JP2019161461A - Image reading device and control method thereof - Google Patents

Abstract

To realize stable sensor anomaly detection that is not easily affected by dust or the like by checking shading data of a reading unit.SOLUTION: An image reading device includes a control unit that reads an image by a reading unit having an image sensor including a plurality of groups including a plurality of reading elements and corrects a signal based on an output from the reading element, and the control unit generates a characteristic value indicating a shading characteristic for each of the reading elements, obtains an average value of the characteristic values for each of the groups on the basis of the characteristic value for the reading element, and determines that the reading unit is abnormal when at least one of average values of the characteristic values is greater than a first threshold value.SELECTED DRAWING: Figure 5

Description

  The present invention relates to an image reading apparatus, and more particularly to an image reading apparatus that detects an abnormality of an image sensor before image reading and a control method thereof.

  In the image reading apparatus, the sensitivity characteristic varies from pixel to pixel depending on the offset and sensitivity variation of each pixel of the image sensor and the distribution characteristics of the light source. In order to correct this, shading correction is widely performed.

  The image sensor abnormality can be detected by checking the shading correction value here, but if the image sensor or shading plate is attached with dust, it is erroneously detected as an abnormality in the shading data even though it is not an image sensor abnormality. there is a possibility. In contact image sensors, an image sensor is configured by arranging multiple sensor chips in the main scanning direction. However, abnormalities in only a single chip or several chips may occur due to electrostatic breakdown of the sensor chip or disconnection of wire bonding. There is.

  Patent document 1 calculates the average value of the reference plate data for each block and detects abnormalities in LEDs and sensors that determine abnormalities when there are singular values. Patent document 2 reads shading plates for each CCD chip. Describes a technique for detecting an abnormality of a sensor by obtaining an average value, a minimum value, and a maximum value.

Patent No. 4894247 Specification Japanese Patent No. 3449490

  The method described in Patent Document 1 is for the case where the light source is an LED light source having a plurality of chips. To do. However, in the case of using a light guide without using a plurality of LEDs as the light source, the distribution characteristic of the light source changes continuously, so that an abnormality cannot be detected. Also, a sensor chip with a small difference in sensitivity cannot be detected as abnormal.

  In the method of Japanese Patent No. 3449490, it is determined whether the CCD is abnormal or dust at another position on the shading plate. However, if there is a large dust, dust may be deposited on either position and there is a possibility of false detection. .

  SUMMARY An advantage of some aspects of the invention is that it provides an image reading apparatus that detects an abnormality of an image sensor or illumination with high accuracy and a control method thereof.

In order to achieve the above object, the image reading apparatus of the present invention has the following configuration.
That is, according to one aspect of the present invention, an image reading apparatus that reads an image with a reading unit having an image sensor including a plurality of groups each including a plurality of reading elements,
A controller for correcting a signal based on the output from the reading element;
The controller is
Generating a characteristic value indicating a shading characteristic for each reading element;
Based on the characteristic value for each reading element, obtain an average value of the characteristic values for each group,
The image is characterized in that the reading unit is determined to be abnormal when an average value of at least one of the characteristic values of the group is greater than a first threshold value. A reader is provided.

  According to the present invention, it is possible to detect an abnormality in an image sensor or illumination with high accuracy.

FIG. 3 is a diagram when the image forming apparatus is in a standby state. 3 is a cross-sectional view of a scanner unit 3. FIG. 3 is a control configuration diagram of a scanner unit 3. FIG. It is a block diagram of a shading process It is a flowchart of a shading acquisition process. It is a flowchart of a black shading data acquisition process. It is a flowchart of a white shading data acquisition process. It is a flowchart of a white shading abnormality detection process. It is an image figure of chip averaging of a white shading correction value. It is an image figure of an abnormal state. It is an image figure of an error display screen. It is an image figure of abnormality determination of Embodiment 2. 10 is a flowchart of white shading abnormality detection processing according to the second embodiment. 4 is a configuration of an image sensor according to a third embodiment. It is an image figure of abnormality determination of Embodiment 3. 10 is a flowchart of white shading abnormality detection processing according to the third embodiment. 10 is a flowchart of white shading abnormality factor determination according to the third embodiment.

  Embodiments of the present invention will be described below with reference to the drawings.

[Embodiment 1]
(1) Description of Image Forming Apparatus FIG. 1 is an internal configuration diagram of an ink jet image forming apparatus 1 (hereinafter, image forming apparatus 1) used in the present embodiment. In the figure, the x direction indicates the horizontal direction, the y direction (perpendicular to the paper surface) indicates the direction in which the ejection openings are arranged in the recording head 8 described later, and the z direction indicates the vertical direction. In this embodiment, a shading correction value is generated as a characteristic value indicating a shading characteristic for each reading element of the image sensor, and an abnormality of the image sensor or illumination (for example, LED) is detected based on the generated shading correction value.

