JPS6080368A - Binary coding device of picture signal - Google Patents

Abstract

PURPOSE:To binary-code properly a picture signal for a small or large density by providing upper/lower limits to a threshold value so as to set the range varying the threshold value. CONSTITUTION:An average value operating section 1 calculates an average the value Ma of the density at the surrounding of a noted picture element desired to be binary-coded based on density signal of a picture element applied from a scanner. The average value Ma is fed to a threshold value operating section 2 and the threshold value TH is corrected at a threshold value correcting section 3. The maximum value Lh and the minimum value Ll of the threshold value are preset to the threshold value correcting section 3 so as to output a corrected threshold value THm whose threshold value TH is corrected to obtain the range. A signal limited to the maximum value Lh is obtained from the corrected threshold value THm when the threshold value TH is larger than the maximum value Lh and a signal limited to the minimum value Ll is obtained when the TH is smaller than the minimum value Ll. The corrected threshold value THm is fed to a binary-coding circuit 4, where the density signal Px is binary-coded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an image signal binarization device that properly binarizes both signals even if the density of the background of a document changes.

[Prior art] In order to decompose an image on a document whose background color or density has partially changed, such as a pasted document, a mixed-picture document, or a color document, into pixels and convert it into a binary image, it is necessary to perform binarization. It is necessary to change the threshold value according to the background color and density of the document.

Conventionally, the density level signal of each pixel is divided at a predetermined ratio using a dividing resistor, the divided signal is low-pass filtered to form a so-called floating threshold, and the density level signal is binarized using this floating threshold. A device that does this is in practical use.

However, in such a conventional device, since a time constant element is used in the low-pass filtering means, it is not possible to sufficiently follow density changes in areas where the background density changes rapidly, such as at the boundary between the bonded portions of originals. Furthermore, the floating threshold becomes too small especially in areas with low density, resulting in the inconvenience that a pixel that is clearly a white pixel is determined to be a black pixel.

[Purpose] The present invention has been made to eliminate the above-mentioned drawbacks of the prior art, and by setting a lower limit of 1 for the threshold value, it is possible to prevent binarization errors in areas where the density is very high and areas where the density is very low. An object of the present invention is to provide an image signal binarization device that can prevent the above problems.

[composition] FIG. 1 is a block diagram showing one embodiment of the present invention.

In this embodiment, a threshold value that takes into account the average value of the density of white pixels is used.

In the figure, an average value calculation unit 1 calculates the average value Ma of the density around the pixel to be binarized (hereinafter referred to as the pixel of interest) based on the density signal (both signals) Px for each pixel added from the scanner. Calculated using equation (1).

Ma=(+-nt)So 4-mMa' --=(1)
Where, So is the density of the pixel of interest, n is the weighting coefficient, and Ma'
is the average value when the immediately previous pixel is the pixel of interest.

This equation (1) will be explained below.

Generally, there is a correlation between the density of a pixel of interest and the density of a pixel located near the pixel of interest (hereinafter referred to as a reference pixel). It is strongly related to pixels and has a strong correlation with density.

Therefore, the weighted average value calculated by assigning a weighting coefficient that decreases exponentially as the distance from the pixel of interest increases to each reference pixel according to its distance is:
This reflects the density distribution of the image near the pixel of interest, that is, the background density.

Therefore, by performing a predetermined calculation (for example, conversion using a linear function) on this weighted average value, a threshold value that reflects the background density can be obtained.

Now, let the density of the pixel of interest be So, and the distance ta from the pixel of interest is
The density and weighting coefficient of the reference pixel separated by I (i X r> (i is an integer, r is the width of one pixel) are Si,
m''(0<m<]), the weighted average value Ma of the reference pixels including the pixel of interest is expressed by the following equation (+1).

By transforming this equation (1), the following equation (Ill) is obtained.

Here, since Ill'' is 0, the above formula (m) +'
i It becomes as follows.

Ma = (] - m)S o + mMa' In this way, the above formula (1) was obtained.

Now, the average calculated by the average value calculation unit 1 (directly Mal) is added to the threshold value calculation unit 2.

The threshold calculation section 2 calculates the threshold value TH using the linear equation (IV), and adds this to the threshold correction section 3.

TI=Ia+k2--(IV)where l(1, 2 are constants).

The threshold correction value 3 includes the maximum threshold value Lh and the maximum value L (l is set in advance, and the corrected threshold T) 1m is set in advance to be within this range. .

That is, the threshold Tl+ is between the maximum value 1, h and the minimum value LQ; in some cases, the threshold Tl+ is set to a value LN thicker than the maximum value Lh.
When the maximum value I, 11 is smaller than the minimum value 1, Q), JX (direct LQ) is output.

Therefore, the concentration as shown in FIG. 2 (if Px is added, the corrected threshold T) is increased by 1 ml, the threshold Tl+ is polished, trxr 1. The part larger than h is set to Lh,
Minimum [portion smaller than LQ is 1. The signal is limited to Q.

This modified threshold THm is applied to the binarization circuit 4.

The binarization circuit 4 binarizes the density signal Px using a modified threshold value THm to form a binarized signal SB, which is sent to the next stage device (
(not shown).

In addition, in the two consecutive embodiments, the maximum value [, h (i.e., upper limit) and the minimum value Lll
! (that is, the lower limit) is fixed, but it may be changed stepwise depending on the value of the average value Ma.

[Effects] As explained above, according to the present invention, the possible range of the threshold value is set, so there is an advantage that the image signal can be properly binarized in areas where the density is high or where the density is low.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG. 2 is a waveform diagram illustrating a density signal, a threshold value, and a modified threshold value. 1. Average value calculation section, 2.. Threshold calculation section, 3. Threshold correction section, 4.. Binarization circuit. (-, Yono

Claims (1)

[Claims] (1) An average value calculation means for calculating the average value of the density of a pixel of interest to be binarized and a plurality of reference pixels consecutive to the pixel of interest, and a predetermined calculation method applied to the calculation result of the average value calculation means. 1. An image signal binarization device comprising: a threshold value calculation means for calculating a threshold value; and a threshold value correction means for correcting the threshold value calculated by the threshold value calculation means to within a preset range. (2. In claim 1, the average value calculation means calculates a weighted average value by assigning to each reference pixel a weighting coefficient that decreases exponentially according to the distance from the pixel of interest. An image signal binarization device characterized by: