JP2002027239A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor which can suppress the deterioration of image quality of an scaled-down image after thinning processing to the utmost. SOLUTION: This image processor can suppress the deterioration of image quality of a scaled-down image after thinning, since this is so arranged as to thin out the lines in thinning positions or the line of one line ahead except lines which bring about larger change of images with the lines just before, when the user performs thinning processing at scale-down processing in the direction of sub scanning.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of reading image data obtained by reading a line by a line image sensor.
The present invention relates to an image processing device that sequentially stores at least two lines of toggle line buffers, and reads out and outputs one line of image data stored in the toggle line buffers from the toggle line buffers in line order.

[0002]

2. Description of the Related Art For example, in a facsimile apparatus, the size of a transmitted document is compared with the size of a document that can be received by a receiving side. If the size of the transmitted document is larger than the size of a receivable document, the transmitted image is reduced. The image is transmitted as an image of a size that can be received by the receiving side.

[0003] When the transmission image is reduced in this way, the reduction processing in the sub-scanning direction can be realized by the thinning processing of thinning out the main scanning lines determined according to the reduction ratio because of the simplicity of the processing. Many.

[0004]

However, in such a conventional thinning process, the main scanning lines determined according to the reduction ratio are simply thinned, so that the image quality of the reduced image may be greatly deteriorated. For example, there is a case where there are many changing points in the sub-scanning direction in the thinning line, that is, a case where the image changes in the vertical line direction.

The present invention has been made in view of such circumstances, and has as its object to provide an image processing apparatus capable of minimizing image quality deterioration of a reduced image after thinning processing.

[0006]

SUMMARY OF THE INVENTION According to the present invention, image data obtained by reading a line by a line image sensor is sequentially stored in each of at least two toggle line buffers, and line data is read from the toggle line buffer. In an image processing apparatus that reads and outputs one line of image data stored in the toggle line buffer in order, the line image data of interest stored in the toggle line buffer and the image data during line thinning processing. Prior to the reference line image data stored in the toggle line buffer, a difference value is calculated for each pixel, the calculated difference value is accumulated, and when the line of interest corresponds to a thinned line, The first difference value between the line of interest and the reference line
Is compared with the reference line and a second cumulative result of the difference value calculated for the reference line when the reference line corresponds to the target line, and the first cumulative result is If the second accumulation result is larger than the second accumulation result, the line image data of interest is thinned out of the image data stored in the toggle line buffer, while the first accumulation result is the second accumulation result. In the following case, the reference line image data is thinned out.

In addition, the image data obtained by reading the image data in line units by the line image sensor is sequentially stored in each of the toggle line buffers for at least two lines. In an image processing apparatus for reading and outputting one line of stored image data, during line thinning processing, target line image data stored in the toggle line buffer and the toggle line buffer before the image data For the reference line image data stored in, a difference value is calculated for each pixel, the calculated difference value is compared with a predetermined value, the number of the difference values exceeding a predetermined value is counted, and the line of interest is thinned out. Difference value between the noted line and the reference line when corresponding to a line The first count value is compared with the reference line and a second count value related to the difference value calculated for the reference line when the reference line corresponds to the target line, and the first count value is compared with the first count value. The counted value is the second
When the first count value is smaller than the second count value, the line image data of interest is thinned out of the image data stored in the toggle line buffer. In this example, the reference line image data is thinned out.

[0008] The toggle line buffer is used commonly for main scanning magnification / reduction processing. The toggle line buffer is used for the main scanning permutation processing.

[0009]

BRIEF DESCRIPTION OF THE DRAWINGS FIG.
An embodiment of the present invention will be described in detail.

FIG. 1 shows a configuration example of a reading system of an image reading apparatus according to an embodiment of the present invention.

In FIG. 1, a line image sensor LI
S is for reading an image of a read document placed on a document table (not shown) line by line, and an analog image signal output from the line image sensor LIS is supplied to an analog / digital converter ADC. The digital signal is converted into a corresponding digital image signal through the shading correction unit SHD.

The shading correction section SHD applies a predetermined shading correction operation to the input digital image signal, and the output signal is applied to the main scanning reduction processing section RDM.