  The image forming apparatus 1 is a multifunction machine including a printing unit 2 and a scanner unit 3, and can perform various processes related to recording operations and reading operations individually or in conjunction with the printing unit 2 and the scanner unit 3. . The scanner unit 3 includes an ADF (automatic document feeder) and an FBS (flatbed scanner), and reads a document automatically fed by the ADF and reads a document placed on a document table of the FBS by a user (scanning). )It can be performed. Although the present embodiment is a multi-function peripheral having both the print unit 2 and the scanner unit 3, a form without the print unit 2 may be used. FIG. 1 shows a state in which the image forming apparatus 1 is in a standby state in which neither a recording operation nor a reading operation is performed.

  In the print unit 2, a first cassette 5 </ b> A and a second cassette 5 </ b> B for accommodating a recording medium (cut sheet) S are detachably installed on the bottom of the casing 4 in the vertical direction. A relatively small recording medium up to A4 size is accommodated in the first cassette 5A, and a relatively large recording medium up to A3 size is accommodated in the second cassette 5B. Near the first cassette 5A, a first feeding unit 6A for separating and feeding the stored recording media one by one is provided. Similarly, a second feeding unit 6B is provided in the vicinity of the second cassette 5B. When a recording operation is performed, the recording medium S is selectively fed from one of the cassettes.

  The transport roller 7, the discharge roller 12, the pinch roller 7a, the spur 7b, the guide 18, the inner guide 19 and the flapper 11 are transport mechanisms for guiding the recording medium S in a predetermined direction. The conveyance roller 7 is a driving roller that is arranged on the upstream side and the downstream side of the recording head 8 and is driven by a conveyance motor (not shown). The pinch roller 7 a is a driven roller that rotates while nipping the recording medium S together with the conveying roller 7. The discharge roller 12 is a drive roller that is disposed on the downstream side of the transport roller 7 and is driven by a transport motor (not shown). The spur 7 b sandwiches and transports the recording medium S together with the transport roller 7 and the discharge roller 12 disposed on the downstream side of the recording head 8.

  The guide 18 is provided in the conveyance path of the recording medium S and guides the recording medium S in a predetermined direction. The inner guide 19 has a side surface curved by a member extending in the y direction, and guides the recording medium S along the side surface. The flapper 11 is a member for switching the direction in which the recording medium S is conveyed during the double-side recording operation. The discharge tray 13 is a tray for stacking and holding the recording medium S that has completed the recording operation and is discharged by the discharge roller 12.

  The recording head 8 of the present embodiment is a full-line type color inkjet recording head, and has a plurality of ejection openings corresponding to the width of the recording medium S along the y direction in FIG. It is arranged. When the recording head 8 is in the standby position, the ejection port surface 8a of the recording head 8 is capped by the cap unit 10 as shown in FIG. When performing the recording operation, the orientation of the recording head 8 is changed by the print controller so that the discharge port surface 8 a faces the platen 9. The platen 9 is constituted by a flat plate extending in the y direction, and supports the recording medium S on which the recording operation is performed by the recording head 8 from the back side.

  The ink tank unit 14 stores the four colors of ink supplied to the recording head 8. The ink supply unit 15 is provided in the middle of the flow path connecting the ink tank unit 14 and the recording head 8, and adjusts the pressure and flow rate of the ink in the recording head 8 to an appropriate range. In this embodiment, a circulation type ink supply system is employed, and the ink supply unit 15 adjusts the pressure of the ink supplied to the recording head 8 and the flow rate of the ink recovered from the recording head 8 to an appropriate range.

  The maintenance unit 16 includes a cap unit 10 and a wiping unit 17 and operates them at a predetermined timing to perform a maintenance operation on the recording head 8.

  FIG. 2 is a configuration diagram of the scanner unit 3. As shown in FIG. 2, when performing simultaneous double-sided reading, a document to be read is conveyed in the direction of arrow A. The document is conveyed in the direction of arrow A by the conveyance roller 220, the conveyance roller 221, the conveyance roller 222, the conveyance roller 223, the conveyance roller 224, and the conveyance roller 225. The conveyed document reaches the reading position of the contact image sensor for front surface 201 which is a device for reading the image recorded on the surface. The document that has reached the reading position on the surface is irradiated with irradiation light from an LED, which is a light source for illuminating the document, provided inside the surface contact image sensor 201 through a light guide. When the conveyance is continued, the document reaches the reading position of the contact image sensor 202 for the back surface which is a device for reading the image recorded on the back surface. The document that has reached the position on the back side of the reading is irradiated with light emitted from the light guide from an LED that is a light source for illuminating the document provided inside the contact image sensor 202 for the back side. These contact image sensors have a reading width corresponding to the width of the document. The direction of the reading width is a direction perpendicular to the arrow A. This direction is called the main scanning direction, and the document transport direction indicated by the arrow A is also called the sub-scanning direction. Therefore, by arranging the two contact image sensors so as to face each other and conveying the document in the direction of arrow A, it is possible to simultaneously read images on both sides of the document. Alternatively, a reading unit including a contact image sensor may be conveyed with respect to the document. Thus, the document image is read line-sequentially by the CIS that forms the line sensor. Each contact image sensor includes a blue LED, a red LED, and a green LED for reading a color image.

