JP2007267079A - Shading data inspection method and image reader - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a shading data inspection method capable of detecting lines with a similar margin, even near the center region and the end parts of a test chart. <P>SOLUTION: The test chart, in the middle of which a black belt with a prescribed density is printed, is scanned and thereafter, when inspection software captures scanned data, the black belt with a height in 45 dots in the middle of the scanned image is extracted, a value j is set to 0; and thereafter, the standard deviations σ of Line average X(j) of the j-th column, and a matrix average Y(j) are calculated (steps 101 to 104). Then whether the relation Y(j) less (a permissible error<X(j)<Y(j)) plus (allowed error) holds is discriminated far for X(j); and when the X(j) is not within the permissible range, the value j as coordinates whereat a color line or a white stripe is detected is stored, and the processing above is executed for all the columns (steps 105 to 108). Then the relation is discriminated up to the final column, the value j is stored, the value j is converted into coordinates and displayed (steps 109 to 111). <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a shading data inspection method for an image reading apparatus such as a digital multifunction peripheral, a facsimile machine, or a copier, that is, a shading data inspection method for inspecting an abnormality of shading data when a white reference plate is read.

  Recent digital multi-function peripherals having a copying function, a facsimile function, a printing function, a scanner function, etc., facsimile apparatuses, copiers, and other image reading apparatuses are CCD image sensors and contact sensors (CIS) that read digital image data on original paper. It has. In such a sensor, the luminance and density of the read image data are not uniform due to uneven illumination, sensitivity variations between the light receiving elements, or lens characteristics, even though a document image with uniform brightness and density is read. May be. This phenomenon is called shading and causes a loss of fidelity of a reproduced image formed on a copy sheet.

  For this reason, in the conventional image reading apparatus, density data for one line of black reference (image data of a white reference plate read when the light source is turned off) and white reference (image data of a white reference plate read when the light source is turned on). (Hereinafter referred to as “shading data”), an error between the shading data and the density data that should be obtained is obtained for each pixel, and based on this, digital image data obtained in the subsequent image reading process is corrected. As a result, the reproducibility of the reproduced image is improved.

Although there are few abnormalities in the black reference data, the white reference data shows a streak in the read shading waveform if there are scratches, uneven density, or foreign matter such as dust or water droplets on the surface of the white reference plate. Therefore, when shading correction is performed based on such a white reference plate, there is a problem in that the scanned original image has white streaks in the sub-scanning direction and the image quality deteriorates.
In recent image processing apparatuses such as digital multi-function peripherals and copiers, the number of devices having a color scanner function is increasing. Thus, the multi-value scan image of the image reading apparatus equipped with the color scanner function is as follows. Unlike black-and-white copying, even if the shading data is a little strange, it can easily be affected by a visible line on the image.

For this reason, the noise component superimposed on the shading data read from the reference white reference plate is removed using MTF (spatial frequency) correction or a low-pass filter, and shading correction is performed using the shading data from which noise is removed. (For example, refer to Patent Document 1).
JP-A-8-307664

  On the other hand, as an approach for detecting abnormalities in shading data, after acquiring shading data, a test chart having a predetermined density band is read, the read image data is displayed on a display, and a line in the predetermined density band is visually inspected. When there is an abnormality, the white reference plate and the optical system from the white reference plate to the CCD are cleaned.

As described above, in the conventional image reading apparatus, the noise component superimposed on the shading data obtained by reading the reference white reference plate is removed using MTF (spatial frequency) correction or a low-pass filter. If the image cannot be completely removed, there is a problem in that a white line appears in the black image.
Also, when the test chart is read and the read image data is projected on the display and inspected by visual inspection, there are oversights and individual differences, and there are places where reliability is lacking, and the display has high resolution and accurate color correction. In addition, there is a problem that the expensive and expensive ones are necessary and costly, and that the man-hours required for the inspection increase when the original is large or the resolution is high.

  For this reason, the line detection function is implemented in the inspection software, an image having a uniform density (band) in the horizontal direction is scanned, and whether or not there is a line in the band is also examined. However, the density of the band is ideally constant, but the shading data of the CCD sensor is generally bowl-shaped as shown in FIG. Because of the difference, the margin detected as a line differs between the center and the end, and if a fixed threshold is used as shown in FIG. 14, the problem (b) that should be determined to be dirty is not determined at the center. Arise.

