JPH02177668A - Picture processor - Google Patents

Abstract

PURPOSE:To obtain an output picture having high picture quality by securing the maximum dynamic range even when the density of an input picture is biased to either a light side or a shade side. CONSTITUTION:The first input means inputs multilevel picture data, a preparing means prepares a density distribution based on the input multilevel picture data, a setting means sets the threshold level of a density level based on this density distribution, and the second input means reinputs the multilevel picture data. A binarizing means binarizes the input multilevel picture data by a binarization processing method to execute binarization while retaining the error between input picture density and output picture density based on the threshold level of the density level. Thus, the deterioration of picture quality caused by an omitted picture element after the binarization can be prevented, and the output picture having the high picture quality can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image processing apparatus, and, for example, to an image forming apparatus that outputs multivalued (for example, 8-bit) image data as binary data.

[Prior Art] Conventionally, an error diffusion method has been devised as a method of binarizing multivalued data as a means of obtaining a high-resolution image.

This involves binarizing a certain pixel based on a certain threshold, that is, the slice level, and comparing the data before binarization (“O” to “255” in the case of 8-bit data) and the data after binarization ( 8
In the case of bit data, an operation of adding an error of "o" or "255" to surrounding pixels is performed sequentially over all pixels.

[Problems to be Solved by the Invention] However, in the conventional example described above, the density distribution of actual image data may be biased toward the darker or lighter side. This will be explained using the histogram shown in FIG.

For example, in the error diffusion method, if the slice level is set to the center of the density (“128” in the case of 8 bits), as shown in Figure 4(a), for light image data, the dots will be The dot will appear later than the position where the dot exists. As a result, it was not possible to reproduce an image faithful to the original, and the image quality deteriorated. In addition, as shown in Figure 4 (C), for dark image data, "25
It is often output as 5°'. This increases the number of missing pixels in the original document, which causes deterioration in image quality. Alternatively, as shown in FIG. 4(b), the highest quality output can be obtained for medium density image data. In this way, in the conventional error diffusion method, since the slice level is constant, high image quality can be obtained for data with many pixels at the density near the slice level, but the image quality is lower for data with uneven density. The disadvantage was that it deteriorated.

The present invention has been made in view of the above-mentioned drawbacks of the conventional example, and its purpose is to provide an image processing device that prevents deterioration in image quality due to missing pixels after binarization.

[Means for Solving the Problems] In order to solve the above-mentioned problems and achieve the purpose, an image processing device according to the present invention processes multivalued image data while preserving the error between the input image density and the output image density. Perform binarization 2
An image processing device that converts into binary image data using a value processing method includes a first input means for manually inputting multivalued image data, and a density distribution based on the multivalued image data inputted by the first input means. a creating means for creating, a setting means for setting a density level threshold based on the density distribution, a second input means for re-inputting the multi-value image data, and a multi-value input by the second input means. The apparatus is characterized by comprising a binarization means for binarizing the image data based on the threshold value of the density level. [According to the configuration described above, the first input means manually inputs multivalued image data, the creation means creates a density distribution based on this manual multivalued image data, and the setting means creates a density distribution based on this density distribution. to set the concentration level threshold, and
The manual means re-inputs this multivalued image data, and the binarization means converts this input multivalued image data into two values based on the density level threshold while preserving the error between the input image density and the output image density. Binarization is performed using a binarization processing method that performs digitization.

[Embodiments] Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram showing the configuration of an image processing apparatus according to this embodiment. In the figure, l represents an exposure lamp that irradiates light onto the original, and 2 represents a CCD image sensor that reads light reflected from the original by exposure from the exposure lamp l.