  In addition, a shading correction sheet 210 (for the front surface) is provided to correct the sensitivity of each pixel of the contact image sensors 201 and 202 (a shading correction sheet for the back surface is not shown in FIG. 2). The contact image sensor is moved onto the shading correction sheet before reading the document. Although details will be described later, black shading data can be acquired with the LED of the light source turned off, and white shading data can be acquired with the LED turned on. Since the shading correction sheet 210 may be a member for shading correction, the shading correction sheet 210 is not limited to a member such as a sheet, and may be a plate member called a so-called shading plate.

  Reflected light on the original of each color light irradiated by each color LED provided on the front surface contact image sensor 201 forms an image on an image sensor disposed on the substrate of the contact image sensor through a lens. The reflected light imaged by the contact image sensor 201 is photoelectrically converted and analog image data (analog signal) is output. An A / D converter (not shown) is mounted on the contact image sensor substrate. The analog signal is converted into a digital signal by this A / D converter and output as digital image data. On the other hand, the reflected light of the light emitted from each color LED provided on the contact image sensor 202 for the back surface is imaged on the image sensor arranged on the substrate of the contact image sensor 202 through the lens. In the contact image sensor 202, the formed reflected light is photoelectrically converted into analog image data (analog signal). The analog signal is converted into a digital signal by an A / D converter (not shown) mounted on the contact image sensor substrate, and is output as digital image data. In the present embodiment, the contact image sensors 201 and 202 are provided with a plurality of sensor chips, and in this example, 17 image sensor chips are mounted.

  In this example, each image sensor chip has a resolution of 600 dpi and a reading element having an opening width of 42 μm per chip is arranged in 432 pixels in the main scanning direction. That is, 432 elements are continuously arranged per chip. By arranging this image sensor chip in the 17-chip main scanning direction, it is possible to read 308 mm of A3 width or more with 600 dpi resolution. In addition, the reading element of this example has a CMOS configuration, but can be configured similarly in a CCD configuration.

  The image processing block 300 in FIG. 3 includes front side and back side applications. A digital signal of the surface image from the contact image sensor 201 is input to the surface image processing block 301. Similarly, a digital signal of the back side image is input from the contact sensor 202 to the back side image processing block 302. The circuit in FIG. 3 may be driven and controlled by a sequencer that generates various signals such as an enable signal for each block at a predetermined timing by first applying a start signal. It is assumed that it is controlled according to a program by a processor (not shown) that controls In this case, for example, the processor outputs a control signal for each circuit block by executing a predetermined program. Control targets include not only each block in FIG. 3 but also an optical system including an LED, an image sensor, a document transport system, and the like.

  The image processing blocks 301 and 302 have the same circuit configuration and will be described in detail with reference to FIG. The signals processed in the image processing blocks 301 and 302 are input to the printer control interface block 303 and output to the printer control unit. The internal configuration of the front surface image processing block 301 will be briefly described. AFE IF 321 receives a signal from the contact image sensor. The dot-sequential line-sequential conversion unit 322 rearranges the data of one line and three colors in the order of RGB in the order of R / G / B pixels. At this time, it becomes a data format in which one pixel is constituted by R / G / B as in the case of a CCD machine. An SHD (shading processing unit) 323 performs shading processing for correcting pixel offset and sensitivity. The linearity of the image sensor 201 is corrected by the LUT 324, and the output is proportional to the luminance by the table. The image cutout unit 325 cuts out the valid signal portion of the data and masks the portion outside the range with white.

  FIG. 4 shows a block diagram of the SHD block. This figure corresponds to SHD 323 and SHD 333 in FIG. The SHD block includes a black SHD correction unit (black shading correction unit) 411, BSHD_RAM 412, white SHD correction unit (white shading correction unit) 421, WSHD_RAM 422, line integrated value calculation unit 431, WORK_RAM 432, white SHD correction value calculation unit 441, and SHDconst register. 442, a black SHD correction value calculation unit 451, a BKoffset register 452, a selector 461, a maximum / minimum integrated value calculation unit 471, a MAX register 481, a MIN register 482, and an integration register 483.

  First, acquisition of black shading data will be described. The black shading data is acquired in a state where the black SHD correction unit 411 is set to “through” and the white SHD correction unit 421 is set to “through” and the LED is turned off. The line integration value calculation unit 431 acquires R / G / B pixel data, and the WORK_RAM 432 integrates a plurality of lines. For example, the values of corresponding pixels in a plurality of lines are integrated for each RGB. The data acquisition is stopped after integrating the predetermined number of lines, and the black SHD correction value calculation unit 451 reads the data (integrated value) of the WORK_RAM 432 for each pixel, and obtains the number of acquired lines (that is, the integrated number of lines). Divide to find the average value. The value set in the BkOffset register 452 is subtracted from the average value. When the result is a positive number, it is clamped to that value, and when it is a negative number, it is clamped to 0 and the black SHD correction value of each pixel is calculated. The black SHD correction value is written in the BSHD_RAM 412. Black shading acquisition ends when all pixel calculations are completed. That is, a black shading correction value is obtained for pixels for one line (in this example, 432 × 17 = 7344 pixels).

