JP2000050082A - Image input device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image input devise capable of obtaining excellent restored images in which image noise is reduced. SOLUTION: An image processing circuit 205 is composed of a dust data extraction circuit 2051, a dust data shift circuit 2052, a variable power circuit 2053, a dust data correction circuit 2054, a selector 2055, a CPU 2056 and the other image processing circuit 2057. The dust data extraction circuit 2051 detects the information of 'presence/absence of dust', 'position of dust' in a main scanning direction and 'width of dust' from the image data of a CCD sensor 204. The dust data shift circuit 2052 shifts and outputs the image data so as to forcibly omit the image data of a dust part based on the information of the 'position of dust' in the main scanning direction and 'width of dust'. The variable power circuit 2053 enlarges the image data and the dust data correction circuit 2054 performs prescribed computation from peripheral image data and corrects the image data omitted by dust. The selector 2055 switches the image data based on an extracted result in the dust data extraction circuit 2051.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image input apparatus for scanning a document to generate a digital image signal, and more particularly, to an input optical system for optically scanning a document by adhering dust or the like thereto. The present invention relates to an image input device that reduces generated image noise.

[0002]

2. Description of the Related Art Conventionally, as an image input device of this type, a CPU reads shading data,
It is known that it is determined whether each pixel detected by the sensor is a normal pixel or an abnormal pixel, and in the case of an abnormal pixel, the shading data is rewritten based on peripheral data ( For example, JP-A-6-65
No. 89).

[0003]

By the way, in the above-mentioned conventional image input apparatus, the CPU reads the shading data to determine whether each pixel of the CCD sensor is a normal pixel or an abnormal pixel. Although correction for dust and dust adhering to the CCD is effective, there is a problem that it is impossible to recover dust and dust adhering to the CCD sensor and the lens.

An object of the present invention is to provide an image input apparatus in which image data of pixels including noise information corresponding to image noise is omitted from image data to reduce image noise generated in an output image.

Another object of the present invention is to provide, in addition to the above object, an image input apparatus capable of obtaining an output image which does not cause a sense of incongruity even if image data of pixels including noise information is lost. .

Still another object of the present invention is to obtain a good output image by restoring image data of a pixel missing due to noise based on image data of surrounding pixels. To provide an image input device.

It is still another object of the present invention to provide an image input apparatus capable of obtaining a good color output image free from color bleeding and the like in addition to the above objects.

Another object of the present invention is to select whether to delete image data of a pixel including noise information or to restore data of a pixel which has been lost due to noise in accordance with the state of image noise. It is another object of the present invention to provide an image input device capable of obtaining a good output image.

Still another object of the present invention is to determine whether to delete image data of a pixel including noise information or to restore data held by a pixel which has been lost due to noise, in addition to the above object. It is an object of the present invention to provide an image input device which is selected according to the width.

[0010]

In order to achieve the above object, according to the first aspect of the present invention, image data obtained by scanning an original is taken in, and noise corresponding to image noise included in the image data is taken. Image noise position detecting means for detecting a main scanning address where information exists, and image data shifting means for shifting image data in the main scanning direction based on the information of the main scanning address detected by the image noise position detecting means. , Is provided.

According to a second aspect of the present invention, in the image input device according to the first aspect, the image data shifted by the image data shifting means is scaled based on a scaling factor determined by a shift width. Image data scaling means for outputting the image data.

[0012] Further, the invention according to claim 3 is characterized in that image data obtained by scanning an original is taken in, and an image noise for detecting a main scanning address at which noise information corresponding to the image noise included in the image data exists. A position detection unit, and a missing line of the main scanning address based on image data including at least two lines on both sides of the main scanning address among peripheral pixel addresses of the main scanning address detected by the image noise position detection unit. Image data restoring means for restoring data.

