JPH11215371A - Shading correction device and shading correction method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device and method by which a fault in a shading reference value is discriminated properly, even when there are steps in shading reference values for consecutive pixels. SOLUTION: A mean value A of white reference values (however normal values only) of a pixel block F0, including 4 pixels P-4-P-1 that have been processed immediately before a noted pixel P0 is calculated and a range of normal white reference values of the noted pixel P is set based thereon. Then a mean value B of white reference values of a pixel block F1 consisting of 4 pixels, including the noted pixel P0 and three successive unprocessed pixels P+1 to P+3, a mean value C of white reference values of a pixel block F2 succeeding the block F1, and a mean value C of white reference values of a pixel block F3 succeeding the block F3 are calculated. Based on the mean values A, B, C, D, whether or not the noted pixel P0 is a pixel resident in a step is discriminated. When the noted pixel P0 is resident in the step, a range of the normal white reference values of the noted pixel P0 is set, based on the mean value B.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a shading correction device and a shading correction method for reducing density unevenness in a read image in an image reading device such as a facsimile device and an image scanner.

[0002]

2. Description of the Related Art In an image reading apparatus such as a facsimile apparatus or an image scanner, in order to compensate for uneven pixel density caused by uneven illumination of a reading light source, etc.
So-called shading correction is performed. In the shading correction, correction based on the following equation (1) is performed on digital data obtained by converting an analog signal output from an image sensor by an analog / digital converter.

[0003]

(Equation 1)

In the equation (1), the "black reference value" is a value obtained by converting a pixel signal output from the image sensor into digital data when a black black reference image is read, and a "white reference value". "Is a value obtained by converting a pixel signal output from the image sensor into a digital data when a pure white reference image is read. For example, the black reference value can be obtained by turning off the reading light source of the image sensor and performing a reading operation, and the white reference value can be obtained by turning on the reading light source and reading the previously provided white reference plate. Can be obtained.

In this manner, the white reference value and the black reference value are obtained for each pixel read by the image sensor. Therefore, when reading an original, the pixel data output from each pixel read out of the image sensor corresponds to the pixel. By performing the correction of the above equation (1) using the white reference value and the black reference value, it is possible to compensate for uneven density between read pixels. However, if local stains adhere to the white reference plate with use of the image reading apparatus, an accurate white reference value cannot be obtained for a pixel corresponding to the spot of the stain. As a result, shading correction becomes poor.

Accordingly, for example, Japanese Patent Application Laid-Open No.
As disclosed in Japanese Patent Application Laid-Open No. 8-208, processing for correcting a white reference value having an abnormal value is performed. In particular,
When the white reference value of a certain pixel exceeds a certain threshold value, the white reference value of that pixel is replaced with a predetermined correction value. This publication further discloses that the threshold value and the correction value are respectively corrected in order to solve the problem when dirt adheres to the entire white reference plate. The value and the correction value are uniformly applied to all pixels of the image sensor.

[0007]

As an image sensor for reading an image, a contact type image sensor is sometimes applied. The contact type image sensor has a configuration in which read pixels are arrayed over a length that can cover the entire width of a read target. However, since such a large image sensor cannot be constituted by a single chip semiconductor element, a large image sensor unit is constituted as a whole by connecting a plurality of one-dimensional image sensor chips in series.

However, since the characteristics of a plurality of image sensor chips cannot be strictly matched, when the white reference value is detected as described above, as shown by reference numeral 60 in FIG. The occurrence of a step corresponding to the part cannot be avoided. Even if a normal / abnormal judgment of the white reference value is made using a fixed threshold value for the white reference value having such a step between the chips, such judgment is correctly performed for all the pixels. Can not be done. Similarly, even if the abnormal white reference value is corrected to a fixed correction value, such correction is not always appropriate.

On the other hand, when performing a process on a white reference value of a certain pixel of interest, an average value of white reference values of a plurality of pixels near the pixel of interest is used as a threshold value for abnormality determination,
If the average value is used as a correction value, it is considered that abnormality determination and white reference value correction reflecting a local situation near the pixel of interest can be performed. However, in the vicinity of the step of the white reference value, the average value gradually follows the step as the pixel of interest is sequentially updated. Therefore, in the vicinity of the step, even if the average value is an appropriate abnormality determination value or It cannot be a correction value.