  The present invention has been made in view of the above problems, and in a shading data inspection method for reading a band of a predetermined density on a test chart and detecting a line from the read band, the vicinity of the center or at the end portion is used. It is an object of the present invention to provide a shading data inspection method and an image reading apparatus capable of detecting a line with a similar margin.

  In order to achieve the above object, a shading data inspection method according to the first aspect of the present invention is a shading data inspection method that reads a band of a predetermined density on a test chart and detects a line from an image of the read band. A first average value of pixels belonging to the pixel column and a second average value of pixels belonging to a plurality of columns near the target pixel column are calculated, and the first average value is within an allowable range from the second average value. If not, the shading data is determined to be abnormal.

  The shading data inspection method of the invention according to claim 2 is the shading data inspection method for reading a predetermined density band on a test chart and detecting a line from the read band. And the standard deviation value is calculated from the average value of each column, and the first average value of the pixels belonging to the target pixel column is calculated from the second average value of the pixels belonging to a plurality of columns in the vicinity of the target pixel column. When the deviation is not within a predetermined multiple range, the shading data is determined to be abnormal.

  Furthermore, a shading data inspection method according to a third aspect of the present invention is the shading data inspection method according to the first or second aspect, wherein the band having the predetermined density has a predetermined width extending in the main scanning direction. It is characterized by.

  According to a fourth aspect of the present invention, there is provided an image reading apparatus comprising: a scanner that reads a document; and a shading correction unit that performs shading correction based on shading data obtained when the white reference plate is read by the scanner. The first average value calculating means for calculating the first average value of the pixels in the target pixel column when reading the band of the predetermined density on the test chart by the scanner, and the pixels belonging to the plurality of columns in the vicinity of the target pixel column A second average value calculating means for calculating a second average value, and comparing the first average value with the second average value to determine whether the first average value is within a normal range; And a determination means for performing the above.

  Further, an image reading apparatus according to a fifth aspect of the present invention is an image reading apparatus comprising: a scanner that reads a document; and a shading correction unit that performs shading correction based on shading data obtained when the white reference plate is read by the scanner. When the scanner reads a predetermined density band of the test chart, a first average value calculating means for calculating a first average value of pixels in each column, and calculating a standard deviation σ from the average value of each column Standard deviation calculating means, second average value calculating means for calculating a second average value of pixels belonging to a plurality of columns in the vicinity of the target pixel column, and a first average value of the target pixel column being the second average And a judging means for judging whether or not to fall within the range of the value ± ασ (α: constant).

  According to the shading data inspection method of the invention of claim 1 and the image reading apparatus of the invention of claim 4, the average density of the column of interest and the plurality of columns in the vicinity thereof are compared, so that the vicinity of the center However, since it is possible to automatically detect a line in a band having a predetermined density at the end portion with a similar margin, it is possible to improve the reliability of detecting an abnormality in shading data.

  On the other hand, when checking white lines in scanned images with inspection software instead of human visual inspection, it is necessary to detect only lines that are regarded as defective without picking up noise components such as random noise. There are places due to individual differences, and since it is completely different when the model changes, the standard (allowable value) to be NG varies depending on the model, so the allowable value becomes sweeter or stricter depending on the model. However, according to the shading data inspection method of the invention according to claim 2 and the image reading apparatus of the invention of claim 5, the shading data is generated using the standard deviation. In order to determine whether or not is abnormal, it should be determined appropriately without being affected by variations between machines and the degree of noise. It can be.

  Hereinafter, an embodiment in which the image reading apparatus of the present invention is applied to a digital multi-function peripheral will be described. FIG. 1 is a diagram illustrating a network configuration example of a system including a digital multifunction peripheral, and FIG. 2 is a block diagram illustrating a hardware configuration of the digital multifunction peripheral.

  In the network configuration diagram shown in FIG. 1, 1 is a digital multifunction peripheral, 2, 3, 4... Are personal computers, 5 is a public switched telephone network (PSTN), 6 is a LAN (Local Area Network), and 7 is an Internet network. is there. The digital multi-function peripheral 1 has copy mode, printer mode, and fax mode functions, and also has a mail transmission function, and is connected to the PSTN 5 and the LAN 6. A plurality of personal computers 2, 3, 4.・ ・ Is connected. The LAN 6 is also connected to the Internet network 7, and the digital multi-function peripheral 1 can also send and receive mail via the Internet network 7.

