JPS58201139A - Information storing system - Google Patents

Abstract

PURPOSE:To perform accurate identification, by storing a character code in the form of a character pattern, and collating the stored pattern with the same character pattern used in previous character pattern generation and identifying a character. CONSTITUTION:The character code inputted on a keyboard is stored in a memory 4 temporarily through a bus line 2 under the control of a processor 3, and the character code is converted at an interface 7 by a character pattern generator 8 into a character pattern, which is stored on an optical disk 6. The character pattern when read is converted at a read interface 9 by a character pattern reader 10 into the character code, which is outputted on a printer 11 and also sent out through a communication interface 12. It is displayed on a CRT5 during reading and writing at need for corrections and editing. Graphie information is processed through an image reader 13. Consequently, accurate reading and writing are performed even in case of a little bit defect and a mass-storage optical disk is usable.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an information storage method for document files.

Conventional information storage methods of this type are roughly divided into those based on character codes and those based on graphic patterns.The ones based on character codes are configured with magnetic disk devices or flexi-sol disk devices; When using a family of devices, the disadvantage is that you cannot carry it around freely because you cannot replace the disk, and when using a flexible disk device, the capacity is small.
There were drawbacks. In addition, some graphic pattern-based devices are constructed using optical disk devices.
The recording medium has a large capacity, is inexpensive, and is portable, but the bit error rate of the recording medium is 10-3.
Since it is about ~10-', inferior to disks etc. by about 9, it cannot be used to record code information, and therefore it is not possible to perform searches using character codes, making it a full-fledged storage method for document files. could not be used.

In order to eliminate these drawbacks, the present invention makes it possible to record and read code information by attaching, for example, a character pattern generator and a character pattern reader to an optical disk device. Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a diagram showing an embodiment of the present invention, in which 1 is a keyboard, 2 is a system pass line, 3 is a processor, 4 is a memory, 5 is a CRT for display, 8 is a DRAW, (Direct Read After
Write ) type optical disc, T is an optical disc writing interface, 8 is a character pattern generator, 9 is an optical disc continuous output interface, 1G is a character pattern reader, 11 is a printer of a hard copy device, 12 is a communication interface, 13 is an image It's a league.

FIG. 2 is a block diagram showing the details of the structure of the character pattern reader 10 in FIG.
11 is a character pattern memory for comparison, 18 is an encoding unit, 1
9 is a character code buffer memory, and 20 is a graphic buffer memory.

Next, the operation will be explained.

The character code signal input from the keyboard 1 is temporarily stored in the memory 4 under the control of the pulse processor 3 via the system path line 2, and if necessary, modified using the CRT 5Y.
After m collections are performed, they are stored on the optical disc 6. Note that when storing, the character code is not input into the optical disc 6 for 70 times (at the optical disc writing interface 7, the character code is converted into a character pattern figure by the character pattern generator 8, and the character code is stored on the optical disc 6). do.

Next, to the data on the optical disc 6! To print, the character pattern is converted into a character code by the character pattern reader 10 at the optical disk reading interface 9, temporarily stored in the memory 4, and if necessary, after correction 2 editing is performed using the CRT 5, the character pattern is transferred to the printer. 11 or to other terminals via the network via the communication interface 12.

The above processing is for character code information, but in the case of graphic information, it is input as pattern information through the image reader 13Y, and stored as pattern information on the optical disk 6 through the optical disk writing interface 11.

At this time, in the optical disk writing interface 7, the character pattern generator 8 is not involved at all. Graphical information is read out via the optical disk 6 or the optical disk reading interface 9 and the system path line 2 from the CRT 5. This is performed for the printer 111 communication interface 12, etc. At this time, in the optical disc reading interface 9, the character pattern reader 10 is not involved at all, and the pattern information is directly output from the optical disc 6.

The character pattern reader 100 functions are as follows. In the character/figure segmentation section 14 shown in FIG. 2, the character string portion of the pattern information inputted from the optical disk reading interface 9 is extracted and inputted to the single character segmentation section 15. Therefore, a dot pattern matrix for one character is cut out, and in the next character butter/matching section 16, it is compared with all the character patterns stored in the matching character pattern memory 17 and identified as the character with the highest matching degree. , is converted into a character code by the encoding unit 18 and stored in the character code buffer memory 19. On the other hand, the character/figure cutting unit 14 sends the cut figure information to the figure buffer memory 20.

Both buffs 7 notes! The character codes and graphic patterns stored in 719.20 are divided into pages, for example, and input again to the optical disk reading interface 9 for processing.

The functions of the character pattern matching section 16 are as follows.

Matching character patterns is to detect the distance between character patterns.The distance between character patterns is determined by comparing the corresponding pixels of the dot pattern matrix of the two characters in question, and then calculating the contents (white or white). black) is defined as the sum of the number of different pixels. Now, character pattern matching section 1
In B, the pattern matrix for one character sent from the character extraction unit 15 is compared with the character pattern for comparison,
Calculate the distance to all character patterns. Note that the character z and pattern contents of the matching character pattern must completely match those of the character pattern generator 8. If the optical disc 6 is complete and free of errors (and if the character pattern generator 8 → optical disc writing interface 7 → optical disc 6 → optical disc reading interface 8
→If there is no noise on the transmission path between the character pattern readers 10, then the character pattern in question will definitely have a distance of 0 from one of the character patterns for comparison9, and will be identified as that character. However, if the optical disc 6 is incomplete and has bit defects, or if there is noise in the transmission path, there may be no matching character pattern with a distance of 0. The result is to identify the Therefore, if the distance between character patterns can be made sufficiently thick, accurate identification is possible even if a 9-bit defect exists.

In the above embodiment, character pattern figures are generated by the character pattern generator 8, but they may also be generated by using the CPUY. The same applies to other parts.

1 As explained above, 5. This invention converts character hoods into character pattern figures and stores them, and identifies the stored character pattern figures by comparing them with the same character pattern that was previously used for generation. This makes it possible to record and read character code information even using a storage medium with a small number of bit defects, making it possible to use a large-capacity optical disk device for full-scale document files based on code information. It can be used as a storage system, and has the advantage of being less expensive and more portable than conventional magnetic disk devices, and of larger capacity and less expensive than flexible disk drives (FDD). In addition, by making it possible to mix character code information on an optical disc, which conventionally could handle only pattern graphic information, it becomes possible to search for titles, keywords, etc. using character codes, which has great advantages as a storage method for graphic files.

[Brief explanation of the drawing]

FIG. M1 is a block diagram showing one embodiment of the present invention, and FIG. 2 is a block diagram showing details of the character pattern repeater in FIG. 1. In the figure, 1 is a keyboard, 2 is a system pass line, 3 is a processor, 4 is a memory, and 5 is a CRT. 6 is an optical disk, T is an optical disk writing interface, 8 is a character pattern generator, 8 is an optical disk reading interface, 10 is a character pattern reader, 11 is a printer, 12 is a communication interface, 13 is an image reader, 14 is a character Figure cutting is part, 15 is one character cutting part,
16 is a character pattern matching section, 17 is a matching pattern memory, 18 is an encoding section, 19 is a character code buffer memory, and 20 is a graphic buffer 7 memory. Figure 1

Claims (1)

[Claims] In the information storage method of Hatern Information Pace, character codes are converted into character pattern figures and stored, and the stored character pattern figures are identified by comparing them with the same character pattern used to generate the character pattern figure. An information storage method characterized by converting the data into code information and reading it out.