JP2014045443A - Shading correction device and shading correction method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a shading correction device which can detect a white level peak value even if the peak position in one line for scanning a line sensor is unknown.SOLUTION: Shading correction is performed by performing black correction of an image signal read from a manuscript 11 being subjected to image processing to adjust the offset amount, and performing white correction to adjust the gain. The shading correction device comprises: a black correction circuit 15 performing black correction of an image signal; a variable gain amplifier 18 for receiving a signal from the black correction circuit 15 after correction and outputting the signal after amplifying with a desired gain; and a white correction circuit 17 connected with the variable gain amplifier 18 and performing white correction based on a true white level peak value detected from an output signal of the variable gain amplifier 18. Black correction is performed by the black correction circuit 15, and white correction is performed for a signal passed through the variable gain amplifier 18 by the white correction circuit 17.

Description

  The present invention relates to a shading correction apparatus and method, and more particularly to a shading correction apparatus and method for white correction of gain variation after black correction of an offset included in an output signal of a line sensor.

  In general, a line obtained by reading a document to be image-processed by a scanning operation of a line sensor such as a CCD in an image processing system constituting an information processing apparatus such as a facsimile, a digital copying machine, and a scanner as a shading correction apparatus. There is a technique in which offset and gain variations included in the output signal of the line sensor are corrected with respect to the output signal of the sensor, and image processing is performed using the corrected signal as a signal representing a true image processing target.

  Further, for example, in Patent Document 1, as a shading correction device, a document reading device, and an image forming device, high-quality shading correction data is stably generated, and this is automatically realized by each correction data generation device. What to do is disclosed. This generates shading correction data by reading only the data in the appropriate area from the image data read from the reference white board. With the shading correction data generated in this way, highly accurate shading correction without the influence of noise such as reflection on the contact glass surface can be easily realized. In addition, it is possible to easily or automatically set a region free from the influence of individual noise due to variations in assembly dimension differences among individual image reading apparatuses. Thus, highly accurate shading correction can be performed for each individual product.

  Specifically, as means for setting the sub-scanning position for reading the reading data of the reference white board in the position memory, one of the optical means for projecting image light on the image line sensor and the reference white board is sub-scanned with respect to the other. Means for generating a reference whiteboard reading data reading instruction signal SMPL based on the subscanning position of the position memory during the subscanning drive, and reading the reference whiteboard reading image data in response to the SMPL to generate shading correction data The correction data storing means for storing in the FIFO memory and the timing means for automatically detecting the effective area for reading the reference white plate and setting the SMPL occurrence position information.

  For example, Patent Document 2 discloses an image signal processing method capable of quickly obtaining appropriate shading correction data in an image signal processing method for shading correction of a color analog image signal. Specifically, the minimum output level is detected from the levels of the plurality of analog image signals output by the image line sensor of each color component with the light source turned off, and the input range of the analog / digital converter is detected to the detected minimum output level. Shift so that the lowest value matches. Next, the light source is turned on, and the image line sensor for each color component detects the maximum output level from the levels of a plurality of analog image signals read and output from the white reference plate. Then, based on the detected minimum output level, maximum output level, shifted input range, and analog / digital converter resolution, the analog / digital converter input range is set by a predetermined arithmetic expression, and shading correction data is acquired. Is.

  Further, for example, Patent Document 3 discloses a shading device that enables high-precision correction by preventing the offset error and gain error of an analog circuit from being included in data subjected to shading correction and A / D converted. Is disclosed. Specifically, it is as follows. The register stores the output of the A / D converter when the output signal of the line sensor is set to a fixed value and the black correction value (offset correction value) is set to the minimum value. The register stores the output of the A / D converter when the output signal of the line sensor is fixed at a fixed value and the black correction value is set to the maximum value. For example, the memory stores the output result when the A / D converter performs A / D conversion with the black correction value set to the minimum value for the output of the line sensor in the “OFF” state. Is done. The computing unit calculates the black correction value by using a predetermined arithmetic expression using the stored contents.

JP 2006-245717 A JP 2006-180153 A JP 2000-278495 A (Patent No. 4204690)

  The above-described shading correction apparatus of Patent Document 1 mainly produces a wrong reference white value in the output of the line sensor due to reflection on the contact glass surface, particularly strong reflected light at the glass start end, in a copying machine. An attempt to achieve the purpose of eliminating the influence of noise when the level is not indicated and the purpose of easily or automatically setting an area where there is no influence of individual noise due to variations in assembly dimensions of individual copying machines, etc. Is. However, the shading correction apparatus of Patent Document 1 is not always simple, and the white level peak position output by the line sensor is not known where it is in one line scanned by the line sensor. Even in such a case, there is a problem that the white level peak value to be detected for white correction cannot be detected accurately.

