JPS6340467A - Picture processor - Google Patents

Abstract

PURPOSE:To discriminate the three kinds of regions, i.e., the character, the photograph and the print of an original by comparing the each variable quantity of the maximum value and the minimum value of both signals converted from an original picture variabile density signal into a defocus signal, with a prescribed threshold. CONSTITUTION:In a region discriminating device 10, after a variabile density data is stored in a buffer memory 11, the defocus signal is calculated in a defocus signal calculation circuit 12, and is stored in the buffer memory 13. On the other hand, the density data is stored directly in the buffer memory 14, and in a region decision circuit 14, the maximum value max1 and the minimum value mini of the window of 7 X 7 of the original picture density signal are extracted from the buffer memory 14. similarly, the maximum value max2 and the minimum value min2 of the window of 7 X 7 of the defocus signal are extracted from the buffer memory 13. A selection circuit 30 selects pathes (a), (b) or (c) by a region deciding signal obtained from the region decision circuit 15 of the region discriminating device 10.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a document reading device used in a digital copying machine, a facsimile machine, etc. The present invention relates to an image processing method that identifies double-headed areas after reading a document containing a mixture of areas, and performs appropriate processing on each area.

[Conventional technology and its problems]

Many common documents contain a mixture of binary images such as characters and halftone images such as photographs and prints.

If such a document is simply converted into a binary value and recorded, the quality of the text is good, but the quality of the photograph deteriorates.On the other hand, if the document is converted into a binary value and recorded using halftone generation methods such as dithering, the quality of the photograph is good. However, the quality of the characters deteriorates. In order to obtain good quality for each of text and photographs, it is necessary to identify two different types of areas, and for example, it is necessary to perform simple binarization for characters and dither processing for photographs.

As a method for identifying the binary image area and the halftone image area, there is a method that uses the absolute value of the difference between the focus value and the defocus value of a pixel, as shown in Japanese Patent Laid-Open No. 58-220563. be. This is based on the fact that there are many sharp edges in character areas.Also, as shown in Japanese Patent Application Laid-Open No. 58-3374, the maximum and minimum values within a block are Some make judgments based on the difference between

However, since both methods detect only the edges, the identification status of the intermediate part between the edge and edge is poor, and the performance is poor when it comes to hollow characters, printing with halftone dots, etc. The disadvantage was that it worsened the situation.

The present invention was devised to solve the above-mentioned drawbacks, and its purpose is to simultaneously identify three areas with different halftone dots, such as text, photographs, and printing, with high accuracy.

Means for Solving Problem C] In order to achieve the above object, the image processing apparatus of the present invention has the following features:
In a document reading device that reads and records a document in which a value image area and a halftone image area are mixed, a means for generating a defocus signal from a read original image density signal, and a means for generating a defocus signal from a read original image density signal and a predetermined size of the original image density signal and the defocus signal are provided. scan with a window, find the maximum value and minimum value of density for both signals, compare the difference between each maximum value and the difference between each minimum value with a predetermined threshold, and identify each region based on the comparison result. The invention is characterized in that it is provided with a means to do so.

The area identifying means identifies the area as a print area when the difference between the maximum values is equal to or greater than a predetermined threshold and the difference between the respective minimum values is equal to or greater than a predetermined threshold; is greater than or equal to a predetermined threshold and the difference between the minimum values is less than a predetermined threshold, it is identified as a character area, and the difference between the maximum values is less than a predetermined threshold and the difference between the minimum values is a predetermined value. When it is less than the threshold, each area can be identified by identifying it as a photographic area.

[Effect]

The present invention focuses on density changes when a defocus signal is generated from an original grayscale image obtained by reading a document, and the defocus signal and the original grayscale signal are observed and compared through an NXM window. That is, attention is paid to the fact that changes in the maximum value maX% and the minimum value min of each signal are different in three types of areas, such as text, photographs, and halftone dots of printing.

FIG. 1 shows a block diagram of an image processing apparatus according to the present invention.

Reference numeral 1 in the figure is an input device having, for example, a drum-type scanner configuration, which inputs document information at a resolution of 500 dots/inch.
Read as bit density data.

2 is an 8-bit image memory, which stores the scanned original.
For example, it has a capacity to store data for one page of F4 size. 3 is an image processing device that identifies the area of the scanned document and performs delay processing and processing for halftone images;
Simple binarization is performed on the character image to generate binary data of 0.1 without sending it to the output device 5. 4 is a control device that performs signal control of the entire device. Output device 5
The printer records a binary image at a resolution of 500 dots/inch, and although an electrophotographic laser beam printer is taken as an example here, it is a thermal printer.

It is also possible to use an inkjet printer or the like.

The present invention relates to a method for area identification in the image processing device 3.

The present invention will be explained in detail using FIGS. 2 to 7.

