JPH07203210A - Picture processing unit - Google Patents

Abstract

PURPOSE:To obtain the picture processing unit eliminating moire or the like by exchanging signal levels of picture element signals in a received picture element signal string so as to weaken a specific frequency component in the input signal. CONSTITUTION:An input picture element signal S whose picture element consists of 8-picture elements has a value of 256 gradation. A selector 12 acts like selecting usually the signal S or selecting a picture element signal P preceding by B picture elements being a signal to be selected and transferred from a picture input section. A line buffer 11 has a data length of B picture elements in parallel 8-bits, shifts one picture element of the signal S for every input and the picture element signal appears at the output by B-times of shift. On the other hand, an output of the picture element signal received by the line buffer 11 is set to a latch 18. An output Q of the latch 18 and a signal T preceding by B-picture elements being and an output of the buffer 11 are objects of exchange. The signals T, Q are given to a subtractor 14, the difference of them is given to a comparator 15, where the difference is compared with a prescribed value A, and the signals are exchanged only when the signal level difference is within the prescribed value A.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a processing device for removing noise generated at a constant cycle in a continuous signal sequence such as an image signal or an audio signal, and particularly to an image signal input as a pixel signal string of continuous pixels. The present invention relates to an image processing device capable of preventing the generation of moire due to noise generated in a constant cycle.

[0002]

2. Description of the Related Art It is known that moire occurs when an original such as a color printed matter formed of halftone dots is read or when the original is halftone-dotted (dithered) to form an image.
This is caused by the periodicity of the array of photoelectric conversion elements and the periodicity of the input light from the halftone dot original, or the periodicity of the input image information from the halftone original and the halftoning (dithering) in the output stage. The periodicity of the generated signals interferes with each other. Further, as described in Japanese Patent Publication No. 3-32264, in the color image signal, the regularity between the color signals also causes interference and causes moire.

In order to remove such moire noise,
Conventionally, the shape of the photoelectric conversion element light receiving part has been changed to a rhombus to improve on the input side of the photoelectric conversion element, and one frequency component that causes interference is removed by providing a filter before the dithering means of the output stage. is doing. In addition, Japanese Patent Publication No. 3-32
In the technique described in Japanese Patent No. H264, the pixel signal of a specific color is a signal of a constant cycle, but the pixel signals of other colors are randomly changed at intervals with respect to the pixel positions of the specific color, and the rule between the color signals is changed. The sex is eliminated. However, the method as described above has a problem that the apparatus becomes expensive in any case.

[0004]

SUMMARY OF THE INVENTION Therefore, the object of the present invention is to solve the problems of the prior art as described above, to eliminate the frequency components of noise superimposed on a continuous signal sequence at a constant period, and to provide moire with an inexpensive means. An object of the present invention is to provide an image processing device capable of solving the above problem.

[0005]

To achieve this object, an image processing apparatus according to the present invention, as a first means, processes an image signal of a signal level input as a pixel signal string of continuous pixels. In the image processing device, a plurality of pixel signals are sampled from the input pixel signal sequence, the signal level of the sampled target pixel signal is compared with the signal level of a predetermined pixel signal, and the difference is within a predetermined value A. In this case, the signal level is exchanged between the compared pixel signal in the input pixel signal sequence and the predetermined pixel signal. As a second means, in addition to the first means, a pixel signal input is a predetermined pixel signal for sampling all pixel signals of an input pixel signal sequence and comparing the sampled target pixel signal with a signal level. The image processing apparatus is configured so that the signal is located at a position separated from the target pixel signal by a constant distance B in the column.

As a third means, in addition to the second means, a switching determining means for randomly outputting either "1" or "0" is provided, which satisfies the signal level comparison condition and which is the switching determining means. The image processing apparatus is configured to exchange the signal level only when the output value is a predetermined value. As a fourth means, in the second and third means, the constant interval B is set to one half of the moire cycle. As a fifth means,
In the first means, the image processing device is configured to perform sampling at random intervals. As a sixth means, in the fifth means, a predetermined pixel signal whose signal level is compared with that of the target pixel signal is sampled as an adjacent pixel signal. As a seventh means, in the fifth means, the predetermined pixel signal for comparing the signal level with the target pixel signal is a pixel signal at a position separated from the target pixel signal by a constant distance C in the input pixel signal sequence. It was made to be the structure.