  Next, acquisition of white shading data will be described. The white shading data acquisition is performed with the black SHD correction unit 411 set to correction ON, the white SHD correction unit 412 set to through, and the LED lit. The line integration value calculation unit 431 acquires data of each pixel of R / G / B and integrates a plurality of lines in the WORK_RAM 432. The data acquisition is stopped after accumulating a predetermined number of lines, and the white SHD correction calculation unit 441 reads the data in the WORK_RAM 432 for each pixel and divides the data by the number of acquired lines to obtain an average value. The value set in the SHDconst register 442 is divided by the obtained average value of each pixel. Calculate the white SHD correction value for each pixel (each pixel in one line) by clamping it with the maximum value that can be stored if it is larger than the value that can be stored in WSHD_RAM. To do. The white SHD correction value is written in the WSHD_RAM 422. The white shading acquisition ends when all pixel calculations are completed. When white shading data is acquired, the object to be read by the image sensor may be, for example, a white shading plate.

  Next, a description will be given of executing SHD correction. In this case, the black SHD correction unit 411 is turned on, the white SHD correction unit 412 is turned on, and the line integration unit 431 is set to through, and the image data subjected to black SHD correction and white SHD correction is output to the subsequent stage. The black SHD correction unit 411 reads the black SHD correction value of each pixel from the BSHD_RAM 412 and subtracts it from the image data. When it becomes a negative number, it clamps at 0 and outputs image data. The white SHD correction unit 421 reads the white SHD correction value of each pixel from the WSHD_RAM 422 and multiplies the black SHD corrected image data with the white SHD data. When a certain maximum value (for example, the maximum value of each color component) is exceeded, the image data is output after being clamped at the maximum value.

  The operation of the maximum / minimum / integrated value calculation unit 471 will be described. The RAM for which the maximum value / minimum value / integrated value is to be obtained is selected by the selector 461. Among the R / G / B components of BSHD_RAM 412, WSHD_RAM 422, and WORK_RAM 432 that can be selected, a total of nine RAMs can be selected. The maximum / minimum / integrated value calculation unit 471 reads the RAM data selected by the selector 461, divides the data by a predetermined number of pixels, and calculates the maximum value / minimum value / integrated value within the determined number. In the case of a contact image sensor, for example, the number of pixels for one sensor chip is set as the determined number of pixels. The calculated maximum value is written to the maximum value register 481, the minimum value is written to the minimum value register 482, and the integrated value is written to the integrated value register 483. The maximum value register 481, the minimum value register 482, and the integrated value register 483 exist more than the number of chips of the image sensor. In this embodiment, the image sensor has 17 chips and 18 registers.

  Finally, the SHD control unit 491 that controls the entire SHD block will be briefly described. This controls the entire black shading acquisition, white shading acquisition, SHD correction operation, and maximum / minimum / integrated value calculation processing. The trigger of each operation is received, the operation of the block starts and ends, and when the entire processing is completed, an operation for generating an interrupt is performed. In the above description, signals such as on / off switching of the white SHD correction unit and the black SHD correction unit, a selector selection signal, and the like may be given by the above-described processor (not shown).

  FIG. 5 shows a shading data acquisition flow in the scanner unit 3. The procedure in FIG. 5 may be executed by, for example, the above-described processor (not shown) that controls the scanner unit 3, but may be a control procedure by hardware such as a sequencer. The procedure shown in FIG. 5 may be executed, for example, in an initialization process after the image reading apparatus is turned on, or shading data may be updated according to an explicit instruction from a user interface, for example.

  When shading data acquisition is started (S100), the LED is turned off (S101), and black shading data is acquired (S102). The black shading data acquisition flow is shown in FIG. First, the black shading correction unit 411 is set to through, the white shading correction unit 421 is set to through, and the black shading acquisition start trigger is turned on (S202). Then, the black shading data for a plurality of lines is read by the reading unit (S203). The read data is added for a plurality of lines using WORK_RAM (432) for each pixel (S204), and is divided by the number of lines and averaged (S205). The offset is subtracted from the average value (S206) and stored as a black shading correction value in the BSHD_RAM 412 (S207). From the data reading (S203) to the correction value storage (S207), the color components R / G / B are processed at the same time, and upon completion, an interrupt is returned (S208). When the black shading acquisition is completed, it is determined whether or not the acquired black shading correction value is within the correction range, thereby determining black shading abnormality (S103). When the correction value is out of the correction range, it is determined that there is an abnormality and the process proceeds to an abnormality process. The value of the correction range may be determined in advance.

  If the black shading correction value is normal, the LED is lit to obtain white shading (S104). In this embodiment, in order to carry out moving shading, movement of the reading units 201 and 202 is started (S105), and white shading acquisition is started (S106). The white shading acquisition flow is shown in FIG. First, the black shading correction is set to ON, the white shading correction is set to through, the white shading data acquisition start trigger is set to ON (S303), and the reading units 201 and 202 read data for a plurality of lines while moving (S304). For each pixel, the read data is added for a plurality of lines using the WORK_RAM 432 (S305), and is divided by the number of lines and averaged (S306). Thereafter, the value of SHDconst is divided by the average value of each pixel (S307), and the obtained result is stored in the WSHD_RAM 432 as a white shading correction value (S308). From the data reading (S304) to the correction value storage (S308), the color components R / G / B are simultaneously processed, and upon completion, an interrupt is returned (S309).

