JPH0431467B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an image information processing method, and in particular, a transmission system performs thinning processing to extract read image information every other pixel in the main scanning direction, and a reception system The present invention relates to an image information processing method for increasing the speed of facsimile printing which performs interpolation processing on thinned out pixels.

[Conventional technology]

In a conventional facsimile image information processing method, the transmission system thins out the image information read from the document by alternately extracting only odd-numbered pixels and even-numbered pixels for each line in the main scanning direction. After reducing the amount of information to be transmitted to 1/2, it is transferred to encoding means that performs predetermined encoding.

In the reception system, for each thinned pixel in the line information consisting of the extracted pixel string decoded by a decoding means corresponding to a predetermined encoding, thinning is performed based on the pixel color of each of the multiple extracted pixels adjacent to the thinned out pixel. The pixel color of the pixel is determined, and according to the determination result, interpolation processing is performed on the thinned out pixels to create recording line information, which is then transferred to the recording section.

Here, the predetermined encoding is, for example, CCITT
One-dimensional encoding (MH encoding) compliant with recommendation T30
and two-dimensional encoding (MR encoding).

[Problem that the invention seeks to solve]

In such conventional image information processing methods, the number of pixels that make up one line after extraction is halved, so
Both the band compression encoder in the transmission system and the band expansion decoder in the reception system need to be added with a function to treat the number of pixels in one line length as 1/2, so a general-purpose band compression encoder and band expansion decoder must be added. The disadvantage is that it cannot be handled by the decoder.

[Means for solving problems]

In the image information processing method of the present invention, in the transmission system, a thinning process is performed to alternately extract only odd-numbered pixels and even-numbered pixels in the main scanning direction from the image information read from the original for each line. Coding in which white information consisting of a number of white pixels equal to the number of pixels to be thinned out for the corresponding line is subsequently added to the extracted pixel string of the corresponding line to create each transmission line information, and the transmission line information is encoded in a predetermined manner. The receiving system deletes the added white information from the transmission line information decoded by the decoding means corresponding to the encoding means of the transmitting system, and extracts the extracted pixel string of each line from which the white information has been deleted. For each pixel to be thinned out, the pixel color of the thinned out pixel is determined by referring to the pixel color of each of a plurality of extracted pixels adjacent to the thinned out pixel, and interpolation processing is performed for the thinned out pixel according to the determination result to create a recording line. configured to create information;

〔Example〕

Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a pixel arrangement diagram of transmission line information for explaining the principle in the transmitting system of the present invention, and FIG. 2 is a diagram explaining the principle of thinning pixel interpolation operation in the receiving system of the present invention.

In Figure 1, the horizontal axis indicates the main scanning direction, and the vertical axis indicates the sub-scanning direction.The document size is JISA4, and the number of pixels in one line in the image information before extraction is
This is the case of 1728 pieces.

As shown in Figure 1, the odd numbered pixels for the (i-1)th line, the even numbered pixels for the ith line, and the odd numbered pixels only for the (i+1)th line. A staggered thinning process is performed in which the phase is alternately shifted for each line, and the number of pixels for one line is reduced to 1/2 of the number of pixels in the image information PD ₁ of one line before extraction. . Therefore, the number of pixels in each extraction pixel array DA is 864.

864 white pixels are added as white information DB to each extraction pixel row DA, and 1728 pieces of transmission line information DL are formed in which the number of pixels of one line is equal to the number of pixels of image information PD ₁ .

By transferring the transmission line information DL to a band compression encoder based on a predetermined encoding method, encoding and decoding processing can be performed without making any changes to general-purpose band compression encoders and band extension decoders. Become.

Here, a reduction in compression efficiency occurs due to the addition of the white information DB, but in terms of encoding processing, it is processed at once as 864 white run lengths, so it is practically negligible.

In the receiving system, the decoded transmission line information DL
After deleting the white information DB from , interpolation processing of the thinned out pixels is performed.

In the interpolation process, as shown in Figure 2, the thinned out pixels of interest to be interpolated are the pixels for the i-th line.
Let Pi,j be the pixel Pi-i,j for the immediately preceding (i-1)th line, the pixel Pi,j _-1 for the i-th line, and the pixel Pi,j ₊₁ for the subsequent (i
The pixel Pi ₊₁ , j for the +1)th line is extracted as a reference pixel.

Here, the extracted pixels Pi _-1 ,j, Pi,j _-1 ,
Pi, j ₊₁ , Pi ₊₁ , j are all pixels extracted from image information PD ₁ in the transmission system and transmitted, and are the pixels to be interpolated.
Pi is an adjacent pixel of j.

Next, from the pixel colors of the four extracted reference pixels, (1)
The logical operation shown in the formula is performed to determine the pixel color of pixels Pi and j.

Pi, j = Pi, j _-1 ^× Pi, j ₊₁ ⁺ Pi _-1 , j ^× Pi ₊₁ , j ...(1) In other words, when there are two black pixels among the reference pixels, interpolation Whether the adjacent pixels of the adjacent line to Pi,j, which is the exponent thinned out pixel, are both black,
Alternatively, if adjacent pixels on the same line are both black, it is determined that the pixel color of pixels Pi and j is black.
If three or more reference pixels have a black pixel color, it is of course determined that the pixel is black.

