JPH01284173A - Method and device for processing picture - Google Patents

Abstract

PURPOSE:To reproduce the satisfactory picture of high resolution with simple constitution by outputting picture data, which has been quantized by means of an error diffusion method, through the use of a thermal head printer. CONSTITUTION:The thermal head printer 7 records binarization data which has been binarized by the error diffusion method in a read part 1. In the case of the thermal head printer, the diameter of printing dots become large in a dark part where there are many black dots in a periphery picture element, and the diameter of the printing dots become small in a highlight part where there are less black dots. Thus, the occurrence of a stripe pattern which is peculiar to the error diffusion method can be prevented in the highlight part, and a void between the dots can be prevented in the dark part, whereby the satisfactory picture can be reproduced.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an image processing method and apparatus for quantizing image data, and more particularly to an image processing method and apparatus for pseudo-reproducing halftone images. It is something.

[Conventional technology]

2. Description of the Related Art Conventionally, there has been an error diffusion method as an image processing method for reproducing a halftone image as a binary image in digital copying, digital facsimile, etc.

This method calculates the pixel-by-pixel density difference between the original image density and the binarized output image density, that is, the error, and uses the error value that is the result of this calculation to identify the surrounding pixels of the pixel of interest according to the coefficients of the weight matrix. This is a method in which the weights are applied and then distributed.

Since this method spatially eliminates the error, which is the difference in density between the original image and the output image, there is no restriction on the number of gradations due to the matrix size, unlike dither processing, which is another binarization method. Value-dependent threshold processing can be performed.

Therefore, the error diffusion method makes it possible to achieve both gradation and resolution, which are problems in dither processing.

Regarding this error diffusion method, see the literature R, W F 1 o
y dand L, Steinberg “An A
adaptive algorithm for 5pat
ial Gray 5cale” 5ID75 Di
gest (1976).

The error diffusion method can be described as follows. However, the input data is assumed to be 6 bits.

D! , l = X IJ + (ΣΣαI+m, l+
n @ E l+m, l+n) (1/Σαm, n
)YLI=63 (DI, 12T) YtJ=0 (DI, )<T) where DI,l: Density X+ after correction of pixel of interest (i, j), input image of pixel of interest (i, j) Concentration E+, 1:
Error α-, weighting coefficient Yt, + when binarizing the pixel of interest (i, j): Output image density T Two thresholds In other words, in the above equation, the input image density of the pixel of interest X+, j is
Weighting error E generated in surrounding pixels 1+m, I+□ (multiply by αI+m, l+n and divide by Σα□, n)
The calculated values are added, and the resulting value becomes the error-corrected density DLJ of the pixel of interest. Then, set the D +, I to the threshold T (
For example, the output image density YLI is obtained by binarizing at T=32).

The printer performs dot on/off control according to the Yt and H values to form an image.

[Problem that the invention is trying to solve] However,
The error diffusion method has the disadvantage that a unique texture (striped pattern) appears in the highlights and halftones of the image. This occurs because the dots of the binary output are connected in a linear manner.

Let's consider why this texture occurs.

As described above, the error diffusion method weights the error generated in the pixel of interest using a weight matrix and disperses it to surrounding pixels.

For example, this weight matrix αLJ(X, 1), that is,
Consider a case where an error occurring in the pixel of interest X is distributed to the pixel on the right.

Positive errors often occur in the highlight and halftone areas of an image because the probability that the output image will be O is higher than in the dark areas. This is because the input image data has at least a certain degree of density, so if the output image is set to 0, a pressure error will occur.

This positive error is expressed as the weight matrix αLJ(X, 1)
If the dot is dispersed to the pixel on the right side, there is a high possibility that the dot will be turned on in the dispersed pixel. When the processing of one line of input image data is completed and the next line is moved on, the positive error is dispersed to the pixel corresponding to the previous line (the pixel below the previous line), and the dot of this pixel is turned on. It is more likely that

In other words, this increases the possibility that dots will turn on periodically in the sub-scanning direction, and a striped pattern will occur due to the connection of these dots. FIG. 14 shows how the striped pattern occurs in the sub-scanning direction.

Furthermore, depending on the shape of the weight matrix, dots are connected in the main scanning direction or diagonally, resulting in a striped pattern.