  FIG. 2 is a schematic block diagram showing the configuration of the control system of the digital multi-function peripheral 1. As shown in the figure, the digital multi-function peripheral 1 includes an MPU (Micro Processing Unit) 11, a flash memory 12, and an SRAM (Static Random Access Memory). 13, memory controller 14, image memory 15, codec 16, modem 17, network control unit NCU 18, operation panel 19, reading mechanism control circuit 20, light source 21, color CCD 22, analog front end 23, read image signal processing circuit 24, print The image signal processing circuit 25, printer 26, LAN interface (I / F) 27, and the like are connected to each other via a bus 28.

  The MPU 11 controls each part of the hardware of the digital multi-function peripheral 1 via the bus 28 and executes various programs based on the program stored in the flash memory 12. The flash memory 12 is necessary for the operation of the digital multi-function peripheral. Various programs, operation messages, and shading data are stored. The SRAM 13 stores temporary data generated when the program is executed.

  The memory controller 14 controls writing and reading of an image signal to and from the image memory 15, and the image memory 15 is configured using an SDRAM or the like, and stores image data to be transmitted, received image data, or read image data, The codec 16 encodes and decodes in accordance with a predetermined protocol. In order to transmit the image data of the read original, the codec 16 is encoded by the MH, MR, or MMR system, and decodes the image data received from the outside. Also, encoding and decoding are also performed in correspondence with the TIFF method, which is an image format generally used as a file that can be attached to an e-mail.

  The modem 17 is connected to a bus 28 and has a function as a fax modem capable of facsimile communication. The modem 17 is also connected to an NCU 18 connected to the bus 28. The NCU 18 is hardware for closing and opening an analog line, and connects the modem 17 to the PSTN 5 as necessary.

  The operation panel 19 includes a display unit that displays an operation state of the digital multi-function peripheral 1 and displays operation screens of various functions, and a plurality of keys for operating the digital multi-function peripheral 1, and is shown in FIG. As described above, the display unit 31 includes an LCD display unit 31 and a number of operation keys. The LCD display unit 31 is provided with a touch panel switch, and by pressing an item part displayed on the LCD display unit 31, a corresponding item can be selected and a function can be executed. As operation keys, a numeric keypad 32, a start key 33, a reset key 34, a stop key 35, a plurality of one-touch dial keys 36, a cross key 37, a return key 38, a set key 39, a FAX switching key 40, a copy switching key 41, Various keys such as a scanner switching key 42 are provided. Note that some or all of these operation keys can be substituted by the LCD display unit 31.

  Further, the reading mechanism control circuit 20 controls a reading document placing table (not shown) such as an auto document feeder (ADF) or a flatbed scanner (FBS) at the time of reading a document, and turns on the light source 21. The color CCD 22 reads an image of a document placed on a document placement table and inputs it to an analog front end (AFE) 23. The analog front end 23 samples / holds and amplifies a signal from the color CCD 22 and performs read image signal processing. Input to the circuit 24. The read image signal processing circuit 24 generates digital image data of color components including three colors of black and white and R, G, B, and executes shading correction. The corrected image data is imaged via the memory controller 14. Stored in the memory 15.

Further, the print image signal processing circuit 25 processes the print image signal and inputs it to the printer 26. The printer 26 is a printer apparatus of an electrophotographic system or the like, and receives received data, copy original data, or an external personal computer 2, Print out the print data transmitted from 3 and 4.
The LAN interface 27 is connected to the LAN 6 and receives signals from the Internet network 7 and transmits signals and data to the LAN 6 and executes interface processing such as signal conversion and protocol conversion.

  The digital multi-function peripheral 1 has the above-described configuration. When sending a facsimile, the image data of the original is read by the color CCD 22, and the image signal from the read image signal processing circuit 24 is compressed by the codec 16 to be memory controller 14. Are stored in the image memory 15. The compressed image data is read from the image memory 15, modulated by the modem 17, and transmitted from the NCU 18 to the communication partner through the PSTN 5. At the time of facsimile reception, the received image data is demodulated by the modem 17, stored in the image memory 15 via the memory controller 14, then decoded by the codec 16, and the print image signal processing circuit 25 sends a print image to the printer 26. Printed by supplying a signal.