When the reduction processing is designated, the main scanning reduction processing section RDM applies a predetermined nonlinear interpolation operation to reduce the inputted shading-corrected digital image signal, and the output signal thereof is The digital image signal S of the line of interest (hereinafter referred to as “Nth line”) being processed
(N) is added to the digital difference calculation unit DFC and the memory interface MIF. When no reduction processing is designated, the main scanning reduction processing unit RDM directly outputs the input digital image signal after shading correction.

The decimation timing generator TGM outputs a timing signal RLTM for specifying a decimation line in accordance with the demagnification rate in the sub-scanning direction in the demagnification process.
The timing signal RLTM is applied to the memory interface MIF.

The memory interface MIF receives one input signal for two toggle line buffers TLB-A and TLB-B capable of storing digital image signals for one line.
The writing operation and the reading operation of the digital image signal for the line are performed.

The memory interface MIF has the toggle line buffers TLB-A, TL during normal operation.
A write operation and a read operation are alternately performed on BB. That is, in the normal operation, for example, for the Nth line, the input digital image signal (Nth line) is written into the toggle line buffer TLB-A, and the data ((N−1) line) is output from the toggle line buffer TLB-B. (N) is read out and output to the next stage device, and (N + 1)
For the line, the input digital image signal ((N +
1) Write the (line) into the toggle line buffer TLB-B, read out the data (Nth line) from the toggle line buffer TLB-A, and output it to the next-stage device (see FIG. 2A). .

That is, the digital image signal on the Nth line is written when the digital image signal on the (N + 1) th line is written.
Output to the next stage device.

Also, the memory interface MIF is:
In the reduction process, for example, when writing the digital image signal S (N) on the Nth line, the digital image signal on the Nth line is written into the toggle line buffer TLB-A.
Read from the toggle line buffer TLB-B (N-
1) The digital image signal of the line is taken as the digital image signal S (N−1) of the previous line, and the digital difference amount calculation unit DFC
Output to Alternatively, the digital image signal on the Nth line is written into the toggle line buffer TLB-B,
Read from the toggle line buffer TLB-A (N-
1) The digital image signal of the line is taken as the digital image signal S (N−1) of the previous line, and the digital difference amount calculation unit DFC
Output to

The digital difference calculation unit DFC includes a digital image signal S (N) and a digital image signal S (N-
1) and the difference (S (N-1) -S (N)) is calculated,
The difference amount calculated for each pixel is accumulated for one line, and when the accumulated value for one line calculated for the line of interest is larger than the accumulated value of one line before, the period of the next one line is calculated. (Period of the (N + 1) th line), the signal RLSL is raised to logic H. If the accumulated value of one line calculated for the line of interest is equal to or less than the accumulated value of the previous line, the next One line period ((N
In the (+1) th line period), the signal RLSL falls to logic L, and the signal RLSL is applied to the memory interface MIF.

When the timing signal RLTM is output from the thinning-out timing generator TGM, the memory interface MIF does not send out a digital image signal to the next-stage device for a period corresponding to one line (ie, thins out). (Write period of the (N + 1) th line), the signal RLS
When L rises to logic H, the image signal on the (N-1) th line is thinned out and the image signal on the Nth line is sent to the next-stage device (see FIG. 2B).

That is, in this case, the content of the digital image signal on the Nth line is larger than the content of the digital image signal on the (N-1) th line from the previous line, and The (N-1) th line is thinned out and the Nth line is output, so that deterioration in image quality in the sub-scanning direction reduction processing (thinning processing) can be suppressed.

Here, when the digital image signal of the Nth line is written in the toggle line buffer TLB-A,
When the digital image signal stored in the toggle line buffer TLB-A is selected and the digital image signal of the Nth line is written in the toggle line buffer TLB-B,
The digital image signal stored in the toggle line buffer TLB-B is selected.

On the other hand, if the signal RLSL falls to logic L in the next period (write period of the (N + 1) th line), the image signal of the Nth line is thinned out and the (N-1) th line is thinned out. The image signal is sent to the next device (see FIG. 2C).

That is, in this case, the content of the digital image signal on the Nth line is smaller than the content of the digital image signal on the (N-1) th line from the previous line. By outputting the (N-1) th line by thinning out the lines, it is possible to suppress the image quality deterioration from being noticeable in the reduction processing (thinning processing) in the sub-scanning direction.