In this way, while the conventional error diffusion method has better resolution than dither processing, it produces a unique texture (striped pattern) in the highlights and halftones of the image, making it difficult to reproduce good images. There wasn't.

[Means and operations for solving the problems] The present invention eliminates the above-mentioned drawbacks of the prior art, and provides an image processing method and apparatus that can satisfactorily reproduce any input image.

That is, according to the present invention, by providing a quantization means for quantizing input image data using an error diffusion method, and an output means for outputting the image data quantized by the quantization means to a thermal head printer, the stripes can be reduced. This suppresses the occurrence of patterns and reproduces good images.

Further, according to the present invention, in an image processing method in which image information is communicated from one device to another device, the one device converts image information obtained by reading a document into quantum data using an error diffusion method. and transmitting it to the other device,
The other device is capable of communicating good images by recording the image information received from the one device and quantized by the error diffusion method using a thermal head printer.

〔Example〕

Hereinafter, one embodiment of the present invention will be described in detail with reference to the drawings.

First, as a first example, a case will be described in which a laser beam printer (LBP) is used as the recording section.

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

1 in the figure is a reading unit, and an image input unit 101 reads the original image.
The image processing unit 102 quantizes manually generated image data into binary data using an error diffusion method. 1
A random access memory (RAM) 03 is used for image processing and as a work area for the CPU 104. 1
04 is a CPU that controls the reading section 1 and the LBP recording section to be described later.

105 is a system bus for transporting control signals and image data in each block; 106 is a read-only memory (RO) that stores control programs executed by the CPU;
M), 107 is an LBP control circuit that controls the LBP under the control of the CPU 104.

Reference numeral 2 denotes an LBP recording unit that forms an image based on data binarized by the error diffusion method. The LBP recording unit 2 includes a laser control circuit 108 that controls the optical power of the laser beam by controlling the drive current, and a laser control circuit l.
A laser drive circuit 109 that drives a laser with a drive current controlled by O8, a laser light source 110 composed of a semiconductor laser, a rotating polygon mirror 111 that polarizes and scans a light beam on the surface of a photoreceptor 115, and a photoreceptor 115. an imaging lens 112 for imaging a light beam on a surface of 115; and a rotating polygon mirror 1.
11, a corona charger 114 for visualizing the latent image formed on the photoreceptor 115 by the laser 3, a photoreceptor 115, a recording paper 116, a transfer charger 117, and a developer 118. be done. Note that the size of the dot diameter can be controlled by controlling the drive current using the laser control circuit 108.

4 is a communication control unit that performs control when communicating binary data using the error diffusion method in the image processing unit 102, and includes an image memory, a code/decoding unit that encodes or decodes data, or a data modulation/demodulation unit. Consists of etc.

Note that the communication control unit 4 is controlled by a communication program stored in the ROM 106.

5 is a receiving device connected to the communication control unit 4 via a communication line. This receiving device 5 has an LBP recording section like the present embodiment. Reference numeral 6 denotes an operation unit for inputting information such as a communication partner.

FIG. 2 is a block diagram showing details of the reading section 1 in FIG. 1.

The input sensor unit 201 is composed of a photoelectric conversion element such as a CCD and a driving device that scans the element, and reads and scans an original.

202 is an AD converter that converts the image data read by the input sensor unit 201 into a digital signal with a quantization number of 6 bits. Here, there are 64 levels of gradation, with the lowest brightness data oooooo representing the darkest black, and the highest brightness data 111111 representing white.

The brightness data from the A/D converter 202 is sent to the correction circuit 302.
sent to. Here, the sensitivity unevenness of the CCD of the input sensor unit 201 and shading distortion, which is distortion of the light distribution characteristics of the light source, are corrected.

204 is a conversion table for converting the luminance data from the correction circuit 203 into density data;
It is composed of a ROM that outputs the density data as 6-bit density data. In general, the relationship between brightness and density is (density) = -γ
log (luminance) γ: There is a positive constant relationship, and data based on this formula is written in the table 303. FIG. 3 shows an example of the contents of the conversion table.

205 is a binarization processing unit which converts the 6-bit density data sent from the conversion table 204 into 1b by the error diffusion method.
Quantization processing is performed on the binary signal of it.