On the other hand, when the digital multi-function peripheral 1 is shipped or when the CCD is calibrated, the image reading of the shading plate (white reference plate) 8 is performed in order to correct the illuminance unevenness of the light source, the sensitivity variation among the elements of the CCD image sensor, or the lens characteristics. The value of the signal is stored in the flash memory 12 as shading data. At this time, using the test chart in which the band of the predetermined density is recorded, the operation for inspecting the abnormality of the stored shading data will be described below. explain.
When inspecting the shading data that has been taken in, the operator activates the personal computer connected to the LAN 6 or the personal computer directly connected to the digital multi-function peripheral 1 via the PC interface (not shown), and starts the inspection software. increase.

FIG. 4 is a block diagram showing the configuration of a personal computer 50 in which inspection software is installed. As shown in the figure, MPU 51, ROM 52, RAM 53, display unit 54, input unit 55, hard disk drive (HDD) 56, and LAN interface 57 are used. These are connected via a bus 58.
The MPU 51 controls each part of the hardware of the personal computer 50 via the bus 58 and executes various programs based on the software program stored in the ROM 52. The RAM 53 is configured by SRAM or the like, and is generated when the program is executed. Store temporary data.

  The display unit 54 is a display device such as a liquid crystal display (LCD) or a CRT display, and the input unit 55 is a keyboard for operating a personal computer. The HDD 56 is a storage device with a built-in storage medium that stores programs of various functions and stores various data. In the HDD 56, software for inspecting abnormal shading data is installed. Can do. The LAN interface 57 is connected to the LAN 6 and functions in the same manner as the LAN interface 27 described above.

  After starting the inspection software of the personal computer 50, the operator sets a test chart printed with a black belt of a predetermined density at the center as shown in FIG. When the 19 scanner switching key 42 is pressed, a scan menu selection screen shown in FIG. 6 is displayed on the LCD display unit 31. On this screen, the columns of color (1) and folder (2) are selected.

  As a result of this selection, the scanning condition input screen shown in FIG. 7 is displayed on the LCD display unit 31, so that the operator is registered with the picture quality (3), resolution 300 dpi (4), and default 01. After selecting the folder (5), the test key is scanned by the reading mechanism control circuit 20, the light source 21, and the color CCD 22 by pressing the start key 33. The scanned image data is input to the read image signal processing circuit 24 via the analog front end 23, and after shading correction is performed, the scanned image data is taken into the selected folder of the image memory 15 via the memory controller 14.

  Next, when the operator takes in the image data of the test chart stored in the selected folder of the image memory 15 of the digital multi-function peripheral 1 on the inspection screen of the display unit 54 of the personal computer 50 by drag drop, The MPU 51 automatically inspects whether there is no color line or white streak in the center of the test chart. If a color line or white streak is detected, the color line appears on the PC monitor screen. Alternatively, the coordinates where the white stripe is detected are displayed.

  That is, when the image data of the test chart stored in the image memory 15 is captured by drag drop, the MPU 51 of the personal computer 50 starts the shading data abnormality inspection program shown in the flowchart of FIG. 8, and first scans the test chart. After extracting a 45-dot-high black belt at the center of the image (step 101), j is set to 0 (step 102).

  Next, the MPU 51 calculates the line average X (j) of the jth column (step 103) and calculates the matrix average Y (j) (step 104). That is, the MPU 51 calculates the average value X (j) of the image density in the sub-scanning direction of the jth column as shown in FIG. 9 and also calculates the matrix average Y (j) from the j-8th column to the j−1th column. The average value of the image density of 8 columns × 45 lines up to the eye is calculated. When j is 0 to 8, the average from the first column to the eighth column is calculated as the matrix average Y (j).

When the matrix average Y (j) is calculated, the MPU 51 determines whether X (j) has a relationship of Y (j) −allowable error <X (j) <Y (j) + allowable error. , X (j) is determined to be within the allowable range (step 105), and if X (j) is determined not to be within the allowable range, j is set as the coordinate where the color line or the white stripe is detected. (Step 106).
As the allowable error, for example, when the gradation of the image density is 0 to 255, a value around 12 can be adopted.

  When it is determined in step 105 that X (j) is within the allowable range, or after j is stored in step 106, the MPU 51 determines whether j is the final column by determining whether j is n. (Step 107) If it is determined that it is not the last column, 1 is added to j (Step 108), and then the procedure returns to Step 103 to calculate the Line average X (j) of the jth column again.