Here, when the digital image signal of the (N-1) th line is written to the toggle line buffer TLB-A, the digital image signal stored in the toggle line buffer TLB-A is selected, and When the digital image signal of the (N-1) th line is written in the toggle line buffer TLB-B, the digital image signal stored in the toggle line buffer TLB-B is selected.

FIG. 5 shows an example of the configuration of the digital difference calculator DFC.

In the figure, the digital image signal S (N) of the Nth line, which is the line of interest, and the digital image signal S (N-1) of the (N-1) th line just before the line of interest,
Are added to a comparator CMP and a difference unit DEC, respectively.

The comparator CMP converts the digital image signal S (N) to the digital image signal S (N-
1) comparing the digital image signal S (N) with the digital image signal S (N-1), and setting the signal SS1 to a logical L level;
The digital image signal S (N) is the digital image signal S (N-1)
In the case of the following large values, the signal SS1 is set to the logic H level. This signal SS1 is applied to a differentiator DEC.

The differentiator DEC determines that the value of the signal SS1 is logic L
In the case of the level, the operation (S (N) -S (N-1)) is executed. When the value of the signal SS1 is the logical H level, the operation (S (N-1) -S (N) is performed. )), And the operation result is added to the adder ADD1.

The adder ADD1 accumulates the output of the differentiator DEC for all pixels for one line, and the addition result is obtained as an accumulated value DF (N) as a thinning-out determination unit D.
DC1 and the data buffer DBF1.

The data buffer DBF1 has an adder ADD
1 and outputs the accumulated value DF (N-1) of the line (N-1) immediately before the line of interest N, and the output is applied to the thinning-out determination unit DDC1. ing.

The thinning-out determination unit DDC1 calculates the accumulated value DF
If (N) is greater than the accumulated value DF (N-1),
During the next one line period, the signal RLSL rises to the logical H level, and the accumulated value DF (N) is changed to the accumulated value DF (N
-1) In the following cases, during the next one line period, the signal RL
SL is lowered to the logic L level.

FIG. 6 shows one of the digital difference calculation units DFC.
An example of processing for a line is shown.

When reading of the Nth line is started (Step 10
0), the counter m for storing the pixel address is 1
Is substituted into the variable DF (N: 0) for storing the accumulated value of the difference values up to the m-th pixel in the N-th line (Process 1).
01).

Next, at that time, the toggle line buffer T
When the digital image signal of the Nth line is written into LB-A (the result of determination 102 is YES), the digital image signal S (N−1: m) of the m-th pixel one line before is converted to the toggle line buffer TLB-B. And the difference value df (m)
Is calculated (process 103). At this time, if the digital image signal of the Nth line is written to the toggle line buffer TLB-B (the result of the determination 102 is NO), m of the immediately preceding line is obtained
The digital image signal S (N−1: m) of the pixel is read from the toggle line buffer TLB-A, and the difference value df is read.
(M) is calculated (process 104), and the value of the difference value df (m) calculated in processes 103 and 104 is used as a variable DF (N: m).
Is updated (process 105).

Next, it is checked whether or not the value of the counter m matches the number M of pixels for one line (decision 106). When the result of the decision 106 is NO, the value of the counter m is increased by one ( (Process 107) Returning to the judgment 102, the same process is executed for the next pixel address.

When the addition of the difference values for the digital image signals for one line is completed and the result of determination 106 is YES, the value DF (N) which is the accumulated value of the difference values on the Nth line : M) is checked to see if it is greater than the value DF (N−1: M) that is the accumulated value of the difference value one line before (decision 108), and the result of decision 108 is Y
When it becomes ES, the value of the signal RLSL is set to data 1 (corresponding to a logical H level) (process 109).
If the result of decision 108 is NO, signal RLSL
Is set to data 0 (corresponding to the logical L level) (step 110), and the operation for one line is completed.

FIGS. 7 and 8 show an example of processing for one line of the memory interface MIF.

When the reading of the Nth line is started (Process 20)
1) It is checked whether the signal RLTM is data 1 (logic H level) (decision 202). When the result of the determination 202 is NO, the line is not a thinned line, so that a normal toggle operation is performed (step 20).
3).

When the result of the determination 202 is YES, since the line is a thinned line, first, image processing after the toggle line buffer is invalidated (process 2).
04). Then, it is checked whether or not an operation of writing the digital image signal of the Nth line into the toggle line buffer TLB-A is performed (decision 205).