As an eighth means, in the first means, a plurality of pixel signals for comparing the signal levels of the target pixel signals are provided, and the difference between the levels of the plurality of pixel signals and the target pixel signal is a predetermined value A. If it is less than or equal to the above, the image processing apparatus is configured to randomly exchange signal levels among three or more pixel signals subjected to signal level comparison in the input pixel signal sequence. As a ninth means, a signal level exchange condition is that the sampled signal level is higher than a predetermined value B. As a tenth means, the image processing apparatus is configured so that the signal level is exchanged a plurality of times in line units or predetermined units of more than that. As the eleventh means,
In the above-mentioned fourth means, means for allowing an input pixel signal sequence to pass through the moiré removal circuit in the moiré test mode, and moiré cycle detection means for detecting a moiré cycle in the moiré test mode provided after the dithering means. A means is provided for setting the input / output delay of the line buffer to a pixel interval of ½ of the moiré cycle given from the moiré cycle detecting means when not in the moiré test mode. As a twelfth means, in the eleventh means, the moire period detecting means is a means for measuring the time from peak to peak of the image signal, and a time from peak to peak corresponding to the time within a predetermined error range. A means is provided for setting the time as a moire cycle when the detection is performed a predetermined number of times in succession.

[0013]

Since the image device according to the present invention is provided with the means as described above, the specific frequency component superimposed on the input pixel signal sequence is weakened and the sampled signal is generated without causing side effect noise. With the configuration in which the signal level exchange condition is that the level is larger than the predetermined value A, the background level signal is not exchanged with the non-background level signal. Further, the configuration in which the signal levels are exchanged a plurality of times in line units or predetermined units more than that can more effectively weaken the specific frequency component superimposed on the input pixel signal sequence.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to the embodiments shown in the drawings. FIG. 1 is a block diagram showing an embodiment of an image output section in an image processing apparatus according to the present invention.
The configuration for removing moire is shown. In the figure,
Reference numeral 1 is a moire removing circuit, 2 is a dithering unit, 3 is a recording control circuit, and 4 is an image forming unit. Each pixel signal of the pixel signal sequence input to the moire removal circuit 1 is a multivalued gradation signal composed of a plurality of bits, and is not a gradation signal by the dither method or the like. This signal sequence includes periodic noise generated in an image input unit (not shown), that is, noise having a frequency component close to the frequency generated by the dithering unit 2. As will be described later, the moire removing circuit 1 does not completely remove the noise superimposed on the input pixel signal sequence, which is one of two frequency components having strong power and interfering with each other, which causes moire. It has a function of weakening the interfering frequency component in the noise in the same manner as other frequency components, and the pixel signal string output from the moire removing circuit 1 is converted into a dither signal by the dithering means 2.

In the dithering unit 2, when the image forming unit 4 has a low gradation expression ability by a large number of gradations, but the resolution (the number of pixels per unit length) is high, each input pixel is inputted. Is expressed as a matrix of multiple dots,
The number of gradations of one halftone dot may be low. As a result, if the matrix is an n × n matrix, the halftone dot signal string output from the dithering means 2 contains n times the frequency components of the output signal string of the moire removing circuit 1.
If the moire removing circuit 1 is not provided, a frequency component close to the above frequency component of the halftone dot signal sequence, which is a frequency component of noise superimposed on the input pixel signal sequence, is a new frequency component of the halftone dot signal sequence ( N times the frequency component of the input signal sequence)
However, due to the insertion of the moire removing circuit 1, such interference does not occur in the image processing apparatus according to the present invention.

The image forming means 4 is based on, for example, an electrophotographic method, and the recording control circuit 3 is configured to include a circuit for controlling the emission time of laser light. The laser light emission time is a time proportional to a signal having a pulse width corresponding to each halftone dot signal value given from the dithering means 2, and the image forming means 4
A dot having a size proportional to the light emission time is formed on the photoconductor drum included in, and toner adheres corresponding to the size of the dot, so that multi-tone density is realized.