FIG. 4 is a block diagram showing details of the binarization processing section 205 of FIG. 2.

The data XLJ sent from the conversion table 204 is
The adder 401 adds the error data ELI from the adder 406 that was generated when the binarization process was already performed.

This error-corrected data D t H is expressed by the following equation.

DI, 1=X+, 1+L, + This DI, 1 is set to the threshold T (T
= 32). That is, the binarized output YLI is expressed as follows.

DLI≧T...Y+, 1 = 63D+, 1<T
...Y+, 1=0 On the other hand, Dl,j is sent to the error calculator 403. The error calculator 403 calculates the error ELI distributed to surrounding pixels based on Dtj and the binarized output Y+g. That is, E+4
can be expressed as follows.

E+, 1=D+, 1-Ytl This ELI is sent to the error distribution value calculation circuit 404, and the error distribution value calculation circuit 404 calculates the value of the error to be distributed to the four surrounding pixels of the pixel of interest.

FIG. 5 is a diagram showing a weight matrix, and this matrix shows the position and ratio of pixels to which the error EtI generated at the pixel of interest X is distributed.

In the error distribution value calculation circuit 404, AI, 1 and Bl in FIG.
, ] are determined as follows.

AI, I = 2x Int (E+, I x )
B +, j= Int (E+, I X ) However,
This error distribution value calculation circuit 404 has a configuration in which the decimal places are rounded down. In other words, only integer operations can be executed. Incidentally, Inl indicates that the fraction below the decimal point is rounded down. Then, by rounding down the decimal point, the error El, J generated in the pixel of interest and the error distribution value calculation circuit 40
AI distributed to 4 surrounding pixels calculated in 4, 1 and BLI
A remainder RI,j is generated between . This can be expressed as the following formula.

RI, 1=E+, 1-2X (AI, I+B1.J) This remainder RLi is sent to latch 405, delayed by one pixel, and added to the input data X +++, 1 of the next pixel.

On the other hand, AI,1 is allocated to pixel (i+1.D) by adder 413 and pixel (i, j+1) by adder 4
Sent to 08. Also, Btl is pixel (i+1.j+1)
Latch 407 and pixel (i-1, j+1)
is sent to adder 410 for distribution.

The memory 411 is a memory that stores errors distributed to the j+1th line, and can store error data of pixels for at least one line.

Timing generation circuit 415 is latch circuit 405.407
It generates various signals such as latch signals to 409, 412, and 414, and address signals to memory 411.

Next, the aforementioned error distribution method will be explained in more detail with reference to FIG.

Figure 6 is a diagram showing the flow of binarization processing using the error diffusion method. First, if the error generated at the pixel of interest X is weighted as PI + Ql + RI + SI, Figure 6 (a) As shown, it is distributed to four surrounding pixels. Here, Pl goes to adder 413 in FIG. 4, Q1 goes to adder 410, R1 goes to adder 408, and Sl goes to latch 40.
Sent to 7. Ql is then written to address 1 of memory 411.

Next, when the pixel of interest moves to X2, the error P2+Q2+R2+S2 is dispersed to the surrounding four pixels as shown in FIG. 6(b). Here, R2 is sent to adder 413.

Further, Q2 is added to R generated at XI by an adder 410 and written to address 2 of the memory 411.

R2 is added to S generated at xl in an adder 408. S2 is sent to latch 407.

Next, when the pixel of interest moves to x3, as shown in FIG. 6(c), the error P3+Q3. R3 and S3 are distributed to four surrounding pixels. Here, R3 is sent to adder 413.

Further, Q3 is added to sl generated in XI and R2 generated in X2 in adder 410, and is written to address 3 in memory 411. R3 is added to s2 generated in X2 by an adder 408. S3 is sent to latch 407.

When the above processing is performed for one line, the following values will be written to the memory 411.

Memory 1 address...M, =Qt Memory 2 address...M2=R, 10Q2 Memory 3rd address...M3 =S1 + R2+ Q3 Memory 4 address...M 4 =S 2 +R3+ Q 4 memory i
Address...M+ =S+-2+R+-++Q+When the processing for this one line is completed and the processing moves to the next line, the error that occurred in the previous line is read from the memory.