When the result of the judgment 205 is YES, the value of the counter m for storing the pixel address is set to 1
(Process 206), and the toggle line buffer T
Digital image signal S of m-th pixel on line N to LB-A
While writing (N: m), the digital image signal S (N-1: m) of the m-th pixel of the (N-1) th line is read from the toggle line buffer TLB-B (process 207).

Here, it is checked whether or not the value of the counter m coincides with the number of pixels M for one line (decision 208). When the result of the decision 208 is NO, the value of the counter m is increased by one (step 208). Process 209), and return to Process 207, and execute the same process for the next pixel address.

When the result of decision 208 is YES, the value of signal RLSL is set to data 1 (logic level H).
Is checked to see if they match (decision 210). Judgment 21
When the result of 0 is YES, the next line (N +
At the start of reading (1) (process 211), the digital image signal S of the line (N + 1) is stored in the toggle line buffer TLB-B.
In addition to writing (N + 1), the digital image signal S (N) of the Nth line stored in the toggle line buffer TLB-A is read and output to the next-stage device (process 212).

If the result of the judgment 210 is NO, the reading of the next line (N + 1) is started (step 21).
3) The line (N +) is applied to the toggle line buffer TLB-A.
While writing the digital image signal S (N + 1) of 1), the digital image signal S (N) of the Nth line stored in the toggle line buffer TLB-B is read and output to the next-stage device (processing 214). ).

On the other hand, when writing the digital image signal of the Nth line to the toggle line buffer TLB-B,
When the result of Step 05 is NO, the value of the counter m for storing the pixel address is initialized to 1 (Step 21).
5) Write the digital image signal S (N: m) of the m-th pixel in the N-th line to the toggle line buffer TLB-B, and write (N-1) from the toggle line buffer TLB-A.
Digital image signal S (N−1: m) of the m-th pixel in the line
Is read (process 216).

Here, it is checked whether or not the value of the counter m coincides with the number of pixels M for one line (decision 217). When the result of the decision 217 is NO, the value of the counter m is increased by one ( Processing 218) and returning to Processing 216, the same processing is performed for the next pixel address.

When the result of determination 217 is YES, the value of signal RLSL is data 1 (logical level H)
It is checked whether or not they match (decision 219). Judgment 21
9 is YES, the next line (N +
At the start of reading (1) (process 220), the digital image signal S of the line (N + 1) is stored in the toggle line buffer TLB-A.
In addition to writing (N + 1), the digital image signal S (N) of the Nth line stored in the toggle line buffer TLB-B is read and output to the next-stage device (process 221).

If the result of the determination 219 is NO, the reading of the next line (N + 1) is started (step 22).
2) The line (N +) is applied to the toggle line buffer TLB-B.
The digital image signal S (N + 1) of 1) is written, and the digital image signal S (N) of the Nth line stored in the toggle line buffer TLB-A is read and output to the next-stage device (process 223). ).

As described above, in the present embodiment, when performing the thinning process in the reduction process in the sub-scanning direction, the line at the thinning position or the image immediately before the line at the thinning position is more likely to change. Since a large line is left and the other line is thinned out, it is possible to suppress image quality deterioration of the reduced image after thinning out.

In the above-described embodiment, the digital difference calculation unit DFC calculates the difference value of each pixel address for the digital image signal of two lines and accumulates the difference value. Rather than accumulating the difference values, binarizing the difference values with a predetermined threshold value and accumulating the binarization results can reduce the circuit configuration and reduce the device cost. it can.

FIG. 9 shows another example of the digital difference calculation unit DFC. In the figure, the same parts as those in FIG. 5 or corresponding parts are denoted by the same reference numerals.

In the figure, the digital image signal S (N) of the Nth line which is the line of interest and the digital image signal S (N-1) of the (N-1) th line immediately before the line of interest are shown.
Are added to a comparator CMP and a difference unit DEC, respectively.

The comparator CMP converts the digital image signal S (N) to the digital image signal S (N-
1) comparing the digital image signal S (N) with the digital image signal S (N-1), and setting the signal SS1 to a logical L level;
The digital image signal S (N) is the digital image signal S (N-1)
In the case of the following large values, the signal SS1 is set to the logic H level. This signal SS1 is applied to a differentiator DEC.