Further, according to a fourth aspect of the present invention, in the image input device of the third aspect, the image data of the peripheral pixel address is converted into brightness data and chromaticity data, and the converted brightness data is converted into one data. The chromaticity data of the missing line data is restored by processing the brightness data of the missing line data by processing based on an arithmetic expression, and the chromaticity data of the missing line data is restored by processing the converted chromaticity data based on another arithmetic expression. And

Further, according to a fifth aspect of the present invention, there is provided an image processing method for capturing image data obtained by scanning a document and detecting a main scanning address at which noise information corresponding to image noise included in the image data exists. Noise position detecting means, image data shifting means for shifting image data in the main scanning direction based on the information of the main scanning address detected by the image noise position detecting means, Image data restoring means for restoring missing line data of the main scanning address based on image data including at least two lines on both sides of the main scanning address among peripheral pixel addresses of the main scanning address; and detecting the image noise position. The output of the image data shifting means and the output of the image data restoring means are switched based on the detection result of the means. An image input device, wherein the image data control unit, further comprising a city to obtain.

According to a sixth aspect of the present invention, in the image input device according to the fifth aspect, the image data control means is configured to output the image data when the width of occurrence of image noise is less than a predetermined value. The output is switched to the output of the restoring means, and when the value is equal to or more than a predetermined value, the output is switched to the output of the image data shifting means.

[0016]

The image data obtained by scanning the original is captured by the image noise position detecting means, and the main scanning address at which noise information corresponding to the image noise included in the captured image data exists is detected. The image data is shifted in the main scanning direction based on the detected information of the main scanning address so that the image data including the image noise is forcibly deleted.

The image data shifted by the image data shift means is scaled based on the scaling factor determined by the shift width and output, and the restored image generated by forcibly deleting image data containing image noise is output. The size reduction is compensated.

The missing line data of the main scanning address where the noise information detected by the image noise position detecting means is present is defined as the missing line data of the peripheral pixel address of the main scanning address where the noise information detected by the image noise position detecting means exists. The image data including at least two lines on both sides of the main scanning address is restored by performing arithmetic processing.

When restoring the missing line data, the image data of the peripheral pixel address of the main scanning address where the noise information exists is converted into brightness data and chromaticity data. For example, linear interpolation processing is performed on the converted brightness data to restore the brightness data of the missing line data.
The chromaticity data of the missing line data is restored using, for example, the maximum saturation as the converted chromaticity data.

The output of the image data shifting means and the output of the image data restoring means are switched based on the detection result of the image noise position detecting means at the main scanning address where the noise information corresponding to the image noise exists. Whether to delete the image data of the pixel including the noise information or to restore the data of the pixel missing due to the noise is selected according to the state of the image noise included in the image data.

Whether the image data of the pixel containing the noise information is lost or the data of the pixel missing due to the noise is restored depends on, for example, the width of the image noise generated in the restored image of the image data. It will be selected accordingly.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an image input device according to the present invention will be described below with reference to the drawings.

In the following, an embodiment in which the image input apparatus according to the present invention is applied to a digital color copying machine (hereinafter simply referred to as a copying machine) will be described with respect to (the overall configuration of the copying machine) and (image processing circuit). The configuration and the control flow will be described separately.

(Overall Configuration of Copier) As shown in FIG.
The copying machine 90 includes an automatic document feeder 100, an image reading unit 200, and an image forming unit 300. The copying machine 90 is usually equipped with an automatic original document feeder 10.
The image reading unit 200 reads the document conveyed to the image reading position by the image reading unit 200, and transmits the read image data to the image forming unit 300 to form an image, thereby achieving a copying function.

The copying machine 90 can be connected to an external device (not shown) such as a personal computer by an interface circuit 207. Therefore, in the copying machine 90, an image can be formed by outputting image data read by the image reading unit 200 to an external device or by sending image data from the external device to the image forming unit 300. It has become.

Next, each of the automatic document feeder 100, image reading section 200, and image forming section 300 will be described.

(A) Automatic Document Feeder 100 The automatic document feeder 100 includes a document set tray 10
This is a device that conveys the document set in 1 to the image reading position of the image reading unit 200 and discharges the document onto the document discharge tray 103 after the image reading is completed. The operation of document conveyance is performed by the operation panel 2 provided above the image reading unit 200.
11 (see FIG. 2), and the operation of discharging the original is performed based on a reading end signal of the image reading unit 200. When a plurality of originals are set, these control signals are continuously generated to convey the original, read,
The document discharging operation is performed efficiently.