FIG. 2 shows a comparison between the original image density signals of three different areas, text, photograph, and printing, and the defocus signal obtained using the defocus aperture shown in FIG. 6. The defocus aperture used here was 50 μm×
50 μm corresponds to one pixel, and it is composed of a 5×5 window. The defocus signal value DF of the pixel of interest is determined by the following equation.

However, d(i,j) indicates the gray scale value of the pixel at address i, j.The frequency characteristic at this time is about 4 lines/pixel as shown in Figure 7.
wia is the cutoff frequency, which is set so that halftone dot components of 135-line printing are removed. Figure 2 (
As shown in 3a) and 3b, it can be seen that the periodic unevenness of the halftone dots that existed in the original image density signal is completely removed. That is, as shown in FIG. 3B, the center value of the unevenness of the original image gradation signal, that is, a value close to the DC value, is obtained by defocus processing.

Now, for each of the original image density signals shown in FIG. 2 (1a) to (3a) and the defocus signals shown in FIG. 2 (1b) to (3b) over a certain length, the maximum value ma
Let's take x and the minimum value min. Furthermore, each maximum value max
The difference between dma! And the difference d +win between each minimum value min is also determined.

Regarding the character area, as shown in Figure 3 (la) and (lb), the deterioration of the top part of the character with high density is large, so the difference dIIIM between the maximum values max shows a large value, but the difference in the character area Most of the area is composed of, for example, a background area with a constant density of white or colored areas, so there is little change in the foreground area, and the difference dmi between the respective minimum values min.

indicates a small value. In addition, Fig. 3 (Ia), (lb
) indicates the case of 6 point characters.

In addition, in the case of photographs, as shown in Figure 3 (2a) and (2b), since there are no high frequency components in most cases, the amount of change dlIa in the maximum value max and minimum value min
The value of x+d mtn is small.

In the case of printing, as shown in FIGS. 3(3a) and 3(3b), it can be seen that the wide range of the maximum value may and the minimum value min is compressed by obtaining a defocus signal. That is, the difference d 1laX between the maximum values max and the difference d wi between the minimum values min in FIGS. 3A and 3B both show large values. In addition, Fig. 3 (3a
) and (3b) show the case of printing with 135 halftone dots.

3 to 5 are histograms of various parts taken to show similar effects.

In the character part of Fig. 3, the amount of change d IIIN of each maximum value maxl and max2 is maxi-max2 = 172 11
8=54, which is a large value. In addition, the minimum 4 fimi
The amount of change in n d mi, l is m1n2 m1nl-2-
0-2, which is a small value.

In the photographic part of Fig. 4, the maximum value ll1ax and the minimum value min
The amount of change in both is small. That is, dmax = 129
-125 = 4, dll, 1l = 61-56 =
It is 5.

In the printing section of FIG. 5, the amount of change is large for both the maximum value max and the minimum value min. That is, d, , =229-1
87 = 42. dll,,1=67-30-37.

Therefore, if we find the maximum value + 118X % change in minimum value min 1d-ax, dmt, we can calculate text, photo,
It becomes possible to identify three areas of printing.

For this purpose, in the present invention, a window of nxm is used to calculate the maximum value and minimum value maxl, m1nl of the original image gray signal and the maximum value and minimum value max2 . m1n2 is extracted, and each change amount d l+aX + d@fi11 is determined to identify the area.

Each change is identified when the states of 1d, aX, d, and □ meet the following conditions. Furthermore, th□. , thll,
n is a threshold value for determination.

That is, ■ If d□8≧th, , and d , Iin≧t
If h min, it is the printing area; ■ If dIIIIX≧thIfillX and d1□<th□.

If so, it is a character area; ■ If dmaw < thffill, and d□
, <th□1, it is determined that the area is a photo area.

〔Example〕

An example for realizing this determination will be described below.

FIG. 8 shows the configuration of an embodiment of an image processing apparatus according to the present invention.

In this embodiment, the image processing device includes an area identification device 1
0, an image forming device 20 that performs appropriate image processing on each image region and generates binarized data, and a region identifying device 1.
The image forming apparatus 20 includes a selection circuit 30 that switches the output of the image forming apparatus 20 based on the area determination signal from 0. The operations are also performed on nxm, eg 7x7 windows.

In the area identification device 10, first, the gray scale data is divided into, for example, 5
After storing the signal in the Batza memory 11 having a capacity for the line, a defocus signal is calculated in the defocus signal calculation circuit 12. Note that this defocus signal is the 6th defocus signal.
It is formed based on the defocus aperture shown in the figure.

Then, ζ is stored in the buffer memo January 3 for m947 minutes (7 lines in this embodiment).