The differentiator DEC determines that the value of the signal SS1 is logic L
In the case of the level, the operation (S (N) -S (N-1)) is executed. When the value of the signal SS1 is the logical H level, the operation (S (N-1) -S (N) is performed. )), And the operation result is applied to the binarizer BIN.

The binarizer BIN binarizes the output of the differentiator DEC with a predetermined threshold, and the result is added to the adder ADD2.

The adder ADD2 accumulates the output of the binarizer BIN for all the pixels of one line, and the addition result is obtained as an accumulated value DF '(N) as a thinning-out decision unit DDC2. It is added to the data buffer DBF2.

The data buffer DBF2 has an adder ADD
2 and outputs the accumulated value DF ′ (N−1) of the line (N−1) one line before the line of interest N. The output is supplied to the thinning-out determination unit DDC2. Have been.

The thinning-out determination unit DDC2 calculates the accumulated value DF '
If (N) is greater than the accumulated value DF (N-1),
During the next one line period, the signal RLSL is raised to the logical H level, and the accumulated value DF ′ (N) is
In the case of (N-1) or less, the signal RLSL falls to the logic L level during the next one line period.

By the way, the present invention relates to processing at the time of image reduction. Normally, when image reduction processing is performed, image expansion processing is not executed. That is, the image reduction process and the image enlargement process are executed exclusively.

On the other hand, at the time of image enlargement processing, a digital image signal for one line is interpolated by a pixel address corresponding to the enlargement ratio to create a digital image signal for one line after enlargement.
For example, in the case of an enlargement ratio of 200%, an enlarged digital image signal can be created by repeatedly outputting a digital image signal of each pixel address continuously for two pixels.

The enlargement processing in the main scanning direction is performed by the toggle line buffer TLB-A,
This can be realized by applying TLB-B.

As described above, since the image enlargement processing and the image reduction processing are performed exclusively, the toggle line buffers TLB-A and TLB-B can be shared for the image enlargement processing and the image reduction processing. it can.

As described above, the toggle line buffer TLB
FIG. 10 shows a configuration example of a reading system of an image reading apparatus in the case where -A and TLB-B are shared for image enlargement processing and image reduction processing. In the figure, the same parts as those in FIG. 1 and corresponding parts are denoted by the same reference numerals.

In the figure, the line image sensor LI
S is for reading an image of a read document placed on a document table (not shown) line by line, and an analog image signal output from the line image sensor LIS is supplied to an analog / digital converter ADC. The digital signal is converted into a corresponding digital image signal through the shading correction unit SHD.

The shading correction section SHD applies a predetermined shading correction operation to the input digital image signal, and the output signal is applied to the main scanning reduction processing section RDM.

The main scanning reduction processing section RDM performs a reduction processing of the input shading-corrected digital image signal by applying a predetermined nonlinear interpolation operation when the reduction processing is designated. The digital image signal S of the line of interest (hereinafter referred to as “Nth line”) being processed
(N) is added to the digital difference calculation unit DFC and the memory interface MIF. When no reduction processing is designated, the main scanning reduction processing unit RDM directly outputs the input digital image signal after shading correction.

The thinning-out timing generator TGM outputs a timing signal RLTM for specifying a thinning-out line in accordance with the reduction ratio in the sub-scanning direction in the reduction process.
The timing signal RLTM is applied to the memory interface MIF.

The memory interface MIF inputs the input 1 to two toggle line buffers TLB-A and TLB-B capable of storing digital image signals for one line.
The writing operation and the reading operation of the digital image signal for the line are performed.

During normal operation, the memory interface MIF has the toggle line buffers TLB-A, TL
A write operation and a read operation are alternately performed on BB. That is, in the normal operation, for example, for the Nth line, the input digital image signal (Nth line) is written into the toggle line buffer TLB-A, and the data ((N−1) line) is output from the toggle line buffer TLB-B. (N) is read out and output to the next stage device, and (N + 1)
For the line, the input digital image signal ((N +
(1) Line) is written into the toggle line buffer TLB-B, and data (Nth line) is read from the toggle line buffer TLB-A and output (FIG. 2).
(A)). This digital image signal is sent to the next-stage device via the main scanning enlargement processing unit ZMM. During the normal operation, the main scanning enlargement processing section ZMM simply functions as an output gate.