Accordingly, an object of the present invention is to solve the above-mentioned technical problem, and it is possible to appropriately determine whether the shading reference value is abnormal even when there is a step in the shading reference value of successive pixels. Accordingly, an object of the present invention is to provide a shading correction device and a shading correction method that can perform good shading correction.

It is a further object of the present invention to correct the shading reference value properly even when there is a step in the shading reference value of successive pixels, thereby achieving good shading correction. It is an object of the present invention to provide a shading correction device and a shading correction method that can be achieved.

[0012]

According to the first aspect of the present invention, there is provided a reading means for reading an image for each pixel and outputting density data of each pixel. A shading correction device for correcting output data based on a shading reference value (WST, BST), wherein a density reference plate is read by the reading means.
(24) a shading reference value storage means (41) for storing the density data of each pixel output from the reading means as a shading reference value when reading the individual pixels, and sequentially outputting individual pixels along a predetermined processing direction. As the target pixel, the shading reference values of a plurality of pixels having a predetermined positional relationship in the vicinity of the target pixel are read from the shading reference value storage means, and based on the read shading reference values of the plurality of pixels, the target pixel Normal value range setting means (44,
S7), if the shading reference value of the pixel of interest is the normal value range value set by the normal value range setting means, the shading reference value is determined to be a normal value, and the shading reference value of the pixel of interest is If the value is out of the range of the normal value set by the normal value range setting means, the shading reference value is determined to be an abnormal value by the abnormal reference value determining means (43, S7); A first block average value calculating means for calculating an average value of shading reference values of constituent pixels of a processed pixel block including a processed pixel having a predetermined positional relationship and having undergone the determination processing by the abnormal reference value determining means; (44, S1) and an unprocessed pixel block including an unprocessed pixel that has a predetermined positional relationship in the vicinity of the pixel of interest and has not been processed by the abnormal reference value determination unit. The second block average value calculating means (47 for determining the average value of the shading reference value adult pixels,
S2) and the step determining means (48, S3, S4), the normal value range setting means, if the step determining means determines that the target pixel is a pixel of the step portion, the average value calculated by the second block average value calculating means A shading correction device characterized by including means (44, S5) for changing the range of the normal value based on
Note that the reference numerals in parentheses indicate corresponding components or processing steps in the embodiment described later. Hereinafter, the same applies in this section.

The reading means may be a scanner unit formed by combining a plurality of semiconductor chips. The unprocessed pixel block may or may not include the target pixel. Further, it is preferable that the second block average value calculation means obtains an average value of shading reference values of constituent pixels for a plurality of (preferably three or more) unprocessed pixel blocks. The step determining means preferably determines whether or not the pixel of interest is a pixel corresponding to a step portion of the shading reference value based on the plurality of average values obtained for the plurality of unprocessed pixel blocks. .

More specifically, the step determining means determines the target pixel as the pixel of the step portion when the average value of the plurality of unprocessed pixel blocks is all larger than a predetermined upper threshold value. S3) and means (S4) for determining the target pixel as a pixel of the step portion when the average value of the plurality of unprocessed pixel blocks is all smaller than a predetermined lower threshold value. Good. In this case, the upper threshold value may be a value obtained by adding a predetermined positive constant (β) to the average value of the unprocessed pixel blocks obtained by the first block average value calculation means. The lower threshold is
It may be a value obtained by subtracting a predetermined positive constant (γ) from the average value of the unprocessed pixel blocks obtained by the first block average value calculation means.

The processed pixel block and the unprocessed pixel block each have at least one, preferably a plurality of pixels (more preferably four or more pixels).
including. The normal value range setting means may set a normal value range based on an average value of shading reference values of a plurality of pixels having a predetermined positional relationship in the vicinity of the target pixel. In this case, the normal value range setting means:
Using a mean value of the shading reference values of the processed pixel blocks calculated by the first block mean value calculating means, for example, a certain range around the mean value is set as a normal value range. Is also good.

According to the first aspect of the present invention, the range of the normal value of the shading reference value is determined based on the shading reference values of a plurality of pixels near the target pixel. The determined local reference value is determined. Therefore, even if the shading reference value greatly varies within the reading range of the reading unit, it is possible to appropriately determine whether the shading reference value of the target pixel is normal or abnormal. As a result, an abnormal shading reference value can be reliably eliminated, and good shading correction can be performed.