FIG. 2 is a block diagram showing details of the moire removing circuit 1 shown in FIG. In the figure, 11 is a line buffer that temporarily stores pixel signals for consecutive B pixels of the input pixel signal sequence, and 12 is a pixel signal that is input to the line buffer 11 from an input pixel that has been transferred from the image input unit side. A selector 13 for selecting whether to use the signal sequence S or the pixel signal P that precedes the B pixels that have been exchanged, and 13 outputs the data (pixel signal) to be output to the dithering means 2 from the line buffer 11. A selector for selecting whether to use the data (pixel signal) T or the data (pixel signal) Q that follows the exchanged B pixel, 14 is a selector of two pixel signals separated by B pixels in the pixel signal column. A subtracter for calculating a signal level difference, 15 is a comparator for comparing whether or not the signal level difference calculated by the subtractor 14 is within a predetermined value A. When the difference is within the predetermined value A, '1' (high Le)
Is output. Reference numeral 16 is an exchange determining means which is one condition for determining whether or not to exchange the signal levels of the two pixel signals for which the subtractor 14 has obtained the signal level difference, and is "1" or "0". Is output at random. Reference numeral 17 is an AND gate, and when the output of the comparator 15 is "1" and the output of the exchange determining means 16 is "1", the output becomes "1", at which time the selectors 12 and 13 are set to P.
The selectors 12 and 13 are controlled so as to select Q and Q.

The operation of the embodiment shown in FIG. 2 will be described below step by step. In the input pixel signal sequence S, one pixel is composed of, for example, 8 bits and has a value of 256 gradations, the selector 12 normally selects S, and an 8-bit parallel pixel signal is input to the line buffer 11. The line buffer 11 is 8-bit parallel and has a length of B pixels. One line pixel of the input pixel signal sequence S is line buffer 11 in the line buffer 11.
Each time it is input to, the pixel is shifted from the left to the right in the figure, and the pixel signal appears at the output by shifting B times. On the other hand, the pixel signal input to the line buffer 11 is also set in the latch 18. Output signal Q from the latch 18 and line buffer 1
The pixel signal T preceding by B pixels output from 1 is a pixel signal to be exchanged under a certain condition, and these image signals T and Q are given to the subtractor 14 and the difference output signal Is supplied to the comparator 15, and the comparator 1
Reference numeral 5 compares the output signal of the difference with a predetermined value A.

That is, the signal exchange is performed only when the signal level difference between the two pixel signals to be exchanged is within the predetermined value A. The signal exchange is performed in order to weaken the specific frequency component of the noise superimposed on the input pixel signal sequence, and from the viewpoint of noise reduction, the signal level difference between the two pixel signals to be exchanged is rather large. It is more effective to weaken the specific frequency component because it breaks the original signal waveform more, but it is configured to exchange signals even if the signal level difference is large, and both signals are effective. In the case of a signal, the signal level after the exchange is significantly different from that before the exchange, and there arises a new problem that the signal level of the pixel adjacent to the exchanged pixel does not match. For example, the signal level before exchange was 5, immediately before it was 7, and immediately after that it was 7, 5, 3
Replacing 20 with 20 results in a transition of 7, 20, and 3, and the exchanged signal 20 disturbs the signal waveform.

Therefore, as described above, it is necessary to permit the exchange only when the difference between the compared two signals is within the predetermined value A. The predetermined value A is experimentally obtained. As the condition for replacement, as described above, the difference between the signal levels is within a predetermined value, and the replacement determining means 16
The condition that the output of "1" is added is added. The exchange determining means 16 is provided with a randomizer and randomly generates "1" or "0". Therefore, by adding such a condition, the exchange is performed at random frequency. The reason for adding such exchange determination means is
If the signal exchange is performed using only the signal level difference without adding the exchange determining means, the exchange is periodically performed, and the noise of the frequency component corresponding to the period is generated, which causes a new interference. This is because it becomes one frequency component. Therefore, it can be seen that if the exchange determining means is set to repeat "1" and "0" alternately, only the period of the periodic exchange is doubled, and the problem is none other than the periodic exchange. That is, the appearance of '1'0' must be random. These two conditions are given to the AND gate 17, and when the AND condition is satisfied, the selectors 12 and 13 select the exchanged signals and incorporate them into the pixel signal train. Hereinafter, such processing is performed for each pixel input.

