JPS6020277A - Multiple tone wedge data compression system - Google Patents

Abstract

PURPOSE:To transfer and hold data through compression of images with variable density by holding multiple tone wedge image data in multiple tone wedge bit picture buffers, and switching, sending and compressing tone wedge bit picture buffers. CONSTITUTION:Image data are read by the image sensor 23 and held in the image buffer 25 by changing 8-tone wedge 3-bit image data of the stamp 12 to 4 tone wedges and 2 bits at the converter 24. For image data which are held in the image buffer 25, pick-up in the prescribed manner is conducted, data are sent from the upper-rank tone wedge bit picture buffer 25a and they are sent to the registration part 2 through the circuit by compressing them at the compression part 19. Data are sent from the lower-rank tone wedge bit picture buffer 25b and sent to the registration part 2 after compression by successively switching part 26.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field of the Invention) The present invention relates to a compression method for image data read in seal stamp verification, etc., and in particular to a method for improving verification accuracy through compression processing of multi-level image data. Regarding the possible multi-gradation data compression method.

(b) Technical Background In recent years, facsimile machines that read documents such as 'ζ2 drawings and transmit image data have rapidly become popular as part of the streamlining of office work.

On the other hand, in financial institutions and the like, a method is being attempted in which a seal matching device automatically matches a customer's stamp on a bill, check, etc. with a registered seal.

In stamp verification, image data corresponding to the stamp is obtained by projecting the stamp to be verified onto an image sensor through an optical lens.
This is to compare it with the image data of the registered seal.

In the seal verification system, not only the seal stamped on the seal registration slip, but also the address written on the slip.

Names and other information are also read as image data and recorded in a registration file, searched by inputting account numbers, etc., and output on a display for use.

There is no problem as long as the character data can be read even in two-tone data, but stamps often have shading due to uneven adhesion and pressure of vermilion, so read the shading depending on the position within the registration mark and verification mark. Since delicate shading cannot be obtained with two-tone data, attempts are being made to improve the reading density and increase the number of gradations.

(C) Prior Art and Problems Hereinafter, a conventional method will be explained using a seal registration system as an example with reference to FIGS. 1 and 2.

FIG. 1 is a block diagram showing a conventional method, and FIGS. 2 and 3 are explanatory diagrams of FIG. 1.

In FIG. 1, the ranges indicated by dashed dotted lines are the reading section 1 and the center 2, respectively.

Further, 11 is a document, 12 is a stamp, 13 is a transport section, 14 is a light source lamp, 15 is a lens system, 16 is a mirror, 17 is an image sensor, 18 is an image buffer, 19 is a compression section, 21
indicates a file control unit, and 22 indicates a registered file.

The compression section 19 of the reading section 1 and the file control section 21 of the registration section 2 are connected via a line.

The document 11 is transported by the transport section 13 at a constant speed.

A light source lamp 14 is arranged so as to emit light onto the stamp 12 and the document 11.

A lens system 15. A mirror 16 and an image sensor 17 are provided.

The irradiated light is not reflected by the stamp 12 and the characters, but is reflected by the white parts other than the stamp 12 and the characters, and the reflected light is collected by the lens system 15 and sent to the image sensor 17 via the mirror 16. .

That is, images of the seal 12 and characters recorded on the document 11 are projected onto the light receiving surface of the image sensor 17, and the optical image is converted by the image sensor 17 into two-tone image data.

Further, the image buffer 18 holds image data of the seal 12 and characters, and has a function of cutting out a part or all of it into a registered cutting pattern and sending it to the compression section 1/9.

The compression unit 19 has a function of compressing the image data sent from the image buffer 18 and sending it out to the line.

The file control section 21 has a function of registering data sent from the reading section 1 via the line into the registration file 22.

The registration file 22 is a large-capacity electronic file composed of, for example, a magnetic disk, and holds registered image data.

With such a configuration and function, when registering the stamp 12, when the transport section 13 of the reading section 1 is driven and the document 11 is transferred in response to a reading instruction from the operator's keyboard (not shown), the stamp 12 is registered. The image is read by the image sensor 17 and sent as image data to the image buffer 18 and held there. Then, the data is cut out, sent out, compressed by the compression unit 19, sent to the file control unit 21 via a line, and registered in the registration file 22. The image data in the image buffer 18 is then erased.

In this case, the compression methods include the Modify Huffman (MH) method, which is a line-by-line run-length compression method, the Modify Read (MR) method, which compresses the data of the corresponding row based on the previous row data, Alternatively, an MH-MR method using both methods may be used.

In this way, the seal 12 can be registered.

However, in this method, as shown in Figure 2, 4)
- The image data held in the Jibasofa 18 has two gradations, and one pixel a is expressed by one bit, and has a value of 0# or ゛1〃, where 0'' represents white and ゛1# represents black. However, the disadvantage is that it is not possible to express the delicate shading of the stamp 12.