That is, the digital image signal on the Nth line is written when the digital image signal on the (N + 1) th line is written.
Output to the next stage device.

The memory interface MIF is
In the reduction process, for example, when writing the digital image signal S (N) on the Nth line, the digital image signal on the Nth line is written into the toggle line buffer TLB-A.
Read from the toggle line buffer TLB-B (N-
1) The digital image signal of the line is taken as the digital image signal S (N−1) of the previous line, and the digital difference amount calculation unit DFC
Output to Alternatively, the digital image signal on the Nth line is written into the toggle line buffer TLB-B,
Read from the toggle line buffer TLB-A (N-
1) The digital image signal of the line is taken as the digital image signal S (N−1) of the previous line, and the digital difference amount calculation unit DFC
Output to

The digital difference calculation unit DFC includes a digital image signal S (N) and a digital image signal S (N-
1) and the difference (S (N-1) -S (N)) is calculated,
The difference amount calculated for each pixel is accumulated for one line, and when the accumulated value for one line calculated for the line of interest is larger than the accumulated value of one line before, the period of the next one line is calculated. (Period of the (N + 1) th line), the signal RLSL is raised to logic H. If the accumulated value of one line calculated for the line of interest is equal to or less than the accumulated value of the previous line, the next One line period ((N
In the (+1) th line period), the signal RLSL falls to logic L, and the signal RLSL is applied to the memory interface MIF.

When the timing signal RLTM is output from the thinning-out timing generator TGM, the memory interface MIF does not send a digital image signal to the next-stage device during the period of one line (ie, thins out), and (Write period of the (N + 1) th line), the signal RLS
When L rises to logic H, the image signal on the (N-1) th line is thinned out and the image signal on the Nth line is transmitted (FIG. 2).
(B)). This digital image signal is sent to the next-stage device via the main scanning enlargement processing unit ZMM. During the normal operation, the main scanning enlargement processing section ZMM simply functions as an output gate.

In other words, in this case, the content of the digital image signal on the Nth line is larger than the content of the digital image signal on the (N-1) th line from the previous line, and The (N-1) th line is thinned out and the Nth line is output, so that deterioration in image quality in the sub-scanning direction reduction processing (thinning processing) can be suppressed.

Here, when the digital image signal of the Nth line is written in the toggle line buffer TLB-A,
When the digital image signal stored in the toggle line buffer TLB-A is selected and the digital image signal of the Nth line is written in the toggle line buffer TLB-B,
The digital image signal stored in the toggle line buffer TLB-B is selected.

On the other hand, if the signal RLSL falls to logic L in the next period (the writing period of the (N + 1) th line), the image signal of the Nth line is thinned out and the (N-1) th line is thinned out. The image signal is sent to the next device (see FIG. 2C).

That is, in this case, the content of the digital image signal on the Nth line is smaller than the content of the digital image signal on the (N-1) th line from the previous line, so that N By outputting the (N-1) th line by thinning out the lines, it is possible to suppress the image quality deterioration from being noticeable in the reduction processing (thinning processing) in the sub-scanning direction.

When the digital image signal of the (N-1) th line is written in the toggle line buffer TLB-A, the digital image signal stored in the toggle line buffer TLB-A is selected, and When the digital image signal of the (N-1) th line is written in the toggle line buffer TLB-B, the digital image signal stored in the toggle line buffer TLB-B is selected.

In the enlargement processing, the main scanning enlargement processing section ZM
M forms a read address control signal ZMTM and outputs it to the memory interface MIF. As described above, the read address control signal ZMTM is output from the toggle line buffer T on the output side in accordance with the enlargement ratio in the main scanning direction.
It controls the timing of the read addresses of LB-A and TLB-B.

FIG. 11 shows an outline of the address generator of the memory interface MIF.

Referring to FIG. 15, write address generation unit WAD
R generates a write address (pixel address) to be applied to the toggle line buffers TLB-A and TLB-B, and outputs the multiplexers MPX1 and MPX2.
Is input to the input terminal A.

The read address generator RADR generates a read address (pixel address) to be applied to the toggle line buffers TLB-A and TLB-B, and outputs the same to the input terminals B of the multiplexers MPX1 and MPX2. Have been added. Further, the generated read address is controlled by signal ZMTM. For example, the signal ZMTM
Is output, the read address output immediately before is held, and the same read address is repeatedly output.