  On the other hand, if it is determined in step 107 that it is the last column, the MPU 51 determines whether j is stored in the RAM 53 (step 109). If it is determined that j is not stored in the RAM 53, the personal computer 50 For example, “No abnormality in shading correction data” is displayed on the display unit 54 (step 110). If it is determined that j is stored in the RAM 53, the value of j stored in the RAM 53 is converted into a coordinate value and displayed on the display unit 54 (step 111).

  As a result, when there is an abnormality in the shading data, for example, when dust is attached to the 128th element of the CCD at the time of obtaining the shading data, a test is performed on the display unit 54 of the personal computer 50 as shown in FIG. Since the read image and shading data of the chart band are abnormal and the abnormal position is displayed, the operator visually checks whether there is a white line or a color line at the detected coordinate position, If there is a color line, it is possible to clean the shading data again by cleaning the mirror, lens or CCD surface in the scanner unit, cleaning the contact glass.

  As described above, since a line in a band of a predetermined concentration can be automatically detected, there is no uncertainty that the judgment criteria differ depending on an oversight or a person, and the reliability of detection of abnormal shading data is improved. In addition, since the average density of the column of interest and a plurality of columns in the vicinity thereof are compared, it is possible to detect a line with the same margin both near the center and at the end.

On the other hand, as described above, random noise in the scanned image is due to individual differences in the machine and cannot be inspected with a certain standard, so it is necessary to determine whether the shading data is abnormal using the standard deviation of the line average Thus, an abnormality can be appropriately determined without being affected by variations between machines or the degree of noise. Hereinafter, an embodiment in which the standard deviation of the line average is used as an allowable error will be described.
Since the apparatus configuration is the same as that of the first embodiment, description thereof is omitted.

When inspecting the abnormality of the captured shading data, the operator activates the personal computer 50 and starts the inspection software, and then scans the test chart and selects the image data in the image memory 15 as described above. Import to a folder.
Next, when the operator takes in the image data of the test chart stored in the image memory 15 of the digital multi-function peripheral 1 on the inspection screen of the display unit 54 of the personal computer 50 by drag-drop, the MPU 51 of the personal computer 50 is shown in FIG. The shading data abnormality inspection program shown in the flowchart is started, and first, a black belt having a height of 45 dots is extracted from the center of the scanned image of the test chart (step 201). Then, as shown in FIG. Line average X (j) (j = 0, 1,..., N-9), that is, after calculating the average value of the image density in the sub-scanning direction of each column, the standard deviation σ of each average value is calculated (Step 202)

  Next, after setting j to 0 (step 203), the MPU 51 calculates the matrix average Y (j) (step 204). That is, the MPU 51 calculates the average value of the image density of 8 columns × 45 lines from the j-8th column to the j−1th column as the matrix average Y (j), as shown in FIG. . When j is 0 to 8, the average from the first column to the eighth column is calculated as the matrix average Y (j).

When the matrix average Y (j) is calculated, the MPU 51 determines whether or not the line average X (j) of the j-th column has a relationship of Y (j) −4σ <X (j) <Y (j) + 4σ. Thus, it is determined whether or not X (j) is within the allowable range (step 205). When it is determined that X (j) is not within the allowable range, j is detected as a color line or white stripe. The coordinates are stored in the RAM 53 (step 206).
When it is determined in step 205 that X (j) is within the allowable range, or after j is stored in step 106, the MPU 51 determines whether j is the last column by determining whether j is n. (Step 207) If it is determined that it is not the last column, 1 is added to j (Step 208), and then the process returns to Step 204 to calculate the matrix average Y (j) again.

  On the other hand, if it is determined in step 207 that it is the last column, the MPU 51 determines whether j is stored in the RAM 53 (step 209). If it is determined that j is not stored in the RAM 53, the personal computer 50 The display unit 54 displays that there is no abnormality in the shading correction data (step 210). If it is determined that j is stored in the RAM 53, the value of j stored in the RAM 53 is converted into a coordinate value and displayed on the display unit 54 as described above (step 211).

  As described above, by determining whether shading data is abnormal using the standard deviation as an allowable error, it is possible to appropriately determine whether shading data is abnormal without being affected by variations between machines or the degree of noise. Can do.