The error read from the memory is added to the error occurring one pixel before in an adder 413 and output from a latch 414.

Reading of the error from the memory 411 is controlled by a timing generation circuit 415 so as to correspond to the previous line. The timing generation circuit 415 detects that the pixel of interest is X+
If so, control is performed to read the address of MI-3 in the memory 411.

Binarization using the error diffusion method can be performed by performing the processing described above on all input data.

Reference numeral 416 in FIG. 4 is a comparator that determines whether the input image signals X+ and I belong to a highlight signal, dark signal, or halftone signal, and outputs a flag for each signal. .

That is, by comparing X+,4 with the two thresholds TD,, TD2, TD, < TD2), XLI≦TD + , °, Flag=0(
Highlight signal) TD 1> X+, H> TD2
, ', Flag = 1 (halftone signal) X+, 1>T
D2,', F1ag=2 (dark signal) A flag corresponding to each gradation level is output.

Next, the image data binarized by the reading unit 1 in Fig. 1 is converted into LBP.
The recording process when recording in the recording unit 2 will be explained.

Under the control of the CPU 109, the LBP control circuit 107 transfers the binary image data Data from the image processing section 102, the density flag Flag (417), and the clock signal Ck to the laser control circuit 108.

In this embodiment, the diameter of the light beam is changed depending on the contents of each flag. As this means, a method of changing the drive current is used.

FIG. 7 shows the relationship between the driving current of the semiconductor laser 10 and the optical power.

In this example, when F1ag=0 (high contrast pixel density), the drive current is I1, and the optical power is LP.
, becomes.

Further, in the case of F1ag=2 (dark pixel density), the drive current is 12 and the optical power is LP2.

Further, in the case of FIag=2 (intermediate gradation pixel density), the drive current is randomly selected as ■ or I2.

Since laser light has a Gaussian energy density distribution, as shown in FIG. 8, when the driving current is increased with respect to the optical power density distribution, the optical power density distribution becomes D2. If the exposure amount required for recording on the photoreceptor is El, then
The recording dot diameter on the photoreceptor changes depending on the drive current.

The recording dot diameter is r and the laser light output is P. Then, r=a　In　P6　bEr becomes. a and b are constants.

The driving current is converted into a light beam by a laser light source 110, and the irradiated light beam is transmitted to the photoreceptor 1 by a rotating polygon mirror 111.
15 planes are polarized and scanned. Reference numeral 112 denotes an imaging lens for focusing a light beam onto the surface of the photoreceptor 115.

After the photoreceptor 115 is charged by the corona charger 114, an image is irradiated with the light beam 3 to form an electrostatic latent image.

This electrostatic latent image is made visible by a developing device 118 and transferred onto recording paper 116 by a transfer charger 117.

Figure 9 shows data 801 that has been binarized using the error diffusion method.
8 is a diagram showing the size of a printed dot that is controlled according to the content 802 of a flag indicating the input image density.

In other words, if the binarized data 801 is 1 (dot on), F1a
When g802 is 2 (dark part), the size of the printed dot is thick (and the binarized data 801 is 1t'F1ag80).
When 2 is 0 (highlight area), the size of the printed dot is small.

Then, the binarized data 801 is 1 and F1ag 802 is 1.
(halftone area), the size of the printed dots is changed from large to small.

An example recorded using the dot size shown in Fig. 9 is shown in the first O.
As shown in the figure.

FIG. 10(a) shows a highlighted portion of the output image, and since the dot size is small, it is possible to prevent the dots from connecting. In other words, this makes it possible to prevent the unique striped pattern that occurs with the error diffusion method.

Figure 10(b) shows the halftone part of the output image. In this case, by placing small dots, it is possible to prevent the dots from connecting and eliminate the unique stripes that occur with the error diffusion method. It is possible to prevent the occurrence of patterns. Furthermore, in this case, since printing is performed by switching between two types of dot sizes, large and small, it is possible to provide gradation in the intermediate tone portion.

FIG. 10(C) shows a dark part of the output image. In this case, large dots are placed so that white spots (white dots) can be prevented from falling out between the dots.