  When the white shading acquisition is completed, white shading abnormality determination is performed using the acquired white shading correction value (S107). The white shading abnormality determination flow is shown in FIG. The R white shading correction value stored in the WSHD_RAM 422 is read into the WORK_RAM 432, and the chip average value is calculated (S401). The method for calculating the chip average value is shown in FIG. The circuit of FIG. 9 is included in the maximum value / minimum value / integrated value calculation unit 471 in the SHD block diagram of FIG.

  In the R component white shading correction value read out from the WSHD_RAM 432, the correction value is obtained in units of pixels, and is divided into units of chips. That is, with the elements included in the chip as one group, the maximum value, the minimum value, and the integrated value are obtained for each chip, that is, the group, and stored in the registers Max_REGs 481, Min_REGs 482, and Add_REGs 483, respectively. The chip average value WSHD_AVE_R “n” is calculated by dividing the integrated value stored in Add_REGs 483 by the number of pixels per chip. This is repeated for the G / B component, and the chip average of the white shading correction value of each color component is obtained. In addition, the calculation method of a chip | tip average value may use not only an arithmetic mean but calculation methods, such as a weighted average, a geometric mean, and a harmonic average. That is, the chip average value of the present invention may be a value indicating an average within one chip, and is not limited to the arithmetic average value. In the example of FIG. 9, the maximum value, the minimum value, and the integrated value are performed by hardware until they are stored in each register, and the process of reading the value from the register and calculating the average value is executed by, for example, the processor described above.

  Of the chip averages calculated in S401, the maximum value WSHD_Max_R / WSHD_Max_G / WSHD_Max_B and the minimum value WSHD_Min_R / WSHD_Min_G / WSHD_Min_B for each color component are calculated (or determined). Also, the maximum value WSHD_Max is calculated from the chip average values of all color components (S402). When the maximum value WSHD_Max is larger than a predetermined threshold Max_TH (Yes in S403), it is determined that white shading is abnormal (S404), and an error screen indicating white shading abnormality as shown in FIG. 11 is displayed on the operation panel 104. The operation of the image forming apparatus 1 is stopped. When the maximum value WSHD_Max is equal to or less than Max_TH (No in S403), the process proceeds to S405, and the uniformity index WSHD_Up_R / WSHD_Up_G / WSHD_Up_B of the white shading correction value is obtained for each color component. Here, the uniformity index Up is expressed by the following equation. WSHD_Up_R = (WSHD_Max_R−WSHD_Min_R) / (WSHD_Max_R + WSHD_Min_R). The higher the uniformity, the smaller the difference between WSHD_Max_R and WSHD_Min_R, so the value of the uniformity index Up becomes smaller. That is, the greater the uniformity index Up, the greater the variation. When at least one of the uniformity index WSHD_Up_R / WSHD_Up_G / WSHD_Up_B of each color component is larger than a predetermined threshold Up_TH (Yes in S406), it is determined that white shading is abnormal (S407), and white shading abnormality as shown in FIG. Is displayed on the operation panel 104, and the operation of the image forming apparatus 1 is stopped. If all the uniformity indexes WSHD_Up_R / WSHD_Up_G / WSHD_Up_B of each color component are equal to or less than the threshold value Up_TH (No in S406), the process proceeds to S408, and the white shading abnormality detection is terminated. When the white shading abnormality detection S107 is completed, the reading unit returns to the home position and stops (S108), and the shading acquisition process ends (S109). As the uniformity index, for example, the variance or standard deviation of the average value of the white shading data correction value for each chip may be used (* Since the shading data itself is used as the characteristic value, it is considered to be the embodiment 2). , “Correction value” has been added, but please contact us if it is different from the intention of this sentence.Please confirm that the word “Correction value” has also been added in the following description. I will do it.) If these values are smaller than a predetermined threshold value, it can be reversed if the uniformity is high. On the error screen, a message indicating an error and a code indicating the type of error are displayed.

  Here, the operation of the portion for calculating the average value of the white shading correction value for each chip is shown again in FIG. This process corresponds to step S401 in FIG. 8, and is performed by the maximum / minimum integrated value calculation unit 471 in the SHD block diagram of FIG. First, the RAM to be detected is selected by the selector 461 in FIG. The WSHD_RAM 432 is selected to calculate the average value of the white shading correction value. In addition, since the calculation of the average value is a specification executed for each component, one of R / G / B is selected and executed. After selecting the target RAM, a trigger for calculating the maximum value / minimum value / integrated value is issued to the SHD control 491 in FIG. The issuance of the trigger may be performed by a processor that controls the scanner unit 3.