(d) Purpose of the Invention The purpose of the present invention is to solve the above-mentioned drawbacks, and to provide a multi-tone data compression method that can improve matching accuracy through compression processing of multi-tone image data. .

+e) Structure of the Invention The present invention provides a plurality of gradation bit screens having a storage area corresponding to each read pixel and holding multi-gradation data in each corresponding storage area, and a plurality of gradation bit screens. This is a multi-gradation data compression method that is equipped with a switching means for switching and reading the gradation data, and compresses the multi-gradation data held in multiple gradation focus screens for each gradation bit screen by the switching means. can be achieved.

(f) Embodiment of the Invention An embodiment of the invention will be described below with reference to FIGS. 3 to 5. FIG. 3 is a block diagram showing an embodiment according to the present invention;
4 and 5 are explanatory diagrams of FIG. 3. The same reference numerals refer to the same objects throughout the design.

In FIG. 3, 23 is an image sensor, 24 is a converting section, 25 is an image buffer, and 26 is a switching section.

The image sensor 23 has a function of reading the image data of the stamp 12 as 8-gradation 3-bit data.

The conversion unit 24 is located on the 8th floor I! It has a function of converting 13-bit image data into four values by changing the slice level according to the density of the stamp 12, converting it into four-gradation 2-bit image data, and sending it to the image buffer 25.

The image buffer 25 is an upper level steel bit screen buffer 25a
It is composed of a lower gradation focus screen buffer 25b, and holds multi-gradation image data.As shown in FIG.
5a and the corresponding 2 of the lower gradation bit screen buffer 25b.
Represented by pi l-b, c, 00", 01", 10",
It has a value of 11", where 00" represents white, -0'l" light gray, 10" dark gray, and 11" black.

Therefore, when handling 8-gradation data, for example, the image buffer 25 will have three screen buffers.

It also has a function of sending image data to the compression section 19.

When sending image data from the image buffer 25, the switching unit 26 first selects the upper gradation focus screen buffer 25a.
It has a function of switching between sending out the image data of 1 and then sending out the lower gradation bit screen buffer 25b.

With such a configuration and function, when registering the stamp 12, the image data of 8 floors, 1 m3 bits read by the image sensor 23 of the stamp 12 is converted into 2 bits with 4 gradations, and is stored in the image buffer 25. be done.

The image data held in the image buffer 25 is cut out in a predetermined manner and stored in the upper gradation bit screen buffer 25a.
Data is sent out from the lower gradation bit screen buffer 25b, compressed by the compression unit 19, and sent to the registration unit 2 via the line, and then, by switching the switching unit 26, data is sent out from the lower gradation bit screen buffer 25b, compressed, and sent to the registration unit 2. sent to. The MH-MR method is used as the compression method in this case.

Then, as shown in FIG. 5, each is stored in the registration file 22. That is, upper and lower steel bit screen buffers 25a,
Compressed data e for each screen of 25b.

f is stored.

In this way, multi-gradation data is held in multiple gradation focus screen buffers, and the upper gradation bit/7th screen buffer 1
By compressing the data more sequentially, it is possible to retain data expressing the shading of the stamp 12, and it is possible to reproduce it at the time of retrieval, thereby increasing the verification accuracy at the time of verification.

(g) Detailed Description of the Invention According to the present invention, multi-gradation image data is held in a plurality of 1!1 little bit screens, the gradation focus screen buffer is switched and the data is sent out, and the data is sent out in two gradations. Since multi-tone image data can be stored and reproduced by compressing using the same compression method as in There is an effect.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the conventional method, Fig. 2 is an explanatory diagram of Fig. 1, Fig. 3 is a block diagram showing an embodiment according to the present invention, and Figs. 4 and 5 are explanatory diagrams of Fig. 3. It is. In the figure, l is a reading section, 2 is a registration section, 11 is a document,
12 is the stamp, 17.23 is the image sensor, 18.25
19 is an image buffer, 19 is a compression section, 21 is a file control section, 22 is a registered file, 24 is a conversion section, 25a is an upper gradation bit screen buffer, 25b is a lower gradation bit screen buffer, and 26 is a switching section. show. Mine 1 Inhaya 2 Diagram

Claims (1)

[Claims] A multi-level data compression method that reads an image written on a medium as multi-level image data and compresses the read multi-level image data, and has a storage area corresponding to each read pixel. , comprising a plurality of gradation bit screens that hold the multi-gradation data in respective storage areas, and a switching means for switching and reading out the plurality of gradation focus screens, and a switching means for switching and reading out the plurality of gradation bits. A multi-gradation data compression method characterized in that multi-gradation data held on a screen is compressed for each gradation bit screen.