The above-mentioned algorithm for pixel color determination is just one example, and it is possible to change the number of reference pixels, extract multiple reference pixels only from the same line, and perform logical operations from the pixel colors of those reference pixels. It goes without saying that modifications such as determining the pixel color of the pixel to be interpolated are possible.

Next, FIG. 3 is a detailed block diagram of an embodiment of the transmission image processing circuit 1 for realizing the method of the present invention.
FIG. 4 is a time chart for explaining the operation of FIG. 3. The operation of the transmission image processing circuit 1 shown in FIG. 3 will be explained below with reference to FIG. 4.

The timing signal generating circuit 11 generates a clock CL ₂ , an information selection signal SE, and a transmission valid period signal EN ₂ from a clock CL ₁ and a read valid period signal EN ₁ supplied from a reading scanning section (not shown).

That is, as shown in _FIG .
is a clock whose period corresponds to the image information reading speed in the reading scanning section, and the timing signal generation circuit 11
alternately switches the odd-numbered and even-numbered pulses of the clock _CL1 every falling edge of the read valid period signal _EN1 , extracts 1728 bits at a time, and outputs them as the clock _CL2 . Therefore, the period of clock _CL2 is twice the period of clock _CL1 .

The reading valid section signal EN ₁ is clocked from the falling edge.
A signal that is ``low'' for 1728 bits of CL <sub>1</sub> and has an invalid ``high'' period that is at least longer than 1728 bits of CL ₁ before the next falling edge, and is read during the ``low'' period. Image information PD ₁ corresponding to one line is supplied from the scanning section to the data selector 12 .

The timing signal generating circuit 11 counts the clock _CL1 with a counter that is started every time the read valid interval signal _EN1 is converted to "high", and sets the clock to an initial value when 1728 bits have been counted. This counter becomes "high" when counting starts, and remains "high" during the counting period.
generates a continuous information selection signal SE and supplies it to the data selector 12.

The data selector 12 selects and outputs image information PD ₁ while the information selection signal SE is "low" and outputs "high".
During the period, the separately supplied white level W is selectively output.

The timing signal generation circuit 11 becomes "low" when the read valid interval signal EN ₁ becomes "low",
"High" when information selection signal SE becomes "low"
A transmission effective interval signal EN ₂ is generated. Therefore, the active low period of the transmission valid interval signal _EN2 corresponds to 1728 bits of the clock _CL2 .

Processed image information PD ₂ consisting of image information PD ₁ from the data selector 12 and white level W, clock
CL ₂ and transmission effective period signal EN ₂ are transferred to a band compression encoder (not shown).

When the band compression encoder transfers the image information PD ₂ to be processed,
During the period when the transmission valid section signal EN ₂ is active low,
The processed image information PD ₂ is sampled at the rising edge of the clock CL ₂ and consists of an effective extraction pixel array DA followed by a white information DB consisting of the same number of white pixels as shown in FIG. Receive as transmission line information DL for each line.

FIG. 5 is a detailed block diagram showing an embodiment of the received image processing circuit 2 for implementing the method of the present invention, and FIG. 6 is a time chart for explaining the operation of FIG. 5. The operation of the received image processing circuit 2 shown in FIG. 5 will be explained below with reference to FIG. 6.

In FIG. 5, the received image processing circuit 2 includes transmission line information DL decoded from a band expansion decoder (not shown), a clock _CL3 , and _1/2 of the period of the clock CL3.
Periodic clock CL′ ₃ and reception valid interval signal
Receive EN ₃ . Here, the clock _CL'3 is, for example, a clock having the same period as the reading scanning clock _CL1 in the transmission system. The reception valid interval signal _EN3 is a signal that is "low" during the transfer period of the transmission line information DL, and corresponds to 1728 bits of the clock _CL3 .

Clock _CL'3 is supplied to counter 24.
The counter 24 is a 1728-bit counter that is activated when the reception valid interval signal EN ₃ becomes “low” and counts the clock CL′ ₃ by 1728 bits to create the valid information interval signal Q ₁ and send it to the shift register 21. It is supplied to gate 25 which controls the clock supply. Note that the counter 24 is reset when the reception valid period signal _EN3 becomes "high".

The shift register 21 has a capacity to store pixels corresponding to two lines, and receives the clock signal from the gate 25.
During the period when CL′ ₃ is supplied (that is, during the period when the effective information interval signal Q ₁ is “low”), the extracted pixel sequence DA, which is the effective information part of the transmission line information DL, is sampled by the clock CL′ ₃ and stored. . Therefore, at this stage, the white information DB added in the transmission system is deleted.

The data stored in the shift register 21 is, as shown in FIG. ” “4, 4”, even numbered pixels are lined up two by two to form a line of 1728 pixels, followed by odd numbered pixels “1, 1”, “3, 3”
It is shifted sequentially like this.