In this way, according to the first embodiment of the present invention, by changing the size of the printed dots according to the density of the input image, the error diffusion method unique to the highlight and halftone areas of the image can be realized. It is possible to prevent the occurrence of striped patterns, and also to prevent white areas from appearing in dark areas of the image.

Further, as shown in FIG. 1, binary data and a flag indicating the density level of the image can be sent to the receiving device 5 via the communication control section 4.

When transmitting data to a partner device, the CPU 104 performs control in response to a transmission command from the operation unit 6.

The receiving device 5 is provided with a recording section having the same configuration as the LBP recording section 2, and performs recording by controlling the size of the dot diameter using the sent binary data and flag.

In the first embodiment, the density of the pixel of interest is classified into three density levels, and the print dot diameter is changed according to each density level, but the following method can also reduce the striped pattern peculiar to the error diffusion method. I can do it.

(i) Randomly change the dot diameter regardless of the concentration level.

(ii) High contrast density pixels have a dot diameter smaller than the standard dot diameter, and dot diameters are randomly varied for other density levels.

(iii) A standard dot diameter is used for dark density pixels, and the dot diameters are randomly changed for other density levels.

(iv) Only high contrast density pixels have a dot diameter smaller than the standard dot diameter.

(V) Only dark density pixels are set to standard dot diameter.

Note that the standard dot diameter can be obtained using a standard drive current.

The first embodiment described above has a configuration in which the size of printed dots in the LBP recording section is controlled in accordance with binary data and a flag indicating the image density level.

The second embodiment described below is based on the L in the first embodiment.
This configuration uses a printer that records on thermal recording paper instead of BP, and a printer using a thermal head (thermal head printer THP) such as a printer using a thermal transfer method.

FIG. 11 is a block diagram showing the second embodiment. In FIG. 11, the same reference numerals as in FIG. 1 have the same structure as in FIG. 1, and the explanation thereof will be omitted.

7 is a thermal head printer, which turns on and off driving power to a heating resistor used as a thermal head based on binary data binarized by the error diffusion method in the reading section 1; An image is formed by developing color or by transferring ink from an ink ribbon onto plain paper.

8 is a receiving device having a thermal head printer section.

In the second embodiment, the binarization processing unit 205 (
The comparator 416 of FIG. 4, which shows the details of FIG. 2), is not required.

FIG. 12 shows details of the thermal head printer 7.

IO is a thermal head control circuit that controls the thermal head under the control of the CPU 104.

11 is a shift register, which is a thermal head control circuit 10;
Converts serial data, which is binary data sent from the computer, to parallel data. A latch circuit 12 temporarily stores data converted into parallel data by the shift register 11. 13 is a driver circuit, 14 is a heating resistor, 1
5 is a thermal head power supply circuit that supplies driving power to the heat generating resistor 14; 16 is a ceramic substrate including the heat generating resistor 14;

The operation will be explained below.

The thermal head control circuit transfers the binary data binarized by the error diffusion method in the reading section 1 to the shift register 11 as a serial data signal.

When data for l lines is stored in the shift register 11, the data is transferred to the latch circuit 12 as parallel data.
send to The latched l-line data is sent to the driver circuit 13 by a latch signal from the thermal head control circuit 10. In the driver circuit 13, the latch circuit 1
The binary data from 2 and the print strobe width signal (for example, 0, 4 msec) from the control circuit 10 are ANDed to cause the heating resistor 14 on the ceramic substrate 16 to generate heat.

By repeating the above operations for a plurality of lines, data that has been binarized by the error diffusion method can be recorded by a thermal head printer.

FIG. 13 shows an example printed by a thermal head printer.

FIG. 13(a) shows a dark area where the number of black pixels is 1 after the binarization process, and FIG. 13(b) is a diagram in which the binary data of FIG. 13(a) is recorded.

FIG. 13(C) shows a highlight part with a small number of black pixels (1) after the binarization process, and FIG. 13(d) is a diagram in which the binary data of FIG. 13(C) is recorded.

As is clear from Figure 13, in the case of a thermal head printer, the print dot diameter is large for dark areas where there are many black dots in the surrounding pixels, and small for highlight areas where there are few black dots. This is because thermal head printers perform printing operations by heating and cooling the head, so when printing black dots continuously, the print dots become thicker (become white) due to the heat storage effect of the head.Also, there are fewer black dots (white). ), the head is cooled, so
Even if energy is applied to the head, the standard print heat generation temperature is not reached and the print dot diameter becomes small.