  In addition, the image signal processing method described in Patent Document 2 described above can be quickly performed from the output signal that the image line sensor of each color component reads and outputs the white reference plate in a state where the light source is unstable immediately after the light source is turned on. This is a method for achieving the object of obtaining shading correction data suitable for the above. However, the image signal processing method of Patent Document 2 is not always simple, and the white level peak position output from the line sensor is not known in one line scanned by the line sensor. However, there is a problem that the white level peak value to be detected for white correction cannot be detected accurately.

  Further, the above-described shading correction apparatus of Patent Document 3 performs analog / digital (A / D) conversion on the data subjected to black correction and white correction, and as a result, analog included in the result of A / D conversion. This is intended to achieve the purpose of removing the offset and distortion of the circuit system. However, the shading correction apparatus disclosed in Patent Document 3 is not always simple, and it is a case where the white level peak position output from the line sensor is unknown in one line scanned by the line sensor. However, there is a problem that the white level peak value to be detected for white correction cannot be detected accurately.

  Specifically, when the white level peak value is at a position shifted from the intermediate position in one scanning line, the peak value of the white correction value cannot be detected correctly, and cannot be converged to the correct target white correction value. The reason for this is that according to the method of estimating the peak value from the average value over the entire period in one line of scanning, in the scanning one line, before and after the largest first peak representing 100% white level, in a close time. This is because the average value including the second largest second peak that has occurred is calculated. That is, only the peak value at the shifted position is the original 100% white level peak value, but if the peak value is estimated from the average value over the entire period in one line of scanning, the peak value is more erroneous than the original peak value. Estimate to a small value. In this case, the gain of a programmable gain amplifier (PGA) is corrected with an excessively large correction value with respect to the original ideal white target level. As a result, there has been a problem that accurate white correction cannot be performed.

  The present invention has been made in view of such problems, and the object of the present invention is when the peak position of the white level output from the line sensor is unknown in one line scanned by the line sensor. Even so, an object of the present invention is to provide a shading correction apparatus and a shading correction method capable of accurately detecting a white level peak value to be detected for white correction.

  The present invention has been made in order to achieve such an object. The invention according to claim 1 is directed to black correction of an image signal obtained by reading a document (11) to be subjected to image processing, thereby reducing an offset amount. In the shading correction apparatus (Example 1 shown in FIG. 1) that adjusts and adjusts the gain by white correction, the black correction circuit (15) that corrects the image signal to black and the correction by the black correction circuit (15) A variable gain amplifier (18) that inputs the subsequent signal, amplifies it with a desired gain and outputs it, and true white detected from the output signal of the variable gain amplifier (18) connected to the variable gain amplifier (18) A white correction circuit (17) for performing white correction based on the level peak value, and black correction is performed by the black correction circuit (15), and the signal passed through the variable gain amplifier (18) And performing white correction by the white correction circuit (17).

  According to a second aspect of the present invention, the white correction circuit (17) is configured to detect a white level peak value in the shading correction device (first embodiment shown in FIG. 3) according to the first aspect. 1H), a first averaging circuit (31) for calculating a moving average value from a white level value for each of a plurality of moving average sections (1-1... 1-7) provided in 1H), and in the detection section (1H) The moving average value is calculated from the white level value for each of a plurality of moving average sections (2-1... 2-6) different from the plurality of moving average sections (1-1. A second average circuit (32) that compares the first moving average value calculated from the first average circuit (31) with the second moving average value calculated from the second average circuit (32). A selection circuit (33) for selecting one of the moving average values, and the selection circuit (33) Based on the-option moving average value, characterized in that a variable gain amplifier (34, 37) for performing white correction.

  Further, the invention according to claim 3 is the shading correction apparatus according to claim 2 (Example 1 shown in FIG. 3), wherein the plurality of moving average intervals (1-1...) Are included in the detection interval (1H). 1-7) one section (for example, 1-1) and one section (2-1) of the other plurality of moving average sections (2-1 ... 2-6) corresponding to the one section (1-1) ) Is characterized by having a time lag.

According to a fourth aspect of the present invention, in the shading correction apparatus according to the second or third aspect (the first embodiment shown in FIG. 3), the selection circuit (33) includes the first moving average value and the first moving average value. The larger moving average value is selected by comparing with the second moving average value.
According to a fifth aspect of the present invention, in the shading correction apparatus according to any one of the first to fourth aspects (Embodiment 2 shown in FIG. 10), the variable gain amplifier (18) is a digital programmable gain amplifier. (18), an analog programmable gain amplifier (13) is inserted before or after the digital programmable gain amplifier (18), and the white color output from the white correction circuit (17) is output to the analog programmable gain amplifier (13). The correction signal is added and the analog programmable gain amplifier (13) also performs white correction.

  According to a sixth aspect of the present invention, there is provided a shading correction method in which an image signal obtained by reading a document (11) to be subjected to image processing is black corrected to adjust an offset amount, and white correction is performed to adjust a gain. 2 is a first embodiment shown in FIG. 2, in which the image signal is black-corrected (S101 shown in FIG. 2), and the black-corrected signal is input to a variable gain amplifier (18) with a desired gain. Amplifying (S102), and a white correction circuit (17) performing white correction based on the true white level peak value detected from the output signal of the variable gain amplifier (18) (S103). It is characterized by.