  The maximum / minimum integrated value calculation unit 471 reads the WSHD_RAM 422 pixel by pixel and calculates the maximum value / minimum value / integrated value. When the amount corresponding to the number of pixels for one sensor chip is read, the maximum value result is written to MAX_REGs 481, the minimum value result is written to MIN_REGs 482, and the integrated value result is written to ADD_REGs 483. When the calculation for one chip is completed, the temporary register for the maximum value, minimum value, and integrated value is reset, and the next chip is processed. Since MAX_REGs 481, MIN_REGs 482, and ADD_REGs 483 are prepared for the number of chips, the maximum value / minimum value / integrated value of all the chips can be obtained. When the calculation for all the chips is completed, the SHD control unit 491 issues an interrupt, and the maximum value / minimum value integrated value calculation ends.

  When the calculation of the maximum value / minimum value / integrated value for each chip is completed, the result of ADD_REGs (483) is read by the FW process, and is divided by the number of pixels of one chip to calculate the average value of each chip. This is repeated with R / G / B and three components, and the average value of white shading correction values for each chip for three colors is obtained.

  FIG. 10 shows an example of the average value of the white shading correction values for each chip obtained by the processing up to step S402 in FIG. As described above, the sensors 201 and 202 are each composed of 17 sensor chips from chip 0 to chip 16. A value 1001 indicates an average value of white shading correction values of the sensor chips. A predetermined threshold MAX_TH1002 is also shown. In FIG. 10, the average value of the chip 6 is the maximum value of the entire chip, and the value exceeds the threshold value MAX_TH. Therefore, in the case of FIG. 10, it is determined to be abnormal in step S403 of FIG. In this example, it is determined whether the maximum value among the average values of the shading correction values of each chip exceeds a predetermined threshold, but not limited to this example, the average value of the shading correction values of each chip. What is necessary is just to determine whether any is over the threshold value. Further, the average value of the shading correction values of the chip 5 is the minimum value of the average value in the entire chip. Therefore, in the example of FIG. 10, the uniformity index WSHD_Up is calculated with the average value of the shading correction value of the chip 5 as the minimum value WSHD_Min and the average value of the shading correction value of the chip 6 as the maximum value WSHD_Max. Even when the uniformity index WSHD_Up exceeds the threshold value, it is determined to be abnormal.

  With the above configuration, an abnormality in shading, particularly white shading, is determined not by each pixel but by the average value of the chip. For this reason, dust attached to a specific image sensor element is not judged to be abnormal, and abnormalities are caused by multiple image sensor elements (particularly a series of multiple elements) such as sensor chip failures or LED lighting failures. Detect. In other words, by filtering the high-frequency component of the shading correction value and making a determination in units of chips, it is possible to suppress the sensitivity to abnormalities in units of pixels and improve the accuracy of abnormality detection.

[Embodiment 2]
In the white shading abnormality detection flowchart of FIG. 8 described in the first embodiment, the abnormality determination is performed using the white shading correction value obtained by the white shading data acquisition process (S106) (S401). Performs the abnormality determination using the acquired white shading data itself. That is, in this embodiment, shading data itself is used as a characteristic value indicating the shading characteristic for each reading element of the image sensor, and an abnormality in the image sensor or illumination (for example, LED) is detected based on the shading data.

  FIG. 12 shows an image diagram of abnormality detection according to the second embodiment. In this embodiment, since the abnormality determination is performed without calculating the white shading correction value and the black shading correction value, the white shading data is output with the offset level of the image sensor added. That is, the data is not subjected to black SHD correction. Therefore, it is necessary to obtain a mean value of the black shading data for each chip and subtract it from the white shading data. Originally, black SHD correction is performed in units of pixels, but since black SHD correction cannot be performed, black shading data that is simply averaged in units of chips is handled as a black SHD correction value.

A maximum value (WSHD_Max) and a minimum value (WSHD_Min) are obtained by subtracting the average value of black shading data from the average value of white shading data for each chip. In Embodiment 2, since the minimum value is more abnormal, when WSHD_Min falls below a predetermined threshold value (MIN_TH), it is determined as abnormal and error processing is performed. In addition, the calculation of uniformity is the same as in Embodiment 1,
Up = (WSHD_Max−WSHD_Min) / (WSHD_Max + WSHD_Min)
Is calculated by This is because the white shading data and the white shading correction value have a reciprocal relationship, but the uniformity index Up is calculated by the same formula.
As in the first embodiment, when the uniformity index Up exceeds a predetermined threshold, it is determined as abnormal and error processing is performed.

  FIG. 13 shows a flowchart of the second embodiment. Since it is almost the same as the flow of FIG. 1 (FIG. 8), detailed description is omitted. However, the difference is that the black level value is integrated in units of chips before the chip average acquisition of the white shading correction value (S401). Black level chip average acquisition (S1301) that averages the value per pixel is added. Further, instead of obtaining the average chip value of the white shading correction value (S401), the white level chip average acquisition (S1302) is performed in which the white level values are integrated in units of chips and averaged to the value per pixel. Although not shown in the flow, in the chip average and MAX / MIN calculation (S1303), the black level chip average value acquired in S1301 is subtracted from the white level chip average value acquired in S1302, and the maximum value / minimum is obtained. Calculate the value. The last difference is that the minimum value calculated in S1304 is compared with the threshold value MIN_TH, and if it is equal to or less than the threshold value, error processing is performed as an abnormality.