3 indicates an amplifier A/D converter that amplifies and quantizes the output of the CCD image sensor 2, and 4 indicates a shading correction circuit that corrects the output data from the amplifier A/D converter 3 based on the characteristics of the CCD image sensor 2. It shows. Reference numeral 5 indicates an LOG conversion/gamma correction circuit that corrects the gradation characteristics of the COD image scanner 2, and reference numeral 6 indicates an error diffusion processing circuit that diffuses errors to pixels in the vicinity of the pixel of interest. Reference numeral 7 indicates a printing unit that performs printing using a laser beam method or an inkjet method, and 8 indicates a CCD driver that drives the CCD image sensor 2. 9 is a central processing unit 9a (hereinafter referred to as rCPUJ), ROM
9b and RAM 9c, and shows a control section that controls connected circuits and performs data calculations. Reference numeral 10 indicates a lamp driver for driving the original exposure lamp 1. Reference numeral 11 indicates an operation provided with keys for issuing various instructions such as printing a document in this apparatus.

Further, the above-mentioned control unit 9 determines the threshold value used in the error diffusion processing circuit 6, that is, the slice level.

Therefore, FIG. 2 is a circuit diagram schematically showing the configuration of the main parts within the control section 9. As shown in FIG. In the figure, numeral 12 indicates a histogram creation circuit that divides the density level into eight levels and counts pixels corresponding to each level. 100-1
06 indicates a latch that stores a comparison value for determining the density level of human image data, and latches 100 to 106
The comparison value is set by the CPU 9a. In this embodiment, in order to divide the density level "0°°~"256" into 8 levels, the latch 100 has "32", the latch 101 has "64", the latch 102 has "96", and so on.
6 and "224" are respectively set. And 200~
206 indicates a comparator for comparing the input image data and each comparison value of the latches 100 to 106, respectively;
Each comparator outputs a count signal to be output to subsequent counters 400 to 406 when the value of each latch is exceeded. In FIG. 2, A indicates an input terminal for input image data, and B indicates an input terminal for comparison values output from each latch. Further, each comparator outputs the above-mentioned count signal only when A>B.

Here, reference numerals 300 to 306 indicate AND circuits that AND circuits image data and a pixel synchronization clock (VCK) signal for synchronizing the image data and applying it to the control unit. 400-406 indicate counters that increment the count values of the comparators 2°O-206 that have passed through the AND circuits 300-306. and 500 is C
It shows a selector that switches the counters 400 to 406 according to instructions from the PU 9a.

Here, the operation of the histogram creation circuit 12 will be briefly described.

This histogram creation circuit 12 divides the density level into eight levels and counts pixels corresponding to each level. The CPU 9a has each latch 1 in advance.
The density level is written in 00 to 106. First, when the operation starts from the leading edge of the document, image data is generated in synchronization with the VCK signal, and each latch 100 to 10
The comparison value set to 6 is for each comparator 200 to 20.
At step 6, the input image data is compared with the input image data, and if the image data is larger, a count signal is sent to the subsequent counters 400 to 406, and the count is incremented. After sampling the necessary concentration data in this way, the CPU 9a switches the selector 500,
Each counter value is read in sequence, and a histogram is created based on the number of pixels at each density level.

Next, the image processing method of this embodiment will be explained.

FIG. 3 is a flowchart illustrating the operation of the CPU 9a of this embodiment.

First, upon receiving an instruction to start copying from the operation unit 11, the document is bliscanned. In this case, DC (not shown)
The exposure lamp 1 is turned on according to instructions from the controller, and the COD image sensor 2 captures the reflected light from the original (
Step S1, Step S2). This first scan is hereinafter referred to as a briscan. The image data captured by the COD image sensor 2 is processed by an amplifier A/D converter 3.
Amplification and conversion from an analog signal to a digital signal (quantized to 8 bits) are performed in the shading correction circuit 4 and LOG conversion/gamma correction circuit 5 at the subsequent stage.
and undergoes shooding correction, LOG conversion, and gamma correction (step S3). The image data processed in this way is sampled for each pixel via the histogram creation circuit 12 in the control unit 9, and when the sampling for one page is completed, the histogram is The slice level is set based on the creation and histogram. In setting this slice level, the density level at which the number of pixels in the histogram is the maximum value is set as the optimum slice level for binarization. The slice level set in step S7 is then set in the error diffusion circuit 6 as level data (step S
6. Step S7).