Moreover, based on the average of the shading reference values of the processed pixel blocks near the pixel of interest and the average of the shading reference values of the unprocessed pixel blocks,
It is determined whether the pixel of interest is a pixel in the step portion. If the pixel of interest is a pixel in the step, based on the average value of the shading reference values of the unprocessed pixel blocks,
The range of normal values is changed. As a result, the range of the normal value can be appropriately set by eliminating the influence of the step, so that the abnormal shading reference value can be reliably eliminated, and proper shading correction can be achieved.

According to a second aspect of the present invention, the correction value calculating means (44, S1) for obtaining a correction value corresponding to the normal value range set by the normal value range setting means, and the abnormal reference value determining means. Shading reference value correction means (45, S10) for correcting the shading reference value determined to be an abnormal value to the correction value obtained by the correction value calculation means
And the correction value calculating means, based on the average value calculated by the second block average value calculating means, when the step determining means determines that the pixel of interest is a pixel of the step portion. To change the correction value (4
4. The shading correction device according to claim 1, wherein the shading correction device includes:

According to this configuration, the correction value is obtained in accordance with the range of the normal value determined based on the shading reference values of a plurality of pixels around the pixel of interest. Corrected to the value. Therefore, since the correction value can take a value reflecting the local situation around the pixel of interest, it is possible to appropriately correct the shading reference value.

In addition, since the correction value for the pixel in the step portion is changed based on the average value of the shading reference values of the unprocessed pixel blocks, the influence of the step is eliminated.
An appropriate correction value can be set. Thereby, good shading correction can be performed. Note that the correction value calculating means may use the center value of the normal value range as the correction value.

Further, the correction value calculating means may determine the average value of the shading reference values of a plurality of pixels having a predetermined positional relationship in the vicinity of the pixel of interest irrespective of the normal value range as the correction value. . According to a third aspect of the present invention, the first block average value calculation means performs an average value calculation using only the shading reference value determined to be a normal value by the abnormal reference value determination means (S7). , S8),
3. The shading correction device according to claim 1, wherein the normal value range setting means determines the normal value range based on the average value calculated by the first block average value calculation means. It is.

According to this configuration, the range of the normal value is determined by using only the shading reference value determined as the normal value, so that the range of the normal value can be determined more appropriately. Moreover, since the average value obtained by the first block average value calculation means can be used for setting the normal value range, the configuration and processing of the apparatus can be simplified.

According to a fourth aspect of the present invention, there is provided a shading correction method for reading an image for each pixel and correcting output data of reading means for outputting density data of each pixel based on a shading reference value. Acquiring density data of each pixel output from the reading means when the density reference plate is read by the means as a shading reference value, and sequentially setting individual pixels as target pixels along a predetermined processing direction. A normal value range setting step (S7) of determining a normal value range of a shading reference value for each pixel of interest based on the obtained shading reference values of a plurality of pixels having a predetermined positional relationship in the vicinity of the pixel of interest; If the shading reference value of the pixel is a value within the range of the normal value set above, the shading reference value is positive. If the shading reference value of the target pixel is determined to be a normal value and the value of the shading reference value of the pixel of interest is out of the range of the normal value set in the normal value range setting step, the shading reference value is determined to be an abnormal value. A value determination step (S7) and a shading reference of constituent pixels of a processed pixel block including a processed pixel that has a predetermined positional relationship in the vicinity of the pixel of interest and has been subjected to the determination processing by the abnormal reference value determination step. A first block average value calculation step (S1) for obtaining an average value of the values, and an unprocessed pixel which has a predetermined positional relationship near the pixel of interest and has not been processed by the abnormal reference value determination step. Second block average value calculation step (S2) for obtaining the average value of the shading reference values of the constituent pixels of the unprocessed pixel block
And a step determining step (S3, S4) for determining whether the pixel of interest is a pixel corresponding to a step of the shading reference value based on the average values obtained in the first and second block average value calculation steps, respectively. The normal value range setting step is based on the average value calculated in the second block flat value calculation step if the target pixel is determined to be a pixel in the step portion in the step determination step. And a step (S5) of changing the range of the normal value.

According to this method, the same effect as that of the first aspect is achieved. According to a fifth aspect of the present invention, there is provided a correction value calculating step (S1) for obtaining a correction value corresponding to a normal value range set in the normal value range setting step, and an abnormal value is determined in the abnormal reference value determining step. A shading reference value correction step (S10) of correcting the determined shading reference value to a correction value obtained by the correction value calculation step, wherein the correction value calculation step includes: 5. The method according to claim 4, further comprising the step of: (S6) changing a correction value based on the average value calculated in the second block average value calculation step if the pixel is determined to be a pixel in the step portion. It is a shading correction method described.

