JPS62144476A - Image information binary-coding system - Google Patents

Abstract

PURPOSE:To eliminate binary-coding errors due to the effect of surrounding picture elements by detecting 4 picture elements at the upper/lower/left/right position of a noted picture element in binary-coding and reading picture information and correcting the resulting noted picture element level while referring to a correction table. CONSTITUTION:After an analog image from a CCD 11 is digitized, the result is subjected to sample holding by an n-bit latch 13 and a preceding line and a noted binary-coded line are stored in a RAM 14. As soon as a succeeding line is read, after a picture element R2 is extracted from the RAM 14, the preceding line is given in a RAM 15, one bit is shifted by a delay circuit (DEL) 16 to extract a picture element R1. A signal of the binary-coded line is shifted by 1 bit in a DEL 17 to extract a noted picture element D and a picture element R3 is extracted via a DEL 18. A signal of the preceding line is given to a DEL 19, one bit is shifted and a picture element R4 is extracted from the RAM 14. Then the noted picture 1 is divided into a table ROM 21 receiving a longitudinal component (R1, D, R3) and a table 22 receiving a lateral component (R2, D, R3), they are corrected separately, the result is put in a decode ROM 23, a slice level is given to apply binary-coding finally.

Description

[Detailed Description of the Invention] [Summary] The present invention provides an apparatus that handles image information, such as a facsimile machine, when reading image information by binarizing it. Four pixels are detected, and the pixel level of interest of the detection result is corrected by referring to a preset correction table.

This makes it possible to eliminate the binarization error 9 due to the influence of surrounding pixels 1-9 on the pixel of interest.

[Industrial application field]

The present invention relates to an image information binarization method that corrects the influence of surrounding pixels on a pixel of interest when binarizing image information of a facsimile J or the like.

[Conventional technology]

Conventionally, when converting image information into electrical signals in facsimiles and the like, there are generally two methods: using a contact type image sensor and using an optical system to form an image on an OOD (charge-coupled device). The former is currently expensive, so A
The latter is advantageous in devices that can read six originals, and is currently the mainstream.

[Problem that the invention seeks to solve]

Since the above-mentioned method of forming an image on 00D uses an optical system, the MTF of the lens, diffused reflection due to light from a light source, etc. have an adverse effect, causing a reduction in image quality. When this effect becomes significant, a phenomenon occurs in which white or black pixels are crushed by surrounding black or white pixels. This phenomenon is illustrated in Figure 4 (α) to (c).
This is explained by:

The method shown in the figure uses COD to compare an analog image signal with a slice level to binarize image data.

Figure (α) shows examples A and B of the states of white and black (diagonal line) pixels in the binarization line of interest and the lines before and after it, and the solid line in Figure (6) shows the image detected by COD. signal 1
The dotted line is the slice level, and FIG. 2C shows the resulting binarized image data. In this case, the binarization line in the same figure (G) that is attracting attention and the two-value line in the same figure (C)
Comparing the digitized image data, it is clear that it is affected by the above-mentioned phenomenon. That is, when there is one black pixel in a completely white surrounding area, as in surface condition B, the waveform of the image signal becomes dull and becomes white. Conversely, if there is one white pixel in a black pixel, the image will become black for the same reason. Furthermore, even in people with poor image quality, the pixels at both ends of the image are significantly affected by white, so black is detected, but if the conditions are bad, the pixels at both ends may become white.

SUMMARY OF THE INVENTION An object of the present invention is to provide an image information binarization method that corrects the influence of surrounding inverted pixels on pixels to be binarized.

[Means for solving problems]

In order to achieve the above object, in the present invention, when image information is binarized and read, four pixels on the front, back, left and right sides of the pixel of interest to be binarized are detected, and the pixel of interest as a result of this detection is The correction is performed by referring to a correction table in which levels are set in advance.

[For production]

In order to correct the influence of inverted pixels around the pixel of interest, this state is detected and a correction table is set in advance with the level changes in the front, back, left, and right directions relative to the pixel of interest as addresses, and this is referred to to correct the pixel of interest. Correct things. This makes it possible to eliminate binarization errors of the pixel of interest due to the influence of surrounding pixels.

〔Example〕

FIGS. 1(α) to 1(C) are configuration explanatory diagrams of an embodiment of the present invention.