FIG. 3 is a diagram for explaining the above exchange. The signal level of the input pixel signal sequence in FIG. 9A is the signal level before the exchange processing, and FIG. 9B is the signal level after the exchange processing up to the current target pixel. In this example, the exchange partner of the target pixel signal is a pixel signal that precedes by 8 pixels. That is, the pixel interval B is 8
This is for pixels, and in this case, the line buffer 11 shown in FIG. 2 may be for 8 pixels. As is apparent from the figure, the other party of the target pixel has already exchanged, and there is also a pixel signal at the level of the pixel signal preceding by 16 pixels (2 × B) after the exchange. As the exchange process further progresses in the future, pixel signals that have the level of the pixel signal preceding by 24 pixels (3 × B) or 32 pixels (4 × B) should appear.
Although the figure does not show that the signal level fluctuates from the normal signal level due to noise, the superimposed noise level is also exchanged together, and from the state of (b), the noise periodicity It can be easily inferred that the specific frequency component is destroyed.

It has been experimentally confirmed that the effect of suppressing the moire is greater when the interval B between the pixels to be exchanged is set to ½ of the moire period. Since the moire period is determined by the constituent elements of the device, it is possible to confirm the moire period at the time of design and set the value of the pixel interval B to one half of the detected moire period. It is also possible to automatically detect the moire cycle and set it to ½ of the detected moire cycle of the pixel interval B. in this case,
As shown in FIG. 4, the moire cycle test mode is provided in the image processing apparatus, and the selector 13a causes the moire removing circuit 1 to pass to positively generate the moire at the output portion of the dithering means 2 to detect the moire cycle. The means 22 detects the moiré cycle, and the line buffer 11a is provided with a FI having a capacity for pixels equal to or larger than the maximum value of the variable range of the constant interval B.
It can be realized by configuring with FO.

In the figure, the selector 13a selects one of the signals from the three directions (8 bits each) and outputs it to the dithering means 2 in the next stage. Two of the signals from the above three directions are the signals T and Q shown in FIG. 2, and the input pixel signal sequence newly input to the moire removing circuit 1 is directly input. This input pixel signal sequence is selected by the moire test mode C which is a signal from an operation unit (not shown) or the like. The line buffer (FIFO) 11a is written with the input pixel signal input from the selector 12 in response to the write clock a, and the delay circuit 1 is input when B pixels are input.
The read clock b outputs the read clock b delayed by B pixels from the write clock a. The delay circuit 19 is provided with a counter, and when the counter reaches the value given by the B register 23, the count is stopped and the write clock a is passed to be the read clock b. The B register 23 is supplied with a value of one half of the cycle detected by the moire cycle detecting means 22.
The moire period detection means 22 is provided with, for example, a counter and a peak detection means, measures the time from peak to peak, and when the time that coincides with the measured value within a predetermined error range is continuously detected a predetermined number of times or more, the time. The number of pixels corresponding to is the Moal cycle. In the moire removing circuit 1a shown in FIG. 4, the parts common to the moire removing circuit shown in FIG. 2 are omitted.

FIG. 5 shows another embodiment of the moire removing circuit. In this embodiment, instead of sampling for each input pixel signal, sampling is performed at random pixel intervals and comparison is performed. In this configuration, sampling is performed at a constant cycle as described above, and whether or not replacement is performed according to the replacement determination means. It has the same effect as making a decision. In this embodiment, it is the random number generator 24 that determines the random pixel interval, and the random number generator 20 randomly generates a number within a predetermined number. The line buffer 11b is composed of, for example, a RAM, and when cyclically writing to a predetermined area and writing M pixels, one pixel is read in the written order and the operation of writing the output pixel of the selector 12 to the address is repeated. Such access control is performed by the access control circuit 25. The access control circuit 25 includes a counter. After setting a numerical value obtained from the random number generator 24, the access control circuit 25 increments the counter by one every time one pixel is written in the line buffer 11b. When the counter counts down to 0, the pixel signal level at the address before the random number generation numerical value is read from the current write address held by the access control circuit 25, and is given to the subtractor 14. Immediately after that, the access control circuit 25 latches the current pixel signal level of the input pixel signal sequence S in the latch 18 and supplies it to the subtractor 14, and if the difference between the two signal levels is within the predetermined value A, then the AND gate 17a.
Output becomes '1'. The exchange of the signal level read from the line buffer is performed via the gate 26 on the one hand under the control of the access control circuit 25 and the other via the selector 12.