According to this method, the same effect as that of the second aspect is achieved. According to a sixth aspect of the present invention, in the first block average value calculation step, an average value calculation is performed using only the shading reference value determined to be a normal value in the abnormal reference value determination step (S7, S8). )
The normal value range setting step includes a step (S8) of determining the range of the normal value based on the average value calculated in the first block average value calculation step. 5. The shading correction method according to 5.

According to this method, the same effect as the third aspect of the invention is achieved.

[0027]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 is a block diagram showing an electrical configuration of a portion related to image processing of a facsimile apparatus to which an embodiment of the present invention is applied. The document set on the facsimile machine is read by the scanner 1. The scanner 1 includes a contact image scanner unit.

The image data of the document read by the scanner 1 is supplied to an input interface unit 2, where the sample and hold processing of the signal is performed. In this embodiment, the input interface unit 2 is configured by an analog circuit, and the above processing is performed in an analog manner. The image data processed by the input interface unit 2 is applied to a digital AGC circuit 3, where a gain control for keeping the level of a signal (image data) within a desired range is performed, and an analog signal is converted to a digital signal. Is performed.

Next, the gain-controlled image data is applied to a shading correction circuit 4 to reduce or eliminate shading distortion. The shading distortion refers to unevenness in illumination of a reading light source when reading an original by the scanner 1 or unevenness in density between pixels due to other causes. This embodiment is characterized by the configuration and processing of the shading correction circuit 4.

The image data from which the shading distortion has been reduced or removed is next supplied to the area separating circuit 5. In the region separating circuit 5, the input image data is image data (character data) obtained by reading characters, image data (photo data) obtained by reading a photograph, or a printed photograph, such as newspaper or magazine. It is determined whether the image data is image data (halftone data) of halftone dots obtained by reading the photographic image.

If character data, photograph data, and halftone data are mixed in the input image data, area separation of each data is performed. On the output side of the region separation circuit 5, a filter unit 6 that performs different processing for each type of region-separated data is connected. The filter unit 6 includes a differentiation filter 6A, an integration filter 6B, an integration differentiation filter 6C, and a data through filter 6D. Then, the filter unit 6 receives, from the region separation circuit 5, a region separation signal indicating which region each pixel belongs to, and performs an appropriate filter process on each pixel. That is, the character data is stored in the differential filter 6.
A, and a process for sharpening the outline is performed. The halftone data is supplied to the integration filter 6B, and the data is smoothed. Also, the boundary pixels of each area are provided to the integral / differential filter 6C, and are subjected to processing for alleviating a sudden change in the density of different pixel information. Further, data other than the character data and the halftone dot data, that is, the photographic data, is given to the data through filter 6D and sent to the next circuit as it is.

The data from the filter section 6 is supplied to a zoom / smoothing circuit 9. In the case of enlarging or reducing an image, the zoom / smoothing circuit 9 performs an enlarging or reducing process and a process of correcting image distortion accompanying the enlarging or reducing process. If the image is not enlarged or reduced, the zoom / smoothing circuit 9 does not perform any processing on the image data.

The image data that has undergone the processing up to the zoom / smoothing circuit 9 is then subjected to processing in one of the following circuits depending on the type, and multi-valued data (for example,
Conversion from 8 bits (0 to 255) to binary data ("0" or "1") is performed. That is, when the image data is photograph data or halftone data and is to be subjected to halftone processing, the image data is supplied to the γ correction circuit 10 and the sensitivity characteristics of the data are adjusted to match the characteristics of human eyes. Will be corrected. Further, after being given to the error diffusion circuit 11 and subjected to a process for good halftone expression,
It is binarized.

On the other hand, if the image data is character data and data to be subjected to simple binarization processing, it is supplied to the binarization circuit 12. The binarization circuit 12 adjusts a slice level for binarization to distinguish a background from characters, line drawings, and the like. At this time, an automatic density adjustment process is also performed so that the density becomes appropriate. The output of the binarizing circuit 12 is supplied to an isolated point removing circuit 13 to remove isolated black points and white points that appear due to noise or the like.