  Since the white shading correction value and the black shading correction value are not calculated in the procedure of FIG. 13, for example, pixel values that have passed through the correction units 411, 421, and 431 of FIG. The value calculation unit 471 calculates the maximum value, the minimum value, and the average value for each chip. If no abnormality is detected, the black shading correction value and the white shading correction value may be calculated again and stored in BSHD_RAM 412 and WSHD_RAM 422, respectively, and used for shading correction. Alternatively, when the respective shading correction values are already obtained and stored, the stored values may be used if no abnormality is detected.

  In this way, when the chip average value of the white level is equal to or less than the predetermined threshold, it is determined that white shading is abnormal. In the example of FIG. 12, the chip average value of the white level of the chip 6 is the minimum and is close to the black level. If the minimum value is not more than a threshold value (not shown), it is determined that white shading is abnormal. With the above configuration, an abnormality in shading, particularly white shading, is determined not by each pixel but by the average value of the chip. For this reason, dust attached to a specific image sensor element is not determined to be abnormal, and abnormalities are caused by a plurality of image sensor elements (particularly a series of a plurality of elements) such as sensor chip failure or LED lighting failure. Detect. In other words, by filtering the high-frequency component of the white level value and determining in units of chips, it is possible to suppress the sensitivity to abnormality in units of pixels and improve the accuracy of abnormality detection. Furthermore, in this embodiment, since the shading correction value is not calculated, the abnormality determination process can be simplified. Furthermore, an abnormality such as a sensor chip or LED can be detected without changing the shading correction value that has already been calculated and stored in the memory.

[Embodiment 3]
In the third embodiment, an image sensor having a complicated configuration is used in order to perform reading at high speed, and each abnormal portion is determined. FIG. 14 shows the configuration of the image sensor. The image sensor 1401 is on both sides of the light sources 1411 and 1412 and illuminates the document via the light guide 1413. The sensor chips 1431 to 1447 are divided into six outputs in total and output. Its output is 1421-1426. This output is converted into a digital signal by an A / D converter (not shown) and output as a digital signal of the contact image sensor 201 and the contact image sensor 202 of the image block 300.

In the case of the image sensor shown in FIG. 14, a single sensor chip abnormality, a single output abnormality, a right LED light source abnormality, a left LED light source abnormality, and the like can be considered as abnormal states. Of these, the average value of the white shading data of each sensor chip when the right LED is abnormal is shown in FIG. 15A, and the average value of the white shading data of each sensor chip when any one output is abnormal is shown in FIG. Shown in The white shading data may be acquired in the same manner as in the second embodiment. In FIG. 15A, since the right LED light source 1412 is not lit, the white shading data of the right half is lowered. Therefore, if the uniformity is calculated for all the sensors, it becomes abnormal, and if the uniformity is calculated except for the right half, it becomes normal. Thereby, it can be determined that the right LED light source is abnormal. In FIG. 15B, one output (Out5 in this case) is abnormal, and the other outputs are normal. Therefore, when the uniformity is calculated except for the abnormal output chip, it is normal, and therefore it can be determined as a single output abnormality. Next, an abnormality determination flow in the third embodiment will be described. FIG. 16 is an abnormality determination flow according to the third embodiment. This is almost the same as in the second embodiment, but there is a difference in that the abnormality factor determination (S1607) is added before the white shading error (S1608). Details of the abnormality factor determination (S1607) are described in FIG. 17. The flow of determining the abnormality factor in FIG. 17 is described. In this example, the determination of the cause of abnormality is to determine the causes of single chip abnormality, single output abnormality, right LED light source abnormality and left LED light source abnormality. First, in single chip abnormality determination (S1701), white The uniformity index Up is calculated using all the remaining chips except for the minimum chip average value WSHD_Min of the shading data. That is, the uniformity index Up is calculated in the same manner as in the first embodiment using the second smallest value WSHD_2nd_Min instead of the minimum value WSHD_Min. If the uniformity index Up is within the standard (smaller than a predetermined value, that is, Up <UP_TH), it is determined that a single chip is abnormal. If it is determined that there is a single chip abnormality, a flag indicating a single chip abnormality is set (S1702) and white shading abnormality error processing is performed. The flag can be used for replacing the reading unit or calculating the failure rate as necessary.

  Similarly, in the single output abnormality determination (S1703), the uniformity index Up of the remaining chips is calculated except for the chip having the same output as the chip whose chip average of the white shading data is the minimum value WSHD_Min. If the value is within the standard (that is, smaller than the threshold value), it is determined that the single output is abnormal. If it is determined that the single output is abnormal, a flag indicating the single output abnormality is set (S1704), and error processing is performed.