The multiplexer MPX1 selects a write address applied to the input terminal A or a read address applied to the input terminal B, and selects the toggle line buffer TLB-A.
The selection is controlled by the logic level of the signal applied to the control input C. That is, when a logical high level signal is applied to the control input terminal C, the input terminal A is selected, a write address is output to the toggle line buffer TLB-A, and a logical low level signal is applied to the control input terminal C. Is applied, the input terminal B is selected and the read address is set to the toggle line buffer TLB-A.
Output to

The multiplexer MPX2 selects a write address applied to the input terminal A or a read address applied to the input terminal B, and selects the toggle line buffer TL
BB, and the selection is controlled by the logic level of the signal applied to the control input C. That is, when a logical high level signal is applied to the control input terminal C, the input terminal A is selected, a write address is output to the toggle line buffer TLB-B, and a logical low level signal is applied to the control input terminal C. Is input, the input terminal B is selected and the read address is set to the toggle line buffer TLB.
Output to −B.

The toggle control signal TGL is directly applied to the control input C of the multiplexer MPX2, and the toggle control signal TG is applied to the control input C of the multiplexer MPX1.
A signal in which L is inverted by the inverter INV is added.

Therefore, when a logic H level signal is applied to the control input terminal C of the multiplexer MPX1 and a write address is applied to the toggle line buffer TLB-A, a logical input is applied to the control input terminal C of the multiplexer MPX2. Since an L level signal is applied, a read address is applied to the toggle line buffer TLB-B.

On the other hand, when a logic L level signal is applied to the control input terminal C of the multiplexer MPX1 and a read address is applied to the toggle line buffer TLB-A, the control input terminal C of the multiplexer MPX2 is Is applied with a logic H level signal, so that a write address is applied to the toggle line buffer TLB-B.

As described above, in this embodiment, the toggle line buffers TLB-A and TLB-B are shared for the image enlargement processing and the image reduction processing, so that the apparatus configuration can be downsized. It is possible to reduce the apparatus cost.

In the above-described embodiment, the case where the toggle line buffers TLB-A and TLB-B are shared by the image reduction processing and the image enlargement processing has been described. However, the main scanning permutation conversion processing and the image reduction The toggle line buffers TLB-A and TLB-B may be shared for processing.

Here, the main scanning permutation conversion process is a process of rearranging image data in the main scanning direction. For example, when obtaining a mirror image rotated 180 degrees left or right, This is applied to a case where two sensors are used and image signals are alternately output from two line image sensors.

The image reading apparatus according to the present invention can be applied to a facsimile apparatus, a digital copying machine, a digital scanner, and the like.

[0092]

As described above, according to the present invention,
Image data read by the line image sensor in line units is sequentially stored in each of the toggle line buffers for at least two lines, and the image data stored in the toggle line buffer in the line order from the toggle line buffer is stored in the toggle line buffer. In an image processing apparatus that reads and outputs image data for a line, at the time of line thinning processing, the line image data of interest stored in the toggle line buffer and the image data stored in the toggle line buffer before the image data. For the reference line image data, a difference value is calculated for each pixel, and the calculated difference value is accumulated. When the line of interest corresponds to a thinned line, a first value of the difference between the line of interest and the reference line is calculated. , The reference line, and the reference line correspond to the line of interest. When the second cumulative result of the difference value calculated for the reference line is compared with the second cumulative result, and when the first cumulative result is larger than the second cumulative result, the toggle line Of the image data stored in the buffer, the line image data of interest is thinned out, and when the first cumulative result is less than the second cumulative result, the reference line image data is thinned out. In addition, it is possible to obtain an effect that the image quality deterioration of the reduced image after the thinning can be suppressed.