Each of the logic circuits 22 and 23 performs the logical operation of equation (1) on the pixel color of the pixel thinned out in the transmission system based on the pixel color of each of the four adjacent pixels supplied from the shift register 21. The circuit for execution includes two AND circuits and one OR circuit. The operation of the logic circuits 22 and 23 is similar to the operating principle described above, and the explanation thereof will be omitted.

Clock CL ₃ is a flip-flop (hereinafter referred to as FF).
FF26 uses clock CL' ₃ as its clock, and outputs 1/1 of clock CL' ₃ as its clock.
A selection clock Q2 with a delay of 4 cycles (corresponding to a 1/2 cycle delay of clock _CL'3 ) _and an inverted selection clock _Q2 are output.

The selection clock _Q2 is supplied to the data selector 29, and the inverted selection clock ₂ is supplied to the data selector 30.

When the selection clock _Q2 is "low", the data selector 29 selects the pixel Pi from the shift register 21,
j _-1 is selectively output, and when it is "high", pixels Pi, j are selectively output as interpolation data from the logic circuit 22 and supplied to the data selector 31.

When the inverted selection clock ₂ is "high", the data selector 30 selects and outputs the pixel Pi, j as interpolation data from the logic circuit 23, and when it is "low", selects the pixel Pi, j ₊₁ from the shift register 21. It is output and supplied to the data selector 31.

The reception valid period signal EN ₃ is supplied to the D input of the FF 28 , and the FF 28 uses the clock CL' ₃ as a clock to output a recording line information selection signal Q ₃ and supplies it to the data selector 31 . The recording line information selection signal _Q3 goes low with a 1/2 cycle delay of clock _CL'3 from the falling edge of the reception valid period signal _EN3 , and 1/2 cycle delay of clock _CL'3 from the next falling edge. The signal that becomes "high" controls alternate selection of odd-numbered pixel extraction lines and even-numbered pixel extraction lines.

That is, the data selector 31 selectively outputs the output signal from the data selector 29 when the recording line information selection signal _Q3 is "low", and selectively outputs the output signal from the data selector 30 when it is "high". The recording line information DR is transferred to a recording scanning section (not shown). Therefore, the effective output signal from the data selector 29 is the recording line information DRi for the i-th line in the recording line information DR shown in FIG. 6, which is obtained by interpolating the even-numbered pixels thinned out in the transmission system, and the data selector The effective output signal from 30 is (i
+1) Recording line information DRi for the th line ₊₁
It is.

On the other hand, the valid information section signal from the counter 24
_Q1 is supplied to the D input of FF27. The FF 27 outputs a recording section signal _EN4 using the clock _CL'3 as a clock, and transfers it to the recording scanning section. Therefore, the recording section signal _EN4 becomes a signal delayed by 1/2 period of the clock _CL'3 with respect to the effective information section signal _Q1 .

Further, the recording sampling clock _CL4 whose phase has been inverted by the inverting circuit 32 is transferred to the recording scanning section.

The recording scanning section receives the recording line information DR by sampling it at the recording sampling clock _CL4 while the recording interval signal _EN4 is active low.

〔Effect of the invention〕

As explained above, according to the present invention, in the transmission system, image information read from a document is thinned out by alternately extracting odd-numbered pixels and even-numbered pixels for each line. is reduced to 1/2, replaced with the remaining 1/2 white information and encoded, and the receiving system deletes the added white information and determines the thinned out pixels according to the pixel color of the adjacent pixel. By performing interpolation, it is possible to realize an image information processing method in which the number of pixels for one line length is always constant and a general-purpose band compression encoder and band extension decoder can be used as they are.

[Brief explanation of drawings]

FIG. 1 is a pixel arrangement diagram of transmission line information for explaining the principle in the transmission system of the present invention, FIG. 2 is a diagram explaining the principle of thinning pixel interpolation operation in the reception system of the present invention, and FIG. A detailed block diagram of an embodiment of a transmission image processing circuit for realizing the method,
4 is a time chart for explaining the operation of FIG. 3, FIG. 5 is a detailed block diagram of an embodiment of a received image processing circuit for realizing the method of the present invention, and FIG. This is a time chart to explain the operation. In the figure, 1... Transmission image processing circuit, 2... Reception image processing circuit, 11... Timing signal generation circuit, 12, 29, 30, 31... Data selector, 21... Shift register, 22, 23... Logic circuit, 24...counter, 25...gate.

Claims (1)

[Claims] 1 The transmission system performs a thinning process that alternately extracts only odd-numbered pixels and even-numbered pixels in the main scanning direction from the image information read from the original for each line, and the number of pixels is equal to the number of thinned pixels for each line Transmission line information is created by subsequently adding white information to the extracted pixel string of the corresponding line, and the transmission line information is transferred to a predetermined encoding means, and the receiving system performs decoding corresponding to the encoding means. The white information added from the transmission line information decoded by the decoding means is deleted, and the pixel color of the thinned-out pixel is determined based on the pixel color of each of the plurality of extracted pixels adjacent to the thinned-out pixel for each thinned-out pixel. 1. An image information processing method, characterized in that recording line information is created by determining the thinned out pixels and performing interpolation processing on the thinned out pixels.