In the same figure (b), the analog image signal sent from 0OD11 is converted into digital data by the A/D converter 12 (n bits), sampled and held by the n bit latte 13, and the image 1 of the same figure (tL) is stored in the RAM 14. The previous line and the binarized 2-in of interest are stored as shown in . Next, at the same time as reading the line after the third line, first R
After extracting the pixel R2 of the unification line at 2 from the output of AM14, the previous line receives pixel R2 from RAM14 almost simultaneously.
Delay circuit (DEL) 1 (5 shifts 1 bit and takes out pixel 1. In the binarization line, DBL 17 shifts 1 bit and takes out the pixel of interest, and then DEL 1
8, the pixel FL3 is taken out. DEL1 on the back line
The pixel R4 shifted by 1 bit is taken out from the FLAM 14 via the pixel R4.

Next, the relationship between the pixel of interest and the surrounding pixels 1 to 4 is made into a table. For example, each pixel level is expressed in 16 levels with 4 pits. Therefore, the state of pixels 1 to 4 is 4X4 = 1.
Although it is possible to search for the pixel of interest using 6-pit addresses, the number of addresses becomes large and the configuration becomes complicated.

Therefore, in the present application, in FIG.
R2, D, 几3) are input to the table ROM 22, and after correcting each separately, the decode ROM 23
The image is then input into the image, given a slice level, and finally binarized.

FIG. 2 is a detailed explanatory diagram of the example shown in FIG. 1(c).

In the figure, an example is shown in which the pixel of interest and surrounding pixel levels stored in the RAM 14 are detected as in image 1.

That is, the address AAA in which the vertical axis (1.D, R4) is expressed in hexadecimal is entered into the table ROM 21, and the correct target pixel level 10 is output. Similarly, horizontal (fL2.D, R
3) Table the address FAF expressed in hexadecimal.
M22 and outputs the correct target pixel level 6.

Both table ROM O output decimal number 10 (101
0), 6 (0110), and the slice level decimal number 9 (1001) as addresses are entered into the decode ROM 23 to obtain the final level 0 of the pixel of interest.

In this case, it can be seen that even though the pixel of interest is Ol, that is, black, it has been transformed into a 10 due to the influence of the surrounding white.

FIG. 6 is a configuration explanatory diagram of another embodiment of the present invention.

In the same figure, similarly to FIG. 1(b), 6 lines of data of image 1 shown in FIG. ．． Rear twin RAM 33
Store in. After reading 6 lines, MPU 5
4 searches the combination table 0M65 using the levels of the pixels on the front, rear, left, and right sides of the pixel of interest to be binarized as addresses in the same manner as described above, and determines the correct level of the pixel of interest based on the slice level and outputs it.

In order to reduce the number of addresses in the table 0M35 as in FIG. 1(c), it can be divided into two table ROMs and configured as a combination of fewer address bits. That is, the timing configuration by the hardware shown in FIG. 1 is implemented by a program by the MPU, and the table ROM is referred to.

〔Effect of the invention〕

As explained above, according to the present invention, when image information such as a facsimile is binarized and read, surrounding pixels to the pixel of interest to be binarized are detected, and the level of the pixel of interest as a result of this detection is determined in advance. The υ correction is performed by referring to the set correction table. As a result, it is possible to almost eliminate binarization errors caused by the influence of inverted pixels surrounding the pixel of interest, which conventionally occur, and this greatly contributes to improving image quality.

[Brief explanation of drawings]

FIGS. 1(a) to 1(c) are configuration explanatory diagrams of embodiments of the present invention,
Figure 2 is a detailed explanatory diagram based on an example of the embodiment, Figure 3 (α)
, (b) is an explanatory diagram of the configuration of another embodiment of the present invention, and FIG. 4 is an explanatory diagram of the conventional example. In the figure, 1 is image data in memory, 11 is COD, and 12 is n-bit A/ D near: y bark, 13 is n bit latch, 14.15.51-36
is RAM, 16 to 19 are delay circuits, 21, 22.3
5 is a table ROM, 23 is a decode ROM, 34
indicates MPU.

Claims (1)

[Claims] In a processing device that binarizes and uses image information, when the image information is binarized and read, means for detecting four pixels on the front, rear, left, and right sides of a pixel of interest to be binarized. An image information binarization method, comprising: means for correcting the pixel level of interest of the detection result by referring to a preset correction table.