FIG. 6 is another embodiment of the moire removing circuit used in the image processing apparatus according to the present invention. In this embodiment, the sampling interval of the exchange is random and the random number generator 2
The numerical value generated by 4 is set in the counter 27, and this counter is counted down each time the input pixel signal is input to the line buffer, and when it becomes 0, the sampling target pixel signal is latched in the latch 18, and the AND gate 17 is set. Exchange is performed by opening. The pixel signal that is compared and exchanged with the sampling target pixel signal is a pixel signal that precedes the target pixel signal by a constant pixel interval C, and is the output signal T of the line buffer 11d by setting the pixel length of the line buffer 11d to C. Become. Further, by adding a condition that the sampled signal level is higher than the predetermined value B to the replacement condition, it is possible to prevent the background of the recording paper from being contaminated by noise. FIG. 7 is a block diagram showing a main part of the moire removing circuit of the image processing apparatus according to the present invention.

In the above description, two pixel signals are exchanged, but as shown in FIG.
This is an example of exchanging signals between individual pixel signals, but when exchanging signals among four or more pixel signals, it is easily possible by adding a similar circuit. The moire removing circuit for exchanging is more effective than the above-described embodiment. In the illustrated example, the target pixel signal a
Is randomly exchanged between the pixel signal b preceding by n pixels and the pixel signal C preceding by 2n pixels, and the line buffers 11e and 11f both have a capacity of n pixels. Differences in the signal levels among the three sampled signals are respectively obtained by subtractors 14a, 14b and 14c, and it is checked by comparators 15a, 15b and 15c that the differences are within a predetermined value A.

The exchange circuit 31 makes one of the following connections. That is, first, the connection of a and f, b and e, c and d,
Secondly, a and f, b and d, c and e, thirdly a and e, b and f, c and d ...
All combinations of connections between the groups e and f are connected, and which combination is selected is randomly determined by the randomizer built in the switching circuit 31 each time the signal level is switched. The signal levels are exchanged among the three parties by controlling the selectors 30a, 30b and 30c when the AND gate 17d is "1" as described above.

Further, according to the embodiment as shown in FIG.
It is also possible to pass the moiré removal circuit many times in units of one line or more to improve the moiré removal effect. In this example, moire removal is performed for each line, and the 1-line buffer 33 is used for that purpose. With a plurality of types (according to various embodiments as described above) of moire removing circuits (two types 1c and 1d in the figure), the same removing method as the previous time can be selected when repeating. Different removal methods can be selected. This selection is performed by the control unit 32 controlling the selector 34. The pixel signal sequence output from the moire removal circuit 1c or 1d is input to the selector 35 via the OR circuit 36, and the control unit 3
The selector 35 is controlled by 2 and is output to the dithering means 2 or the 1-line buffer 33. That is, by selecting the output destination of the selector 35 on a line-by-line basis, it is possible to pass the moiré removal circuit only once, or to pass the moiré removal circuit any number of times.

[0029]

As described above, according to the present invention,
By exchanging the signal level between the pixel signals in the input pixel signal sequence, the specific frequency component superimposed on the input pixel signal sequence can be weakened, so moire etc. can be eliminated by an inexpensive means. It is possible to provide an image processing device that can do so. Further, by giving a condition for signal level exchange, the side effect of signal level exchange can be eliminated. Furthermore, in a configuration in which the signal level is exchanged a plurality of times in line units or predetermined units more than that, it is possible to more effectively weaken the specific frequency component superimposed on the input pixel signal sequence. Can be eliminated more effectively.

