JPS58104571A - Picture signal binary-coding system - Google Patents

Abstract

PURPOSE:To attain binary-coding taking the resolution characteristics of the reading system into consideration, by discriminating white and black picture for objective picture elements, based on the level of picuture elements before and after the objective picture element and its level. CONSTITUTION:An A/D conversion circuit 1 quantizes an analog picture signal (a). A latch circuit 2 latches a quantized picture signal 3 for for each one picture element and inputs an output (d) to a latch circuit 3. The latch circuit 3 is latched at each one picture element in a timing delayed for a prescribed time from the latch timing of the circuit 2. A digital comparator 4 compares the output (d) of the circuit 2 with an output (e) of the circuit 3. A latch circuit 5 latches an output (f) of the comparator 4. A discrimination circuit 6 performs discrimination of white and black signals depending on an output (g) of the circuit 5, an output (b) of the comparator 4, the output (e) of the circuit 3 and an output (i) of a line memory 7. In this case, the discrimination is done by taking the level for the preceding and succeeding picture element as well as the level itself of the guantized picture signal (c) of the picture element to be discriminated, into consideration.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image signal binarization method for binarizing an image signal obtained by reading a document into two values of white or black in a facsimile machine or the like.

In this type of image signal binarization method, if binarization is simply performed using a fixed threshold value, thin lines or pale lines in the document cannot be reproduced due to the resolution of the document reading system.

Therefore, in the conventional image signal binarization method, consideration is given to the resolution in the scanning direction by detecting a level change point in an image signal sequence and determining black and white based on this change point. Many of them were able to reproduce Ival lines and faint lines in manuscripts. However, in such a method, it is necessary to configure an analog differentiating circuit and an integrating circuit in order to detect the change point, so the precision of the components becomes a problem, and the image signal is sampled and held once. The disadvantage is that stability and reproducibility are poor because there is a necessary force to approximate the ideal waveform through the process.

The present invention has been made in order to eliminate the above-mentioned drawbacks of the conventional technology, and it is possible to realize binarization taking into account the resolution characteristics of the reading system using a digital circuit configuration, and to achieve stability without being affected by variations in components. The present invention also aims to provide an image signal binarization method that can improve reproducibility.

The present invention will be described below based on embodiments shown in the drawings.

FIG. 1 shows an embodiment of a circuit for realizing the image signal binarization method according to the present invention, and FIG. 3 shows a timing chart of the same embodiment. 1 is an A/D conversion circuit having (L-1) analog comparators, which converts analog image signals &
By inputting the reference voltage S and the reference voltage S, the analog image signal a is quantized to L level. Below, as an example, L=5
and the quantized image signal C output from the A/D conversion circuit 1
is a book classified into one of T1 to T6 as shown in FIG. 2 is a latch circuit that launches the quantized image signal C output from the A/D conversion circuit 1 pixel by pixel, and 3 is a latch circuit that launches the output d of the launch circuit 2 at a timing delayed by a certain period of time from the latch timing of the latch circuit 2. This is a latch circuit that latches each pixel. 4 is a digital comparator that compares the output d of the latch circuit 2 and the output e of the launch circuit 3; and 5 is a latch circuit that latches the output f of the comparator 4. 6 is a determination circuit, and the output q of the latch circuit 5,
Using the output b of the comparator 4, the output e of the latch circuit 3, and the output i of the line memory 7 as input, it determines whether each pixel of the analog image signal a is black or white according to an algorithm that will be explained in detail later, and converts it into a binary signal. Output. The line memory 7 is written with the binary signal outputted from the determination circuit 6 by the memory write pulse j, while
The binarized signal of one line before is outputted as the output signal i.

FIG. 3 shows signals of the same main part, and 1 to 1 correspond to signals c to j, respectively.

Next, the operation of this embodiment will be explained. Let the levels of the quantized image signal C of the (N-1), N, and (N+1)th pixels in a certain line be CN-1pCN and CN+1, respectively, and the black and white level for the NIIth pixel in the line immediately before the certain line. If the determination result is Hi, the determination circuit 6 determines that the output e of the latch circuit 3 is CN, and the latch circuit 6
is the magnitude relationship between CN and CN-1 (CN10N,), the output f of the comparator 4 is the magnitude relationship between CN and CN+1 (CN/CN+1), and the output i of the line memory 7 is H
6, a black/white determination is made for the Nth pixel of the certain line according to the algorithm shown in Table 1 (HN indicates the determination result).

Margin 1 Table 1 below As can be seen from Table 1, first of all in the main scanning direction, the black and white determination involves not only the level CN of the quantized image signal C of the pixel to be determined, but also the level CN of the quantized image signal C of the pixel to be determined. Binarization is performed in consideration of the magnitude relationship between 9r1 and CN+1 regarding the pixels before and after the exponent pixel, so the binarization is performed in consideration of the resolution in the main scanning direction.

In addition, regarding the sub-scanning direction, if the determination result Hi for the Nth pixel of the previous line is white, CN is T6
If and only if CN and CN-1 and CN+1 are large/
The determination result HN is white regardless of the h relationship, but H'
If N is black, CN and CN- even if CN is T4.
1 and CN+1":"Regardless of the magnitude relationship, the determination result HN is determined to be white.

For this reason, in this embodiment, binarization is performed taking into account the resolution in the sub-scanning direction, and the inconveniences such as small characters being reproduced in a slightly blurred manner as in the prior art are reduced.

Note that in order to more precisely consider the resolution in the main scanning direction, black and white determination may be performed by also taking into account the level of the quantized image signal C of the pixels two or more before and after the pixel to be determined.

Furthermore, in order to more precisely consider the resolution in the sub-scanning direction, it is sufficient to consider the level of the same pixel in the next line.

As described above, the image signal binarization method according to the present invention quantizes the image signal to be binarized into L (L≧3) steps, and determines the level itself of the target pixel in this quantized image signal, as well as the target pixel. level and M (before and after the target pixel)
By determining whether the target pixel is black or white based on the magnitude relationship between the levels of M≧1) pixels, binarization that takes into account the resolution characteristics of the reading system can be realized using a digital circuit configuration. , it is less susceptible to variations in parts, etc., and has the excellent effect of improving stability and reproducibility.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a circuit for realizing an image signal binarization method according to an embodiment of the present invention, FIG. 2 is a diagram showing the level of quantization of an image signal in the circuit, and FIG. It is a timing chart in a circuit. 1... A/D conversion circuit, 2, 3... U-nochi times g, 4... Digital controller (rake, 5... Launch circuit, 6...-judgment circuit, 7... line memory.

Claims (3)

[Claims] (1) The image signal obtained by main-scanning and sub-scanning the original is quantized to L (L≧3) level, and the level of the target pixel itself in this quantized image signal, the level of the target pixel, and An image signal binarization method characterized in that black and white is determined for the target pixel based on the magnitude relationship between the levels of M (M≧1) pixels before and after the target pixel. (2) The black and white determination for the target pixel is performed by also considering the black and white determination result for the same pixel as the target pixel in the line immediately before the line to which the target pixel belongs. Image signal binarization method. (3) The second aspect of the present invention is to determine whether the target pixel is black or white by also considering the level of the quantized image signal of the same pixel as the target pixel in the line next to the line to which the target pixel belongs. Image signal binarization method described in Section 2.