(B) Image Reading Unit 200 In the image reading unit 200, the reflected light of the original on the original glass 208 irradiated by the exposure lamp 201
Second and third mirrors 202a, 202b and 202
The light is guided to a lens 203 by a mirror group 202 made of c and forms an image on a CCD sensor 204. Exposure lamp 2
01 and the first mirror 202a are scanned by the scanner motor 209 in the direction of arrow A1 at a speed V according to the magnification, and scan the entire original on the original glass 208. Further, with the scanning of the exposure lamp 201 and the first mirror 202a, the second mirror 202b and the third mirror 202 are scanned.
c is scanned in the same direction at a speed V / 2. Note that the scanner motor 209 is controlled by the image processing circuit 205.
This is performed by the CPU 2056 (see FIG. 2).

The position of the scanner composed of the exposure lamp 201 and the first mirror 202a is determined by the amount of movement from the home position by the output of the scanner home sensor 210 and the scanner motor 2.
It is calculated and controlled based on the count value of the step number of 09.

The reflected light of the document incident on the CCD sensor 204 is converted into an image signal in the CCD sensor 204,
The image signal is subjected to analog processing such as amplification, A / D conversion, and digital image processing by an image processing circuit 205 described later in detail, and then sent to the interface circuit 207 and the image forming unit 300.

(C) Image Forming Unit 300 In the image forming unit 300, image data sent from the image reading unit 200 or the interface circuit 207 is cyan (C), magenta (M), yellow (Y), black (K ) Is converted into print data, and the print head 30
Sent to 1. In the print head 301, lasers are respectively emitted according to the image signals of the transmitted printing data of C, M, Y, and K, and the light is one-dimensionally scanned by a polygon mirror 315 to perform each imaging. The photosensitive members Cd, Md, Yd, and Kd in the units 302c, 302m, 302y, and 302k are exposed, respectively.

The imaging units 302c, 302
m, 302y, and 302k, the photosensitive member C
Elements necessary for performing the electrophotographic process are arranged around d, Md, Yd and Kd. The image forming processes are continuously performed by rotating the photoconductors Cd, Md, Yd, and Kd clockwise in FIG. Further, imaging units 302c, 302m, 302y, and 302k required for forming these images.
Are integrated in each process, and are detachable from the copier 90.

The imaging units 302c, 302
m, 302y, 302k
The latent images on d, Yd and Kd are stored in color developing machines 316c and 316, respectively.
16m, 316y and 316k. The toner images on the photoconductors Cd, Md, Yd, and Kd are transferred to transfer chargers 303c, 303m, which are provided inside the loop of the loop-shaped paper transport belt 304 so as to face the photoconductors Cd, Md, Yd, and Kd, respectively. 303y, 3
At 03k, the image is transferred onto the sheet on the sheet transport belt 304.

On the other hand, the sheet on which the latent image is transferred is
The image is supplied to the transfer position in the following order to form an image. Paper of various sizes is set in each of the paper feed cassettes 310a, 310b, 310c, and the desired paper size is determined by paper feed rollers 317a, 317b, and 317c attached to the paper feed cassettes 310a, 310b, 310c, respectively. It is supplied to the transport path 312. The sheet supplied to the conveying path 312 is sent to the sheet conveying belt 304 by a pair of conveying rollers 313. Here, the timing sensor 306 detects a reference mark on the paper transport belt 304 and adjusts the transport timing of the transported paper.

Further, the imaging units 302c,
Each of 02m, 302y, and 302k has a built-in registration correction sensor (not shown). By detecting a reference mark on the paper transport belt 304, the supply timing of the image data is finely adjusted, and the registration shift is performed. Color misregistration is prevented.

The transferred toner image on the sheet is heated and melted by the fixing roller pair 307, is fixed on the sheet, and is discharged to the sheet discharge tray 311. In the case of double-sided copying, the sheet fixed by the fixing roller pair 307 is inverted by the sheet reversing unit 309, guided by the double-sided unit 308, and transferred from the double-sided unit 308 to the transport path 312 by the roller 318. The image is formed on the back side by feeding the sheet.