[0030]

[Brief description of drawings]

FIG. 1 is a block diagram illustrating an example of an image output unit of an image processing system according to an embodiment of the present technology.

FIG. 2 is a block diagram showing an embodiment of the mower main removal circuit shown in FIG.

3A and 3B are diagrams for explaining an image processing system according to the present invention.

FIG. 4 is a block diagram showing another embodiment of the moire removing circuit shown in FIG.

5 is a block diagram showing another embodiment of the moire removing circuit shown in FIG.

FIG. 6 is a block diagram showing another embodiment of the moire removing circuit shown in FIG.

7 is a block diagram showing another embodiment of the moire removing circuit shown in FIG.

8 is a block diagram showing another embodiment of the moire removing circuit shown in FIG.

9 is a block diagram showing another embodiment of the moire removing circuit shown in FIG.

[Explanation of symbols]

1 ... Moire removing circuit, 2 ... Dithering means, 3 ...
..Recording control circuits, 4 ... Image forming means, 11 ...
Line buffer, 12, 13 ... Selector, 14 ...
-Subtractor, 15 ... Comparator, 16 ... Exchange determining means, 19 ... Delay circuit, 22 ... Moire cycle detecting means, 23 ... B register, 24 ... Random number generator, 28 , 29 ... Comparator, 31 ... Exchange circuit, 32 ... Control unit, 33 ... 1 line buffer.

Claims (12)

[Claims] 1. An image processing apparatus for processing an image signal input as a pixel signal sequence of continuous pixels, sampling a plurality of pixel signals from the input pixel signal sequence,
The signal level of the sampled pixel signal is compared with the signal level of a predetermined pixel signal, and if the difference between the sampled signal level and the predetermined pixel signal level is within a predetermined value A, An image processing apparatus, characterized in that signal levels are exchanged between a pixel signal compared with the pixel signal sequence. 2. A predetermined pixel signal for sampling all pixel signals of an input pixel signal sequence and comparing the signal levels of the sampled pixel signals with a predetermined interval from the target pixel signal in the input pixel signal sequence. The image processing apparatus according to claim 1, wherein the signal is at a position apart from B. 3. An exchange determining means for randomly outputting either "1" or "0" is provided, and only when a signal level comparison condition is satisfied and the exchange determining means output value is a predetermined value. The image processing apparatus according to claim 2, wherein the signal levels are exchanged. 4. The constant interval B is ½ of a moire cycle.
The image processing apparatus according to claim 2 or 3, wherein 5. The image processing apparatus according to claim 1, wherein sampling is performed at random intervals. 6. The image processing apparatus according to claim 5, wherein a predetermined pixel signal whose signal level is compared with that of the target pixel signal is sampled as an adjacent pixel signal. 7. The predetermined pixel signal for comparing the signal level with the target pixel signal is a pixel signal at a position separated from the target pixel signal by a constant interval C in the input pixel signal sequence. 5. The image processing device according to item 5. 8. A plurality of predetermined pixel signals for comparing a signal level with a target pixel signal are provided, and when a difference in signal level between each of the plurality of pixel signals and the target pixel signal is within a predetermined value A, The image processing apparatus according to claim 1, wherein signal levels are randomly exchanged among three or more pixel signals subjected to signal level comparison in the input pixel signal train. 9. The sampled signal level is a predetermined value B.
9. The image processing device according to claim 1, wherein the condition that the signal level is larger than that of the signal level is exchanged. 10. The image processing apparatus according to claim 1, wherein the signal levels are exchanged a plurality of times in units of line units or predetermined units of more. 11. A moiré period detecting means for detecting a moiré period in the moiré test mode, which is provided after the dithering means, the means for allowing an input pixel signal sequence to pass through the moiré removing circuit in the moiré test mode. 5. The image processing apparatus according to claim 4, further comprising means for setting the input / output delay of the line buffer to a pixel interval of half of the moiré cycle given by the moiré cycle detection means when not in the test mode. 12. The moire period detecting means detects the time from peak to peak of the image signal, and the time from peak to peak corresponding to the time within a predetermined error range is detected a predetermined number of times in succession. The image processing apparatus according to claim 11, further comprising means for setting the time as a moire cycle.