The binarized data generated by the above processing is supplied to the output circuit 14 and output to a transmission circuit (not shown) or to a printing circuit. FIG. 2 is a diagram for explaining the configuration of the shading correction circuit 4.
In the facsimile apparatus according to this embodiment, the roller 21
A predetermined position on the conveyance path 22 of the document conveyed by is set as a reading position 23. At the reading position 23, for example, a white reference plate 24 as a density reference plate extending in a direction perpendicular to the transport direction of the document transport path 22 is arranged. Further, a lamp 25 for illuminating the original conveyed through the white reference plate 24 or the original conveyance path 22 at the reading position 23, and reading the original content at the white reference plate 24 or the reading position 23 illuminated by the lamp 25. Image scanner unit 26 for
Is arranged.

The contact type image scanner unit 26 includes:
Read pixels are arranged in a straight line over a range of length covering the entire width of the document. Since such a contact type image scanner unit 26 cannot be constituted by a single semiconductor chip, it is constituted by connecting a plurality of line sensor chips in series. However,
Since a plurality of line sensor chips usually have different characteristics, a step occurs in the output level of the contact image scanner unit 26 near the junction of the chips.

The image data read by the image scanner unit 26 is supplied to the shading correction circuit 4 via the input interface unit 2 and the digital AGC circuit 3 as described with reference to FIG. The shading correction circuit 4 includes a white reference value WST corresponding to each of the plurality of read pixels of the image scanner unit 26.
And a line memory 4 for storing the black reference value BST
1 and the white reference value WS stored in the line memory 41.
T and the input image data D based on the black reference value BST.
And a correction calculation unit 42 that performs shading correction calculation according to the following equation (2) to calculate data SOUT after shading correction.

[0038]

(Equation 2)

In the above equation (2), OFFSET represents an offset value. For example, when the analog / digital converter provided in the digital AGC circuit 3 has a data depth of 7 bits, the minimum value of the white reference value is “00”.
If it is "00111", it means that the gradation expression is substantially performed by 4 bits. Therefore, by subtracting a constant offset value OFFSET from the divisor and the dividend in the shading correction operation expression, the data representing the density in 7-bit gradation is represented by the data S after the shading correction.
OUT can be obtained. That is, the dynamic range of the corrected data SOUT can be increased by effectively utilizing the data depth of the analog / digital converter.

The shading correction circuit 4 further includes:
Abnormality determination unit 43 for performing abnormality determination of white reference value WST
A normal value range setting unit 44 for determining a normal value range of the white reference value WST; and a white reference value correction unit 45 for correcting the abnormal white reference value WST. Furthermore, the shading correction circuit 4 performs a series of several pixels (four pixels in this embodiment) when performing an abnormality determination and correction of the white reference value WST using pixels read from the image sensor unit 26 as a pixel of interest sequentially from one end. A block average value calculator 47 for calculating the average value of the white reference value WST of each constituent pixel of the pixel block composed of
A step determining unit 48 for determining whether or not the target pixel is a pixel of the step.

FIG. 3 is an illustrative view for explaining the principle of correction of the white reference value WST in the shading correction circuit 4. In the shading correction circuit 4, after the white reference value WST from the image scanner unit 26 is stored in the line memory 41, a plurality of pixels are sequentially set as a target pixel one by one from one end pixel, and the white reference value of the target pixel is set. WS
It is determined whether or not T is a normal value, and if it is abnormal, a process for replacing it with an appropriate correction value is performed.

For example, when the pixel P10 in FIG. 3 is the pixel of interest, the normal value range setting unit 44 sets the constituent pixels of the processed pixel block consisting of the immediately preceding pixels (for example, four pixels) P9, P8, P7, and P6. The average value A of the white reference value WST is calculated, and a certain range centered on the average value A, that is, a range from A-α (α is a positive constant) to A + α,
This is set as the range of the normal value of the target pixel P10. In other words, A + α is the upper limit threshold of the normal value, and A−α
Is set as the lower threshold of the normal value.

The abnormality determination unit 43 compares the white reference value WST of the pixel of interest P10 read from the line memory 41 with the upper threshold value A + α and the lower threshold value A-α, and the white reference value WST is normal. Determines if the value is within the value range. If the white reference value WST is a value within the range of the normal value,
The abnormality determination unit 43 determines that the white reference value WST is a normal value, ends the process for the target pixel P10, and performs the same process using the next pixel P11 as the target pixel.