The paper transport belt 304 can be retracted from the C, M, and Y imaging units 302c, 302m, and 302y by the action of the belt retracting roller 305. Thus, when a monochrome image is formed, the sheet conveyance belt 304 and the photoconductors Cd, Md, and Yd can be brought into a non-contact state, and thus the driving of the C, M, and Y imaging units 302c, 302m, and 302y is stopped. And the photoconductors Cd, Md and Y
The wear of d and peripheral processes can be reduced.

(Configuration of Image Processing Circuit) Image Processing Circuit 20
5 is shown in FIG. The image processing circuit 205 includes a garbage data extraction circuit 2051 and a garbage data shift circuit 205.
2, a scaling circuit 2053, a garbage data correction circuit 2054,
It comprises a selector 2055, a CPU 2056 (central processing unit) and other image processing circuits 2057.

The garbage data extraction circuit 2051 has a CC
From the image data read by the D sensor 204,
"Waste", "waste position" in the main scanning direction,
And a circuit for detecting the information of the "garbage width". In the garbage data extraction circuit 2051, as will be described in more detail later, the image data subjected to the predetermined processing is stored in a FIFO (first-in first-out) buffer memory (hereinafter, simply referred to as FIFO), and the CPU 2056 outputs the data from the FIFO. By reading the data, the above-mentioned information on the garbage is detected. The garbage data shift circuit 2052 forcibly deletes the image data of the garbage portion based on the information of the “garbage position” and the “garbage width” in the main scanning direction extracted by the garbage data extraction circuit 2051. The image data is shifted and output as described above.

The scaling circuit 2053 enlarges image data. This is because the garbage data shift circuit 2052
This is to compensate for the reduction of the image in the main scanning direction caused by performing the process of dropping the pixels in the main scanning direction. The magnification for enlarging the image data is determined by the garbage data extraction circuit 205.
The CPU 2056 decides based on the information on the “garbage width” in the main scanning direction extracted in step 1. Therefore, the larger the dust size, the larger the enlargement ratio in the scaling circuit 2053. The garbage data correction circuit 2054 corrects image data missing due to garbage by performing a predetermined operation from peripheral image data.

The selector 2055 switches the image data based on the extraction result of the garbage data extraction circuit 2051. The switching of the image data is performed according to a command from the CPU 2056, whether to use the output of the scaling circuit 2053, use the output of the dust data correction circuit 2054, or use the output of the dust data extraction circuit 2051. The other image processing circuit 2057 performs image processing such as MTF correction and color conversion other than the processing described above, and outputs the output to the print head 301.

The image processing circuit 205 of FIG. 2 described above
Of the garbage data extraction circuit 205
1. The garbage data shift circuit 2052 and the garbage data correction circuit 2054 will be described with reference to FIGS.
This will be described in more detail.

(A) Garbage data extraction circuit 2051 The details of the garbage data extraction circuit 2051 of the image processing circuit 205 are shown in FIG. The garbage data extraction circuit 2051 includes a shading correction circuit 511, a shading memory control circuit 512, a shading memory 513, and a FIFO1.
A FIFO memory 8, an average / difference value circuit 514, an average value memory 515, and a difference value memory 516. Image data is input from the CCD sensor 204 to the shading correction circuit 511 and the shading memory control circuit 512.

The above-mentioned shading correction circuit 511
The sensitivity unevenness of each pixel caused by the chip variation of the CD sensor 204 is corrected. When correcting the uneven sensitivity by the shading correction circuit 511, the shading plate, the CCD sensor 204, and the first
If dust or dirt is attached to the third mirrors 202a, 202b, 202c and the lens 203,
Since light does not reach the CCD sensor 204, the level of the shading data signal is lower than that of a portion free from dust and dust.

The result is written into the shading memory 513 by the shading memory control circuit 512 and can be read by the CPU 2056 (see FIG. 2). That is, the CPU 2056 reads out the data in the shading memory 513 to detect an abnormality of the shading data, and in the shading state, can specify the “garbage position” and the “garbage width” in the main scanning direction.

