between sections (35-70) and the 2-byte data.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for calculating font codes such as kanji, alphabetical characters, kana characters, and kanji characters using pattern recognition.

Kanji and other fonts are printed by printers using computers, but the development of methods to read the printed fonts and store the codes in memory is generally delayed, although it relies on OCR techniques. An object of the invention is to provide a method for reading out a font code by detecting an image of its four-edge pattern rather than recognizing the pattern of the font itself.

This objective can be achieved by sequentially reading out corner base points and font codes from the four edge patterns of the font by image detection.

Next, preferred embodiments of the present invention will be described with reference to the drawings.

FIG. 1 shows a character corner 10 made up of 24 x 24 dots, 20 being the font component and 30 its four edges.

FIG. 2 shows an example of a four-edge pattern of a font representing the kanji character "中" defined by JISC6226 with 70 points in 35 sections. In this example, the font corner base points are determined by the four corner points,
After finding this base point by image detection, we then read "35" as the ward information from the top side and the left side,
Also, the point information is "70" based on the bottom side and the right side.
Accordingly, the font "chu" is read out from the # font ROM based on the correspondence between the kuten (35-70) and the 2-byte data. If the dot diameter is, for example, 0.3 mm and it is difficult to detect the image directly, a magnifying glass may be used. Since the diameter of the dots is generally larger than the distance between the dots, it is preferable to position the dots at appropriate distances so that the diameters of the pattern dots can be sufficiently recognized. In this example, the dot diameter as a pattern display is very small, for example, 0.13 mm, and does not impair the font. Note that the example of the four-edge pattern of the font is not limited to the illustrated example, and various examples can be applied depending on the size of the font, ease of detection, etc.

Furthermore, the present invention is applicable not only to fonts having a dot structure, but also to printed characters whose four edges are patterned using a mold.

[Brief explanation of drawings]

FIG. 1 shows a heptagon according to the invention, and FIG. 2 shows an example of application of the invention. 10 is the seven corners, 20 is the font, and 30 is the four edges of the font. (3) Figure 1 0 398- Figure 2

Claims (1)

[Claims] A method for reading font codes such as kanji characters, which is characterized by sequentially reading out corner base points and font codes from the four edge patterns of fonts by image detection.