On the other hand, in FIG.
As shown by 10b, due to local contamination on the surface of the white reference plate 24, adhesion of dust, or electrical noise,
When the white reference value WST of the pixel of interest P10 is out of the range of the normal value, the abnormality determining unit 43 determines that the white reference value WST is an abnormal value, and determines that the abnormality is abnormal. The signal is given to the white reference value correction unit 45. In response to this, the white reference value correction unit 45 replaces the white reference value WST with the correction value CV, and replaces the correction value CV with the target pixel P10.
Is written into the line memory 41 as the white reference value of the. After that, similar processing is performed with the next pixel P11 as a target pixel.

The correction value CV may be equal to the average value A. In this case, the average value A is set to the normal value range setting unit 4.
4 to the white reference value correction unit 45. Note that the normal value range setting unit 44 does not use the white reference value data of the pixel for which the initial white reference value WST is determined to be an abnormal value when calculating the average value of the immediately preceding pixels. An average value calculation is performed using the white reference value WST for several pixels. Specifically, the normal value range setting unit 44 determines the white reference value W of the pixel of interest.
When the determination result that ST is a normal value is received from the abnormality determination unit 43, in order to set the normal value range of the next pixel, the target pixel at that time and several pixels immediately before it are set. Find the average value of On the other hand, when the determination result that the white reference value WST of the target pixel is abnormal is received from the abnormality determination unit 43, the average value calculation is not performed and the normal value range is not updated.

The average value calculation in the normal value range setting section 44 may be a calculation for obtaining a simple average value of the pixel of interest and several pixels immediately before it, or a moving average value obtained by the following equation (3). The required calculation may be used. (Moving average value) = Total− (previous moving average value) + (white reference value of target pixel) (3) Here, Total is the white reference value of the target pixel and several pixels immediately before it. Is the sum of

For some pixels at the beginning of the process, the number of pixels for calculating the average value is insufficient, so that the normal / abnormal judgment and correction of the white reference value are not performed for these pixels, or The normal / abnormal judgment and correction may be performed based on a predetermined normal value range and a predetermined correction value. FIG. 4 is an illustrative view for explaining a principle of detecting a step of a white reference value in a joint portion of the image sensor chip. Block average value calculator 47
Is the average value B of the white reference value WST of the block F1 of a total of four pixels including the pixel of interest P0 and the three unprocessed pixels P + 1, P + 2, and P + 3 that follow it, and the four unprocessed pixels following the pixel P + 3. Average value C of white reference value WST of block F2 including pixels P + 4, P + 5, P + 6, P + 7, and four unprocessed pixels P + 8, P + 9, P + following pixel P + 7.
The average value D of the white reference value WST of the block F3 including 10, P + 11 is calculated. The calculated average values B, C, and D are provided to the level difference determination unit 48. The step determining unit 48 also includes the above-described average value A (the four processed pixels P-1, P-2, P-3, P-3 immediately before the pixel of interest) calculated by the normal value range setting unit 44. -4 containing -4
The white reference value of 0 (however, only the normal value) is given.

When the following condition or the following condition is satisfied, the step determining section 48 determines that the target pixel is a pixel of the step section, and outputs a step detection signal. A + β <B and A + β <C and A + β <D where β is a positive constant.

A−γ> B and A−γ> C and A−γ> D where γ is a positive constant. The above condition is satisfied when the white reference value WST of the target pixel P0 is the first pixel of the step portion where the white reference value WST of the pixel of interest rapidly increases, as shown in FIG.

The above condition is that the target pixel P0 is the top pixel of the step where the white reference value WST is sharply smaller than the white reference value WST of the previous pixel as shown in FIG. 4 (b). Holds if. As shown in FIG. 4C, when the target pixel P0 is not a pixel in the step portion, none of the above conditions is satisfied. In this case, the following condition should be satisfied.

A + β ≧ B ≧ A−γ When none of the above conditions is satisfied, the abnormal value is included in the reference value of the pixel of the block corresponding to the average value B (that is, the block including the target pixel). Therefore, it is determined that the average value B is an abnormal value. When the level difference determination section 48 outputs a level difference detection signal,
The normal value range setting unit 44 fetches the above average value B from the step determining unit 48, and uses this average value B in place of the average value A when setting the normal value range. In this case, the correction value CV is also made equal to the average value B.

FIG. 5 is a flowchart showing the above-mentioned processing by the shading correction circuit 4 collectively. The average value A of the white reference value WST of several pixels (for example, four pixels) immediately before the pixel of interest is calculated by the normal value range setting unit 44 (step S1), and the three blocks after the pixel of interest are calculated by the block average calculation unit 47. Average value B of white reference value WST of
C and D are calculated (step S2).