When the slice level is set in this way, the scanning process starts again from the leading edge of the document (step S8).
), the output signal of the COD image sensor 2 is amplified, and the A/
Based on the error diffusion method, which is a binarization processing method that performs D conversion, shooding correction, LOG conversion, and gamma correction, and also performs binarization while preserving the error between the known input image density and output image density. Binarization processing is performed in the error diffusion processing circuit 6 (Step S9. Step 5IO). Then, the error diffusion processing circuit 6 sends the 62-valued data to the printing unit 7 (step 5ll), and prints one page, that is,
Steps 88 to 11 are performed until binarized data is generated.
The processes up to this point are repeated (step 512). In this way, the slice level can be set based on the density distribution based on the input image data.

FIG. 5 is a block diagram showing an example of the error diffusion processing circuit 6 of FIG. 1.

In FIG. 5, 41 is an adder, 42 is a binarization circuit, 4
Reference numeral 3 indicates an arithmetic unit, 44 an error buffer, and 45 a weighting circuit. First, the multivalued data sent from the LOG conversion/gamma correction circuit 5 is added by the adder 41 to the error data generated when the previous pixels were binarized. Note that error data for multiple pixels is stored in the error buffer 44, and weighting is performed by the circuit 4.
Weighting is performed by 5. The multivalued data to which the error data has been added is binarized by the binarization circuit 42 at the slice level given from the control unit 9. In the binarization circuit 42, 2
Based on the digitization result, binary data of "0" or "255°" is output.The arithmetic unit 43 calculates the error of the multi-value data obtained by adding this binary data and the error data before binarization. and stores it in the error buffer 44. On the other hand, the printing unit 7 forms an image by turning dots on and off in accordance with the binary data sent from the binarization circuit 42.

As a result, binarization processing is performed using the error diffusion method with variable slice levels.

As described above, according to this embodiment, even if the density of the input image is biased toward either the dark or dark side, a high-quality output image can be obtained by ensuring the maximum dynamic range. Furthermore, even if the slice level is changed, the error between the input image density and the output image density is preserved, so it is possible to reproduce an image faithful to the original.

Now, when density measurement is performed during blisscanning, it is desirable to detect a certain range of the document or to perform a specific range after specifying.This can avoid a reduction in the dynamic range due to density measurement of unnecessary areas. .

Furthermore, in the above description, sampling of the measurement area is performed over all pixels at once, but the present invention is not limited to this. For example, sampling may be performed every few mm to increase the number of bits of the counter. It is possible to reduce the circuit scale.

[Effects of the Invention] As explained above, according to the present invention, even if the density of the input image is biased toward either the dark or dark side, a high-quality output image can be obtained by ensuring the maximum dynamic range. .

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of the image processing device of this embodiment, FIG. 2 is a circuit diagram schematically showing the configuration of the main parts in the control unit 9, and FIG. 3 is the operation of the CPU 9a of this embodiment. 4A, 4B, and 4C are histogram diagrams illustrating different density distributions. FIG. 5 is a block diagram illustrating an example of an error diffusion processing circuit. In the figure, l: Exposure lamp, 2: COD image sensor, 3: Amplifier/AD converter, 4: Schuding correction circuit, 5: LOG conversion/gamma correction circuit, 6... error diffusion processing circuit, 7... printing section, 8.
...COD driver, 9...control unit, 9a...CP
U, 9b”-ROM, 9c...RAM% 10...
Lamp driver, ll...operation unit, 41...adder, 42...binarization circuit, 43...arithmetic unit, 44...
- Error buffer, 45... Weighting is a circuit. Figure 4

Claims (1)

[Scope of Claim] An image processing device that converts multivalued image data into binary image data using a binarization processing method that performs binarization while preserving an error between an input image density and an output image density, comprising: a first input means for inputting image data; a creation means for creating a density distribution based on the multivalued image data input by the first input means; and a creation means for setting a density level threshold based on the density distribution. a setting means; a second input means for re-inputting the multi-value image data;
An image processing device comprising: binarization means for converting into values.