The data for which the sensitivity unevenness has been corrected by the shading correction circuit 512 includes a plurality of lines in the sub-scanning direction, and data (eight lines in the present embodiment) are FIFO-registered.
1 to FIFO8.

The average / difference value circuit 514 calculates the average value ((maximum value + minimum value) / 2) of the same pixel in the main scanning direction in the sub-scanning direction based on the data stored in the FIFO1 to FIFO8. The difference value (maximum value−minimum value) is calculated. This result is written to the average value memory 515 and the difference value memory 516, and the
(See FIG. 2), so that reading or writing is possible.

That is, the CPU 2056 reads out the data in the average value memory 515 and the difference value memory 516, thereby obtaining the average value ((maximum value + minimum value) / 2) in a plurality of lines in the sub-scanning direction of the same pixel in the main scanning direction. ) And the difference value (maximum value−minimum value). Based on such information, it is possible to detect an abnormality in the image data in the image reading state.

With the above-described configuration, the CPU 2056 determines the abnormality in the shading state based on the shading data by using the average value memory 515 and the difference value memory 51.
According to 6, the abnormality of the image data in the image reading state can be detected as follows.

In the data extraction circuit 2051, in the optical path diagram shown in FIG. 4 from the original glass 208 to the CCD sensor 204 of the digital copying machine 90 shown in FIG. Since the image is in optical focus, the missing state of image data in the main scanning direction due to dust has a sharp shape (see (1) in FIG. 4). Further, even if the same size of dust and dust is observed, the dust and dust on the original glass 208 are small and the dust and dust on the CCD sensor 204 are projected large when viewed optically. Depending on the size of the document, whether the position where dust is attached is on the original glass 208 or whether the CCD sensor 2
04 can be estimated.

When the position where dust adheres is the first, second, and third mirrors 202a, 202b, 202c and the lens 203, the optical focus is out of focus, so that dust in the main scanning direction is The missing state of the image data has an unsharp shape. In addition, the positions where dust adheres are the first, second, and third mirrors 202a, 202b, and 202c.
In this case, the missing state of the data changes by performing the scan (see (2) to (4) in FIG. 4).
This occurs because the distance between the lens 203 and the first, second, and third mirrors 202a, 202b, and 202c changes as a result of scanning. Therefore, when the missing state of the image data is unsharp, the dust adhesion position can be estimated by controlling the scanner motor 209 and observing the missing state at a plurality of locations.

On the other hand, in normal pixel data, at predetermined pixels in the main scanning direction, the image data changes drastically due to the distribution of image density in the sub-scanning direction. FIG. 5 shows how data changes in the sub-scanning direction at a predetermined pixel in the main scanning direction.

In abnormal pixel data due to dust or dirt, the amount of light incident on a predetermined pixel in the main scanning direction is limited, so that image data in the sub-scanning direction changes little and the luminance signal is low. (High concentration). This situation can be detected by the CPU 2056 observing the average value memory 515 and the difference value memory 516.

(B) Garbage data shift circuit 2052 As described above, the garbage data shift circuit 2052
The image data is shifted and output so that the image data of the dust portion is forcibly deleted based on the information of the "garbage position" and "garbage width" in the main scanning direction extracted by the dust data extraction circuit 2051. I do. FIG. 6 shows the garbage data shift circuit 2052 of the image processing circuit 205.

The garbage data shift circuit 2052 is
O521, write address counter 522, selector 5
23, read address counter 524 and adder 52
5 is provided. The read address of the FIFO 521 in the garbage data shift circuit 2052 is the same as that of the CPU shown in FIG.
The control is performed based on the shift start position signal and the shift width signal output from 2056.

The read address counter 524 counts up an address in synchronization with a clock. The adder 525 adds the address by the data width corresponding to the shift width from the read address counter 524 based on the shift width signal output from the CPU 2056. As a result, the image data of a predetermined width stored in the FIFO memory 521 can be forcibly skipped and deleted.