  In the left LED light source abnormality determination (S1705), as described above, the right half is normal, and thus the normality is calculated when the uniformity index is calculated excluding the left half (that is, the right half uniformity index Up is the threshold value). If it is smaller, a left LED light source abnormality flag is set (S1706) and error processing is performed. The right LED light source abnormality determination (S1707) is the same processing as that on the left side. However, the uniformity index Up is a value obtained by using the minimum value and the maximum value of the white shading data of the left half chip. If the value is smaller than the predetermined threshold, the right LED light source abnormality flag is set (S1708). In this way, the sensor array is divided into two groups at approximately the center (in this example, it may be in units of chips), and the uniformity index of the average value in units of chips is determined for each side. If there is no abnormality on the determined one side, it is determined that there is an abnormality in the illumination on the opposite side that is not the object of determination.

  Finally, the case that does not apply to any of the above is determined as a multi-chip abnormality (S1709), and a flag indicating a multi-chip abnormality is set to perform error handling. In the above procedure, “error processing is performed” is described. For example, an error code corresponding to a flag indicating the type of abnormality may be displayed on the user interface screen of FIG.

  As described above, in the present embodiment, as in each of the above-described embodiments, shading, particularly white shading abnormality, is determined not by each pixel but by the average value of the chip. For this reason, dust attached to a specific image sensor element is not determined to be abnormal, and abnormalities are caused by a plurality of image sensor elements (particularly a series of a plurality of elements) such as sensor chip failure or LED lighting failure. Detect. In other words, by filtering the high-frequency component of the white level value and determining in units of chips, it is possible to suppress the sensitivity to abnormality in units of pixels and improve the accuracy of abnormality detection. Furthermore, in the present embodiment, the value of the white shading data is processed according to an abnormal state that may occur, and is inspected, whereby the type of abnormality that has occurred can be specified and notified to the user.

[Other Embodiments]
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in the computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

411 Black shading correction unit, 412 BSHD_RAM, 421 White shading correction unit, 422 WSHD_RAM, 431 Line integration calculation unit, 441
White shading correction value calculation unit, 451 Black shading correction value calculation unit, 471 Maximum / minimum / integrated value calculation unit

Claims (10)

An image reading apparatus for reading an image by a reading unit having an image sensor including a plurality of groups including a plurality of reading elements,
A controller for correcting a signal based on the output from the reading element;
The controller is
Generating a characteristic value indicating a shading characteristic for each reading element;
Based on the characteristic value for each reading element, obtain an average value of the characteristic values for each group,
The image is characterized in that the reading unit is determined to be abnormal when an average value of at least one of the characteristic values of the group is greater than a first threshold value. Reader. The image reading apparatus according to claim 1,
The controller is
An image reading apparatus that generates a shading correction value as the characteristic value. The image reading apparatus according to claim 1,
The controller is
An image reading apparatus that generates a difference between white shading data and black shading data as the characteristic value. The image reading apparatus according to any one of claims 1 to 3,
The control unit further includes:
Based on the average value of the characteristic values, a uniformity index indicating uniformity of the average value of the characteristic values is obtained, and when the uniformity index exceeds a predetermined value and indicates that it is not uniform, the reading is performed. An image reading apparatus that determines that an abnormality has occurred in a portion. An image reading apparatus according to any one of claims 1 to 4,
The controller is
If it is determined that an abnormality has occurred in the reading unit,
An image reading apparatus characterized by determining the cause of the abnormality. The image reading apparatus according to claim 5,
The controller is
An average value of the shading data is obtained for each group of the plurality of reading elements, a uniformity index indicating uniformity of the average value obtained by removing a minimum value from the average value for each group is obtained, and the uniformity index is An image reading apparatus characterized by determining that an abnormality has occurred in one of the groups when it is shown that the image is uniform. The image reading apparatus according to claim 6,
The output of the plurality of reading elements is output as an output that combines the plurality of groups,
The controller is
From the average value for each group, obtain a uniformity index indicating the uniformity of the average value excluding the average value of the group collected in the same output as the minimum value, and the uniformity index is uniform If so, it is determined that an abnormality has occurred in one of the outputs,
From the average value for each group, a uniformity index indicating the uniformity of the average value of the half group on one side is obtained, and when the uniformity index indicates that it is uniform, the opposite of the one side It is determined that there is an abnormality in the side lighting,
An image reading apparatus characterized by determining that an abnormality has occurred in a plurality of the groups when none of the above applies. An image reading apparatus according to any one of claims 1 to 7,
The image reading apparatus, wherein the error processing includes outputting a message indicating that an error has occurred. An image reading apparatus according to any one of claims 1 to 8,
The image sensor is configured by arranging a plurality of sensor chips having a plurality of reading elements,
The image reading apparatus according to claim 1, wherein the control unit sets a reading element for each sensor chip as a group of the reading elements. A method for controlling an image reading apparatus that reads an image by a reading unit having an image sensor including a plurality of groups each including a plurality of reading elements,
The image reading apparatus includes a control unit that corrects a signal based on an output from the reading element,
The controller is
Generating a characteristic value indicating a shading characteristic for each reading element;
Based on the characteristic value for each reading element, obtain an average value of the characteristic values for each group,
The image is characterized in that the reading unit is determined to be abnormal when an average value of at least one of the characteristic values of the group is greater than a first threshold value. A method for controlling a reader.

Images