On the other hand, is it directly in the buffer memory 14? Agriculture? The flame data is m
It is stored for 547 minutes. Then, the area determination circuit 15 extracts the maximum value maxl and minimum value n+inl of the 7×7 window of the original image gray level signal from the buffer memory 14, and similarly stores the buffer memo! The maximum value max2 and minimum value m1n2 of the 7×7 window of the defocus signal are extracted from month 3. Next, d max = maxl max2+ d mi
An i1 calculation of n=min2-minl is performed, and an area determination signal consisting of 2 bits is sent to the selection circuit 30 based on the determination conditions of ■■■ shown above.

In the image forming apparatus 20, assuming that the image has three regions, the gray level signal is branched into three paths, a, b, and C, so that optimal image processing is performed for each region in advance.

Path a is processed assuming that the gray level signal is a character, and the MTF (modu-Ia) degraded by the input system is used.
tion transfer function), and a simple binarized signal is generated by a binarization circuit 22 having a binarization threshold shown in FIG.

The flow is based on the assumption that the grayscale signal is a photographic signal, and after the MTF correction circuit 21, a dither processing circuit 23 having a dither threshold shown in FIG. 10 performs dither processing to generate a halftone image signal. It is something to do.

Path C is a flow that assumes that the grayscale signal is a print signal, and is passed through a halftone removal circuit 24 such as a low-pass filter circuit to remove halftone dots that become moiré during dither processing, and then passes through an MTF correction circuit. Through R21, the dither processing circuit 23
This generates a halftone image signal.

In the selection circuit 30, the area determination circuit 15 of the area identification device 10
By selecting path a, b, or C based on the area determination signal obtained from the above and sending only one signal to the recording output device, it is possible to record an image that has undergone optimal processing for each area.

In the present embodiment, the percha size for obtaining the defocus signal was 5×5, and was determined by a simple average of the weighting coefficient 1, but the size and shape 2 weighting coefficients may be arbitrary. Further, although the frequency characteristic was to remove 135 VA, °ζ is also effective for different halftone dots such as 100 lines and 175 lines even if used as is.

Furthermore, the size of the window for determining the maximum value and minimum value may be arbitrarily varied depending on the resolution of the reading system and the object. As another example, valid results can be obtained with n-20 and m=1.

〔Effect of the invention〕

As described above, in the present invention, the amount of change in the maximum value and minimum value of both signals when the original image density signal is converted into a defocus signal is determined by the amount of change in the maximum value and minimum value of both signals.

Paying attention to each change in the three areas of photography and printing, the amount of change is compared with a predetermined threshold.

This makes it possible to identify the three types of regions, and perform optimal processing depending on the type of image.In other words, the present invention detects only steep changes at the edges of images, unlike conventional methods. Since it does not perform area identification based on the image, the identification accuracy is high, and image processing can be performed without problems such as hollow characters or poor performance for images printed with halftone dots. .

[Brief explanation of drawings]

FIG. 1 is a schematic block diagram of an image processing device according to the present invention;
Figure 2 is a graph comparing the original image density signal and defocus signal, Figure 3 is a graph showing changes in the histogram in the text area, Figure 4 is a graph showing changes in the histogram in the photograph area, and Figure 5 is the graph in the printing area. 6 is an explanatory diagram of the defocus aperture, FIG. 7 is a graph illustrating the frequency characteristics of the defocus aperture, and FIG. 8 is a block diagram showing an embodiment of the image processing device according to the present invention. 9 are explanatory diagrams showing threshold values for binarization, and FIG. 10 is an explanatory diagram showing threshold values for dither processing. 1: Input device 2: Image memory 3: Image processing device 4: Control device 5: Output device ICl3.14 F buffer memory 10: Area identification device 12: Defocus signal calculation circuit 15; Area determination circuit 20: Image forming device 21 :MTF correction circuit 22:
Binarization circuit 23: Dither processing circuit 30: Selection circuit

Claims (1)

[Scope of Claims] A document reading device that reads and records a document in which a monochrome image area and a halftone image area are mixed, comprising means for generating a defocus signal from a read original image density signal, and the original image density signal. and the defocus signal are scanned with a window of a predetermined size, the maximum value and the minimum value of the density are determined for both the signals, and the difference between each maximum value and the difference between each minimum value is compared with a predetermined threshold value. An image processing apparatus comprising: means for identifying each area based on a comparison result. 2. The means for identifying the area identifies the area as a printing area when the difference between the maximum values is equal to or greater than a predetermined threshold and the difference between the respective minimum values is equal to or greater than a predetermined threshold; When the difference is greater than or equal to a predetermined threshold and the difference between the minimum values is less than a predetermined threshold, it is identified as a character area; 2. The image processing apparatus according to claim 1, wherein the image processing apparatus identifies the area as a photographic area when the area is less than a predetermined threshold value.