Then, it is determined whether or not the above condition is satisfied (step S3). If this condition is not satisfied, it is further determined whether or not the above condition is satisfied (step S4). If either of the conditions is satisfied, the average value A is replaced with the average value B (step S5), and the correction value CV is also replaced with the average value B (step S6), and the white reference of the target pixel is set. Value WS
It is determined whether T is within the normal value range A ± α (step S7). If the white reference value WST of the target pixel is within the normal value range A ± α, an average value A for processing the next pixel is obtained using the white reference value WST of the target pixel (step S8). ), Update the target pixel (step S9), and return to step S2. On the other hand, if the white reference value WST of the target pixel is out of the normal value range A ± α, the white reference value WST of the target pixel is corrected to a correction value CV (step S10), and the correction value CV is set to a line. While writing to the memory 41, the pixel of interest is updated (step S
9) Return to step S2. At this time, the average value A is not updated.

If none of the above conditions are satisfied, it is further determined whether the conditions are satisfied (step S11). Then, if the condition is satisfied, the processing from step S7 is performed. On the other hand, when the condition is not satisfied, the same determination as in step S7 is made (step S12).
The pixel of interest is updated to the next pixel without updating (step S9). This is because if the condition is not satisfied, there is a high possibility that the white reference value WST of the pixel in the pixel block corresponding to the average value B contains an abnormal value. This is to prevent the influence of WST from affecting the normal / abnormal determination of the white reference value of the next pixel. Since the average value B is an average value of the white reference value WST before the determination of normality / abnormality, it is preferable to perform such processing.

As described above, according to this embodiment, based on the average values A, B, C, and D of the white reference values WST of the pixel blocks before and after the pixel of interest, the white reference corresponding to the joint portion of the chip is obtained. The level difference of the value WST is detected, and for the pixels in the level difference portion, the range of the normal value and the correction value are set based on the average value of the white reference values WST of a plurality of pixels after the level difference. Therefore, the image scanner unit 26
Irrespective of the level difference of the output level, it is possible to appropriately determine whether or not the white reference value is normal / abnormal and correct it.

Further, in this embodiment, the range of the normal value and the correction value are set for each pixel of interest based on the average value of the white reference values of several pixels near the pixel of interest. That is, a local normal value range and a correction value reflecting the situation around the target pixel are set. Thereby, even when the output level of the pixel within the reading range of the image scanner unit 26 is not uniform,
It is possible to appropriately determine and correct the normal / abnormal of the white reference value for all the read pixels. as a result,
Good shading correction becomes possible.

Although one embodiment of the present invention has been described, the present invention can be embodied in other forms.
For example, in the above-described embodiment, the configuration for correcting the white reference value has been described. However, similar normal / abnormal determination and correction may be performed for the black reference value. However, since the black reference value is generally detected in a state where the lamp 25 is turned off, there is a low possibility of noise as in the case of the white reference value. Is not necessary.

The present invention can be applied not only to a facsimile apparatus but also to an image scanner and a digital copying machine. In addition, it is possible to make various design changes within the technical scope described in the claims.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration related to image processing of a facsimile apparatus to which an embodiment of the present invention is applied.

FIG. 2 is a block diagram illustrating a configuration related to shading correction.

FIG. 3 is an illustrative view for explaining the principle of normal / abnormal judgment and correction of a white reference value;

FIG. 4 is an illustrative view for explaining a principle of detecting a step of a white reference value generated in a joint portion of a chip;

FIG. 5 is a flowchart illustrating a flow of a process for determining whether or not a white reference value is normal / abnormal and for correcting the white reference value.

FIG. 6 is a diagram illustrating an example of a detection result of a white reference value by a contact image scanner.