Further, the image data obtained by reading the image data in units of lines by the line image sensor is sequentially stored in each of the toggle line buffers for at least two lines. In an image processing apparatus for reading and outputting one line of stored image data, during line thinning processing, target line image data stored in the toggle line buffer and the toggle line buffer before the image data For the reference line image data stored in, a difference value is calculated for each pixel, the calculated difference value is compared with a predetermined value, the number of the difference values exceeding a predetermined value is counted, and the line of interest is thinned out. Difference value between the noted line and the reference line when corresponding to a line The first count value is compared with the reference line and a second count value related to the difference value calculated for the reference line when the reference line corresponds to the target line, and the first count value is compared with the first count value. The counted value is the second
When the first count value is smaller than the second count value, the line image data of interest is thinned out of the image data stored in the toggle line buffer. Since the reference line image data is thinned out, the image quality of the reduced image after the thinning can be suppressed from being deteriorated. In addition, since the binarization result of the difference value is integrated without directly integrating the difference value, the effect that the device configuration can be downsized and the device cost can be reduced can be obtained.

Further, since the toggle line buffer is used in the main scanning magnification changing process, the effect that the apparatus cost can be reduced is obtained. Further, since the toggle line buffer is used for the main scanning permutation conversion processing, the effect that the apparatus cost can be reduced is obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration example of a reading system of an image reading apparatus according to an embodiment of the present invention.

FIG. 2 is a timing chart for explaining the output timing of a digital image signal when there is no thinning line.

FIG. 3 is a timing chart for explaining an output timing of a digital image signal when N lines are lines to be thinned and (N-1) lines are thinned.

FIG. 4 is a timing chart for explaining the output timing of a digital image signal when N lines are to be thinned and N lines are thinned.

FIG. 5 is a block diagram illustrating an example of a configuration of a digital difference calculation unit DFC.

FIG. 6 is a flowchart illustrating an example of processing for one line of a digital difference calculation unit DFC.

FIG. 7 is a flowchart showing an example of processing for one line of the memory interface MIF.

FIG. 8 is a flowchart (continued) showing an example of processing for one line of the memory interface MIF.

FIG. 9 is a block diagram showing another example of the digital difference calculation unit DFC.

FIG. 10 is a toggle line buffer TLB-A, TLB-
FIG. 9 is a block diagram illustrating a configuration example of a reading system of the image reading apparatus when B is used for both image enlargement processing and image reduction processing.

FIG. 11 is a block diagram schematically showing an example of an address generator of a memory interface MIF.

[Explanation of symbols]

 DFC digital difference calculation unit MIF memory interface TLB-A, TLB-B toggle line buffer

Claims (4)

[Claims] 1. Image data obtained by reading a line unit by a line image sensor is sequentially stored in each of at least two lines of toggle line buffers.
In an image processing apparatus that reads out and outputs one line of image data stored in the toggle line buffer, at the time of line thinning processing, the line image data of interest stored in the toggle line buffer and the line data preceding the image data For the reference line image data stored in the toggle line buffer, a difference value is calculated for each pixel,
The calculated difference value is accumulated, and when the line of interest corresponds to a thinned line, a first accumulation result of a difference value between the line of interest and the reference line,
Comparing the second cumulative result of the difference value calculated for the reference line when the reference line corresponds to the target line, and determining that the first cumulative result is larger than the second cumulative result Among the image data stored in the toggle line buffer, the line image data of interest is thinned out while the first accumulation result is stored in the second line data.
An image processing apparatus characterized in that the reference line image data is thinned out in the case where the accumulation result is equal to or less than the accumulation result. 2. An image data obtained by reading a line unit by a line image sensor is sequentially stored in each of at least two toggle line buffers, and from the toggle line buffer in a line order,
In an image processing apparatus that reads out and outputs one line of image data stored in the toggle line buffer, at the time of line thinning processing, the line image data of interest stored in the toggle line buffer and the line data preceding the image data For the reference line image data stored in the toggle line buffer, a difference value is calculated for each pixel,
The calculated difference value is compared with a predetermined value, the number of which the difference value exceeds the predetermined value is counted, and when the line of interest corresponds to the thinned line, the difference between the line of interest and the reference line is calculated. The count value of 1, the reference line,
When the reference line corresponds to the target line, a second count value related to the difference value calculated for the reference line is compared, and the first count value is larger than the second count value. In this case, while the line image data of interest is thinned out of the image data stored in the toggle line buffer, if the first count value is equal to or less than the second count value, the reference line image An image processing apparatus characterized in that data is thinned out. 3. The image processing apparatus according to claim 1, wherein the toggle line buffer is used for main scanning scaling processing. 4. The image processing apparatus according to claim 1, wherein the toggle line buffer is used for a main scanning permutation conversion process.