That is, in the first embodiment, the print dot diameter was changed by controlling the laser drive current of the laser beam printer, but when using a thermal head printer, the print dot diameter is changed by using the heat storage effect of the head. The diameter can be changed.

That is, as shown in FIG. 13(b), the diameter of the dots becomes larger in the dark areas, so that the spaces between the dots can be prevented from falling out.

Furthermore, as shown in FIG. 13(d), the diameter of the dots becomes smaller in the highlighted area, which prevents the dots from connecting and prevents the unique striped pattern that occurs with the error diffusion method. be able to.

Incidentally, as shown in FIG. 11, the binary data is binarized by the error diffusion method and can be transmitted to the receiving device 8 via the communication control section 4. In this case, compared to the embodiment of bundle 1 described above, there is no need to transmit a flag indicating the density level of the image, so communication efficiency can be improved.

When transmitting data to a partner device, the CPU 104 performs control in response to a transmission command from the operation unit 6.

The receiving device 8 is provided with a recording section having the same configuration as the thermal head printer 7, and controls the thermal head based on the sent binary data to perform recording.

As explained above, according to the second embodiment, the size of the dot diameter is controlled according to the density of the image by recording binary data that has been binarized using the error diffusion method using a thermal head printer. can do.

As a result, it is possible to prevent the occurrence of a striped pattern unique to the error diffusion method in highlight areas, and also to prevent white spots between dots in dark areas, making it possible to reproduce a good image. In this embodiment, the case where image data is binarized using the error diffusion method has been described, but the present invention can also be used when image data is subjected to multi-value processing.

Further, in the first embodiment, a case has been described in which the size of the dot diameter is controlled using a laser beam printer, but the size of the dot diameter can also be controlled by controlling the amount of ink ejected using an inkjet printer.

〔Effect of the invention〕

As explained above, according to the present invention, by outputting image data quantized by the error diffusion method using a thermal head printer, it is possible to reproduce or communicate high-resolution, good images with a simple configuration. A possible image processing method and apparatus can be provided.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a first embodiment of the present invention, FIG. 2 is a block diagram showing details of the reading section l shown in FIG. 1, and FIG. 3 is a diagram showing a luminance-density conversion table. , Figure 4 is 2
A block diagram showing the details of the digitization processing section, Fig. 5 is a diagram showing an example of a weight matrix, Fig. 6 is a diagram showing the flow of binarization processing using the error diffusion method, and Fig. 7 is a diagram showing a semiconductor laser. 1
Figure 8 shows the energy distribution of the laser beam; Figure 9 shows the print dot size determined by binary data and density data; Figure 10 shows the relationship between drive current and optical power; 11 is a block diagram showing a second embodiment of the present invention; FIG. 12 is a block diagram showing details of a thermal head printer; FIG. 13 is a block diagram showing details of a thermal head printer; The figure shows a recording example in the second embodiment, and FIG. 14 shows the problems of the conventional technique. 1... Reading section 2... LBP recording section 3.
...Laser light 4...Communication control section 5.8...
-Receiving device 6...Operation unit 7...Thermal head printer 101...Input sensor unit 102...Image processing unit 103...RAM 104...CPU 1
05...System bus 106...ROM107
...LBP control circuit 108...Laser control circuit 1
09... Laser drive circuit 110... Semiconductor laser 115... Photoreceptor 116... Recording paper 41
7 .=F I a g signal 71.・θ5 Pre-tori] Hit 1 (b) (c) (ρ) Highlight Iku (b) Chinese and Korean language #1 part cc) Reparation part c (L n
(G)C)(〆)

Claims (2)

[Claims] (1) Image processing characterized by having a quantization means for quantizing input image data using an error diffusion method, and an output means for outputting the image data quantized by the quantization means using a thermal head printer. Device. (2) In an image processing method in which image information is communicated from one device to another device, the one device quantizes the image information obtained by reading a document using an error diffusion method and the other device , and the other device records the image information received from the one device and quantized using an error diffusion method using a thermal head printer.