The selector 523 is connected to the FIFO memory 5
The address to be given to 21 is read address counter 52
4 and an address based on the data of the adder 525 are switched. The switching is performed by the CPU 205 in FIG.
6 is indicated by a shift start position signal input from
Until the address reaches the predetermined address, the output from the read address counter 524 is adopted, and after the address is reached, switching is performed so as to output data output from the adder 525.

The garbage data shift circuit 205 as described above
7, the black streaks 526 in the sub-scanning direction caused by dust data are forcibly deleted by shifting the data in the main scanning direction, as shown in FIG.
Easy-to-view images can be created.

(C) Garbage data correction circuit 2054 The details of the garbage data correction circuit 2054 of the image processing circuit 205 are shown in FIG. The dust data correction circuit 2054 includes a peripheral six-pixel extraction circuit 541, a color space conversion circuit 542 that performs color space conversion of image data from RGB space to YCrCb space, which is a lightness and chromaticity space, a lightness data restoration filter 543, And a reconversion circuit 545 for reconversion from the YCrCb space to the RGB space.

The peripheral six-pixel extracting circuit 541 extracts a 3 × 3 pixel block centered on a missing line of a missing image in the main scanning direction caused by dust or dust, and removes three pixels of the missing pixel line. The surrounding six pixels are extracted.
The color space conversion circuit 542 performs color space conversion of image data from the RGB space to the YCrCb space, which is a lightness and chromaticity space, for these six pixels. The color space conversion is necessary for performing restoration filtering in a different manner for lightness and chromaticity.

The brightness data restoration filter 543 determines the value of the central pixel by calculation based on the peripheral six pixel data extracted by the peripheral six pixel extracting circuit 541.
In the present embodiment, the value of the center pixel is determined by calculating the total value of all "ones obtained by multiplying adjacent horizontal pixels by 1/4" and "ones obtained by multiplying adjacent diagonal pixels by 1/8". Like that. Note that other arithmetic expressions can be used to determine the value of the central pixel.

On the other hand, the chromaticity data restoration filter 544 determines the value of the central pixel by calculation based on the peripheral six pixel data extracted by the peripheral six pixel extracting circuit 541. In the present embodiment, the value of the center pixel is determined based on the “maximum value of all six adjacent pixels”. Note that other arithmetic expressions can be used for determining the center pixel. The corrected image data is re-converted from the YCrCb space to the RGB space by the re-conversion circuit 545.

As described above, in the present embodiment, since restoration filtering is performed in a different manner for lightness and chromaticity, a sharp restoration with little color fringing can be performed in a character image, and a smooth restoration can be performed in a photographic image. Becomes

(Control Flow) Next, a control flow of the CPU 2056 will be described with respect to a main flow, a dust detection processing flow, and a dust position detection processing flow.

(A) Main Flow FIG. 9 shows a main flow of the CPU 2056 control. When the power is turned on, first, in step S1, initialization processing such as parameter setting such as image processing necessary for controlling the copying machine 90 is executed. Next, if dust is attached, image noise is generated in the copy. Therefore, before performing the copy operation, in steps S2 and S3, a dust presence / absence detection process is performed to determine whether dust is attached. And a determination process is performed.

If it is determined in step S3 that there is dust (yes in step S3), the process proceeds to step S3.
In step S4, a dust position detection process is further performed to detect the position where dust is attached, and in step S5, a warning of the cleaning position is issued from the CPU 2056 to the operation panel 211 (see FIG. 2). Next, a correction method control process for controlling which correction method is used in step S6 based on the detection result of the dust position detection process in step S4. In the present embodiment, when the dust data width is one pixel in the detection result of the dust position detection processing, the dust data correction circuit 2054 is used as a correction method, and in other cases, the dust data shift circuit 2052 performs correction. Switching of the correction method is performed by the selector 2055.

Next, in step S7, dust data shift processing is executed, and dust shift amount information and shift start position information are set based on the detection result of the dust position detection processing.

If it is determined in step S3 that there is no dust (no in step S3), the selector 20
Reference numeral 55 does not use any of the garbage data correction circuit 2054 and the garbage data shift circuit 2052.