  In the above embodiment, the shading correction data is checked for abnormality by using a personal computer connected to the LAN or a personal computer directly connected to the digital multifunction peripheral via the PC interface. Incorporating special equipment for testing, such as a computer, can be eliminated.

In the above embodiment, 12 is used as the allowable error, 4 is used as the multiple of the standard deviation, 45 is used as the number of lines in the sub-scanning direction, and 8 columns are used as the matrix average width. However, these values may be changed as appropriate. In addition, in the above-described embodiment, the pixel column on the left side of the target pixel column is used as the matrix in the vicinity of the target pixel column, but the left and right pixel columns of the target pixel column may be used.
In the above embodiment, the image reading apparatus according to the present invention is applied to a digital multi-function peripheral. However, the image reading apparatus according to the present invention is also applicable to other image reading apparatuses such as a facsimile machine and a copying machine. can do.

It is a figure which shows the network structural example of the system provided with the digital multifunctional device. 2 is a block diagram illustrating a hardware configuration of a digital multifunction peripheral. FIG. It is a figure which shows the detail of an operation panel. It is a block diagram which shows the structure of the personal computer which mounted inspection software. It is an example of the test chart which printed the black belt of the predetermined density in the center. It is an example of a scan menu selection screen. It is an example of a scanning condition input screen. It is a flowchart of an abnormality inspection program for shading data. It is a figure for demonstrating the calculation of Line average X (j) and matrix average Y (j). It is an example of a display when there is an abnormality in shading data. It is a flowchart of the abnormality inspection program of the shading data of another Example. It is a figure for demonstrating the calculation of Line average X (j) of each column. It is a figure for demonstrating the calculation of matrix average Y (j). It is a figure which shows the characteristic of shading data.

Explanation of symbols

1 Digital MFP 2, 3, 4 PC 5 PSTN
6 LAN
7 Internet network 8 Shading plate 11 MPU
12 Flash memory 13 SRAM
14 Memory Controller 15 Image Memory 16 Codec 17 Modem 18 NCU
19 Operation Panel 20 Reading Mechanism Control Circuit 21 Light Source 22 Color CCD
23 Analog Front End 24 Read Image Signal Processing Circuit 25 Print Image Signal Processing Circuit 26 Printer 27 LAN Interface 28 Bus

Claims (5)

In the shading data inspection method of reading a band of a predetermined density on a test chart and detecting a line from the image of the read band,
A first average value of pixels belonging to the target pixel column and a second average value of pixels belonging to a plurality of columns near the target pixel column are calculated, and the first average value is within an allowable range from the second average value. A shading data inspection method characterized by determining that shading data is abnormal when not inside. In a shading data inspection method for reading a predetermined density band on a test chart and detecting a line from the read band,
The average value of the pixels belonging to each column is calculated, the standard deviation value is calculated from the average value of each column, and the first average value of the pixels belonging to the target pixel column belongs to a plurality of columns near the target pixel column A shading data inspection method, wherein the shading data is determined to be abnormal when it is not within a range of a predetermined multiple of the standard deviation from the second average value. In the shading data inspection method according to claim 1 or 2,
The method for inspecting shading data, wherein the band having the predetermined density has a predetermined width extending in the main scanning direction. In an image reading apparatus having a scanner for reading a document, and a shading correction means for performing shading correction based on shading data when a white reference plate is read by the scanner,
A first average value calculating means for calculating a first average value of pixels in the target pixel column when reading a band of a predetermined density on the test chart with the scanner; and a first average value calculating unit for the pixels belonging to the plurality of columns in the vicinity of the target pixel column. A second average value calculating means for calculating an average value of 2, and comparing the first average value and the second average value to determine whether or not the first average value is within a normal range. An image reading apparatus comprising: a determination unit. In an image reading apparatus having a scanner for reading a document, and a shading correction means for performing shading correction based on shading data when a white reference plate is read by the scanner,
When the scanner reads a predetermined density band of the test chart, a first average value calculating means for calculating a first average value of pixels in each column, and a standard deviation σ is calculated from the average value of each column. A standard deviation calculating means; a second average value calculating means for calculating a second average value of pixels belonging to a plurality of columns in the vicinity of the target pixel column; and a first average value of the target pixel column being the second average value. An image reading apparatus comprising: a determination unit that determines whether or not a value falls within a range of ± ασ (α: constant).

Images