[Explanation of symbols]

 Reference Signs List 1 scanner 4 shading correction circuit 26 contact image scanner unit 41 line memory 42 correction calculation unit 43 abnormality determination unit 44 normal value range setting unit 45 white reference value correction unit 47 block average value calculation unit 48 step difference determination unit

Claims (6)

[Claims] 1. A shading correction device for reading an image for each pixel and correcting output data of a reading means for outputting density data of each pixel based on a shading reference value. A shading reference value storage means for storing density data of each pixel outputted from the reading means as a shading reference value when reading is performed, and each pixel is sequentially set as a target pixel along a predetermined processing direction. The shading reference values of a plurality of pixels having a predetermined positional relationship in the vicinity of are read out from the shading reference value storage means, and based on the read out shading reference values of the plurality of pixels, the shading reference value of each pixel of interest is determined to be normal. A normal value range setting means for determining a value range, and a shading base for a pixel of interest. If the value is within the range of normal values set by the normal value range setting means,
It is determined that the shading reference value is a normal value, and if the shading reference value of the pixel of interest is a value outside the normal value set by the normal value range setting means, the shading reference value is an abnormal value. An abnormal reference value judging means for judging, and a constituent pixel of a processed pixel block including a processed pixel having a predetermined positional relationship in the vicinity of the pixel of interest and having undergone the judgment processing by the abnormal reference value judging means. A first block average value calculating means for calculating an average value of the shading reference value; and a non-processed pixel including an unprocessed pixel which is in a predetermined positional relationship near the pixel of interest and has not been processed by the abnormal reference value determining means. A second block average value calculating means for calculating an average value of shading reference values of constituent pixels of the processing pixel block; and the first and second block average value calculations A step determining means for determining whether the pixel of interest is a pixel corresponding to a step portion of the shading reference value, based on the average value obtained by each step, wherein the normal value range setting means is provided by the step determining means If it is determined that the pixel of interest is a pixel of the stepped portion, means for changing the range of normal values based on the average value calculated by the second block average value calculation means is included. Shading correction device. 2. A correction value calculating means for obtaining a correction value corresponding to a normal value range set by the normal value range setting means; and a shading reference value determined to be an abnormal value by the abnormal reference value determining means. And a shading reference value correcting unit that corrects the correction value to the correction value obtained by the correction value calculating unit. The correction value calculating unit determines that the target pixel is a pixel of the step unit by the step determining unit. 2. The shading correction device according to claim 1, further comprising means for changing a correction value based on the average value calculated by said second block average value calculation means. 3. The first block average value calculation means performs an average value calculation using only the shading reference value determined to be a normal value by the abnormal reference value determination means. 3. The shading correction device according to claim 1, wherein the setting means determines the range of the normal value based on the average value calculated by the first block average value calculation means. 4. A shading correction method for reading an image for each pixel and correcting output data of a reading means for outputting density data of each pixel based on a shading reference value. Acquiring the density data of each pixel output from the reading unit as a shading reference value when reading the image data; and sequentially setting each pixel as a target pixel along a predetermined processing direction, and setting a predetermined pixel in the vicinity of the target pixel. A normal value range setting step of determining a normal value range of a shading reference value for each pixel of interest based on the obtained shading reference values of a plurality of pixels having a positional relationship of: If the value is within the range of the normal value, the shading reference value is determined to be a normal value, and If the shading reference value of the pixel is outside the normal value range set in the normal value range setting step, an abnormal reference value determining step of determining that the shading reference value is an abnormal value; A first block average value calculation for obtaining an average value of shading reference values of constituent pixels of a processed pixel block including a processed pixel having a predetermined positional relationship and having undergone the determination processing in the abnormal reference value determination step The average value of the shading reference values of the constituent pixels of the unprocessed pixel block including the unprocessed pixels that have a predetermined positional relationship in the vicinity of the target pixel and that have not been processed by the abnormal reference value determination step are calculated. The second block average value calculating step, and the first and second block average value calculating steps. A step of determining whether the pixel of interest is a pixel corresponding to a step portion of the shading reference value, based on the average value obtained, the normal value range setting step includes: A shading correction method characterized by including a step of changing a range of a normal value based on the average value calculated in the second block average value calculation step if the pixel is determined to be a pixel in the step portion. 5. A correction value calculating step for obtaining a correction value corresponding to a normal value range set by the normal value range setting step; and a shading reference value determined to be an abnormal value by the abnormal reference value determining step. And a shading reference value correction step of correcting the correction value to the correction value obtained by the correction value calculation step. In the correction value calculation step, the target pixel is determined to be a pixel of the step portion by the step determination step. The method according to claim 4, further comprising the step of changing a correction value based on the average value calculated in the second block average value calculation step. 6. The first block average value calculating step,
A step of performing an average value calculation using only the shading reference value determined to be a normal value by the abnormal reference value determination step, wherein the normal value range setting step is calculated by the first block average value calculation step. 6. The shading correction method according to claim 4, further comprising the step of determining the range of the normal value based on the average value obtained.