After performing the above-described steps S1 to S7, the process returns to step S8 after executing other processes such as a copy process.

(B) Dust detection processing flow S2 FIG. 10 shows a dust detection processing flow S2. In the dust presence detection processing flow S2, first, step S11 is executed, and the operation of the scanner motor 209 is started to collect a plurality of lines of image data in the sub-scanning direction. Then, steps S12 and S13 are sequentially executed, and the CPU
2056, the data in the average value memory 515 and the data in the difference value memory 516 of the garbage data extraction circuit 2051 are read.

Next, steps S14 to S16 are sequentially executed to detect whether or not the main scanning all pixel check is completed, and whether both the average value memory 515 and the difference value memory 516 are smaller than a predetermined value. When the main scanning all pixel check is not completed (not completed in step S14) and both conditions are satisfied (steps S15 and S1)
It is determined that there is garbage in all (yes in 6) (step S1)
7) Yes. When the main scanning all pixel check is not completed and at least one of the two conditions is not satisfied, it is determined to be deemed (step S18). These processes are checked for all pixels at the main scanning address until it is determined in step S14 that the process is completed. As a result, it is possible to identify at which position of the main scanning address the dust adheres.

In order to improve the reliability of dust detection, the above operation may be performed a plurality of times, and when the presence of dust is detected a plurality of times, the presence of dust may be determined.

(C) Waste Position Detection Processing Flow S4 FIG. 11 shows a flow S4 of the waste position detection processing. In the flow of the dust position detection processing, first, in step S21, the shading memory 513 is read, and in step S22, the dust data start position is extracted. Then, in step S23, the dust data width is extracted. In addition, since shading data is performed by reading a shading plate, it usually indicates a high luminance level.
On the other hand, when dust is attached, the shading data is lower than a predetermined value, and thus the shading data is estimated to be dust data.

Next, in step S24, focus data is extracted by observing data around the garbage data. In step S24, a difference value between adjacent pixels is calculated, and when the difference data of the difference value is large, it is determined that focus is good (yes in step S25).

If the focus is good, the data width is further focused on, and it is determined in step S26 whether the data width is larger than a predetermined position. If the data width is larger (yes in step S26), the vicinity of the CCD sensor 204 (step S26) is determined. S27),
If smaller (No in step S26), original glass 208
It is estimated that it is near (Step S28), and the cleaning position is determined.

When the focus is not good (n in step S25)
In step o), the position of the scanner is moved (step S2).
9) Reading of the shading memory is performed again (step S30). Then, the amount of change from the previously collected data is obtained, and if the amount of change is large (yes in step S31), the first to third mirrors 202a, 202b, 202
c (step S32), the amount of change is small (step S3).
If “1” is no, the vicinity of the lens 203 (step S3)
3), and the cleaning position is determined.

As described above, the dust adhesion position can be estimated by analyzing the focus data of the dust, the dust data width, and the variation due to the scan position.

[0078]

As described in detail above, according to the present invention,
Since the image data including the image noise is forcibly deleted based on the main scanning address information including the noise information corresponding to the image noise included in the image data,
Image noise is removed from the output image, and a good output image with almost no image noise can be obtained.

Further, according to the present invention, the shifted image data is scaled based on the scaling ratio determined by the shift width and output, thereby forcibly removing image data including image noise. The resulting reduction in the size of the restored image is compensated for, and a restored image without a sense of incongruity can be obtained.

Further, according to the present invention, of the peripheral pixel addresses of the main scanning address where the noise information is present, image data including at least two lines on both sides of the main scanning address is processed to restore the missing line data. This makes it possible to restore the data of a pixel missing due to noise based on the data of surrounding pixels, and obtain a good restored image.

Further, according to the present invention, the image data of the peripheral pixel address of the main scanning address in which the noise information exists is converted into brightness data and chromaticity data, and the brightness data and chromaticity data are arithmetically processed. Thus, a good restored image without color blur can be obtained.

Further, according to the present invention, whether to delete the image data of the pixel including the noise information or to restore the data of the pixel missing due to the noise is determined by the state of the image noise included in the image data. By making a corresponding selection, it is possible to obtain an optimal reproduced image according to the image noise.

Further, in accordance with the magnitude of the width of the image noise generated in the restored image of the image data, the image data of the pixel including the noise information is deleted or the data of the pixel which is missing due to the noise is restored. By switching whether to do so, it is possible to obtain an optimum generated image according to the width of the image noise.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a digital color copying machine including an image input device according to the present invention.

FIG. 2 is a block circuit diagram illustrating a configuration of an image processing circuit.

FIG. 3 is a block circuit diagram illustrating a configuration of a garbage data extraction circuit of the image processing circuit.

FIG. 4 is an explanatory diagram of an operation of a garbage data extraction circuit.

FIG. 5 is an explanatory diagram of another operation of the garbage data extraction circuit.

FIG. 6 is an explanatory diagram illustrating a configuration of a garbage data shift circuit.

FIG. 7 is an explanatory diagram of an operation of the garbage data shift circuit.

FIG. 8 is a block diagram illustrating a configuration of a garbage data correction circuit.

FIG. 9 is a main flow of control.

FIG. 10 is a flowchart of a dust presence detection process.

FIG. 11 is a flowchart of a garbage position detection process.

[Explanation of symbols]

 1 to 8 FIFO 90 Digital copier 100 Automatic document feeder 200 Image reading unit 202 Mirror group 203 Lens 204 CCD sensor 205 Image processing circuit 207 Interface circuit 300 Image forming unit 301 Print head 511 Shading correction circuit 512 Shading memory control circuit 513 Shading Memory 514 Average / difference value circuit 515 Average value memory 516 Difference value memory 521 FIFO memory 522 Write address counter 523 Selector 524 Read address counter 525 Adder 426 Black streak 541 Peripheral 6 pixel cutout circuit 542 Color space conversion circuit 543 Brightness data restoration filter 544 Chromaticity data restoration filter 545 Re-conversion circuit 2051 Waste data extraction circuit 2052 Waste data system Door circuit 2053 scaling circuit 2054 garbage data correction circuit 2055 selector 2056 CPU

Claims (6)

[Claims] 1. An image noise position detecting means for fetching image data obtained by scanning a document and detecting a main scanning address at which noise information corresponding to image noise contained in the image data exists. An image input device, comprising: image data shifting means for shifting image data in the main scanning direction based on the information on the main scanning address detected by the noise position detecting means. 2. The apparatus according to claim 1, further comprising image data scaling means for scaling and outputting the image data shifted by said image data shifting means based on a scaling factor determined by a shift width. Item 2. The image input device according to Item 1. 3. An image noise position detecting means for fetching image data obtained by scanning a document and detecting a main scanning address at which noise information corresponding to image noise included in the image data exists. Image data restoring means for restoring missing line data of the main scanning address based on image data including at least two lines on both sides of the main scanning address among peripheral pixel addresses of the main scanning address detected by the position detecting means. An image input device comprising: 4. The image data of the peripheral pixel address is converted into brightness data and chromaticity data, and the converted brightness data is processed based on a single arithmetic expression to restore the brightness data of the missing line data. 4. The image input device according to claim 3, wherein the chromaticity data obtained is processed based on another arithmetic expression to restore the chromaticity data of the missing line data. 5. An image noise position detecting means for fetching image data obtained by scanning a document and detecting a main scanning address where noise information corresponding to image noise included in the image data exists. Image data shifting means for shifting image data in the main scanning direction based on information of the main scanning address detected by the position detecting means; and a peripheral pixel address of the main scanning address detected by the image noise position detecting means. An image data restoration unit for restoring missing line data of the main scanning address based on image data including at least two lines on both sides of the main scanning address; Image data control means for switching between the output of the data shift means and the output of the image data restoration means; and An image input apparatus characterized by comprising. 6. The image data control means switches to the output of the image data restoration means when the width of occurrence of image noise is less than a predetermined value, and to the output of the image data shift means when the width is more than a predetermined value. The image input device according to claim 5, wherein the switching is performed.