  According to a seventh aspect of the present invention, in the shading correction method according to the sixth aspect (Example 1 shown in FIG. 9), the step of performing the white correction (S103 shown in FIG. 2) includes a white level peak value. A plurality of moving average sections (1-1... 1-7 shown in FIG. 8C) are provided in the detection section (1H) for detecting the moving average sections (1-1... 1-7). The step of calculating the first moving average value from the white level value (S111 shown in FIG. 9) is different from the plurality of moving average sections (1-1... 1-7) in the detection section (1H). A plurality of moving average sections (2-1... 2-6) are provided, and the white level value of each of the other plurality of moving average sections (2-1... 2-6 shown in FIG. 8D) is calculated. A step of calculating two moving average values (S112), and comparing the first moving average value and the second moving average value. A step (S113) for, characterized in that it comprises a step (S114) of performing white correction based on one of the first moving average value and the second moving average value the comparison.

  The invention according to claim 8 is the shading correction method according to claim 7 (the first embodiment shown in FIGS. 1 to 9), wherein the plurality of moving average intervals (1− 1 ... 1-7) (e.g., 1-1) and one of the other moving average intervals (2-1 ... 2-6) corresponding to the interval (1-1) (2- 1) is characterized in that there is a time lag.

  The invention according to claim 9 is the shading correction method according to claim 7 or claim 8 (Example 1 shown in FIGS. 1 to 9), wherein the compared first moving average value and second movement are compared. In the step of performing white correction based on either one of the average values (S114), the moving average determined to be larger than the step (S113) of comparing the first moving average value and the second moving average value It is characterized by using a value.

  The invention described in claim 10 is the shading method according to any one of claims 6 to 9 (second embodiment shown in FIG. 10), wherein the variable gain amplifier (18) is a digital programmable gain amplifier ( 18), and adding the white correction signal output from the white correction circuit 17 to the analog programmable gain amplifier (13) inserted before or after the digital programmable gain amplifier (18), the analog programmable gain amplifier ( 13) is also characterized in that white correction is performed.

  The invention described in claim 11 is the shading correction method according to claim 7 (the first embodiment shown in FIG. 9), which is one of the compared first moving average value and second moving average value. In the step of performing white correction based on one (S114), at the end of each moving average section in the plurality of moving average sections (1-1... 1-7, 2-1. A moving average value is selected.

  According to the present invention, the white level peak value to be detected for white correction can be obtained even when the peak position of the white level output from the line sensor is unknown in one line scanned by the line sensor. It is possible to provide a shading correction apparatus and a shading correction method that can be accurately detected.

It is a block block diagram for demonstrating Example 1 of the shading correction apparatus which concerns on this invention. It is a figure which shows the flowchart for demonstrating Example 1 of the shading correction method which concerns on this invention. It is a block block diagram for demonstrating the white correction circuit in Example 1 of the shading correction apparatus which concerns on this invention. It is a figure which shows the flowchart which shows the whole operation | movement of the white correction by the moving average in Example 1, 2 of the shading correction apparatus based on this invention. It is explanatory drawing regarding the offset and gain variation of a line sensor. (A) thru | or (c) is explanatory drawing in case a peak value is obtained correctly in one line of scanning, since there is one line sensor output peak. (A) is a white level peak value. (B) is a white level peak detection section. (C) is a white level peak detection value. (A) thru | or (c) are explanatory drawings in case a white level peak value becomes inaccurate because it has two peaks in the line sensor output in one line of scanning. (A) is a white level peak value. (B) is a white level peak detection section. (C) is a white level peak detection value. (A) thru | or (f) is explanatory drawing which detects a moving average value since it has two peaks in a line sensor output in 1 line of scanning. (A) is a white level peak value. (B) A detection interval. (C) is a moving average section at detection timing 1. (D) is a moving average section based on the detection timing 2. (E) is a white level peak value at detection timing 1. (F) is a white level peak value at the detection timing 2. It is a figure which shows the flowchart explaining the operation | movement of the white correction by a moving average in Example 1 of the shading correction method which concerns on this invention. It is a block block diagram for demonstrating Example 2 of the shading correction apparatus which concerns on this invention.

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

Hereinafter, Embodiment 1 of the present invention will be described with reference to FIGS.
FIG. 1 is a block diagram for explaining a first embodiment of a shading correction apparatus according to the present invention. A shading correction apparatus 100 shown in FIG. 1 includes a line sensor 12, an analog / digital converter (ADC) 16, a black correction circuit 15, a digital / analog converter (DAC) 14, an adder / subtractor 19, and a digital PGA 18. And a white correction circuit 17. The line sensor 12 is constituted by a CCD or the like, and outputs a signal including a true signal component based on an analog value corresponding to reading of the document 11 and variations in offset and gain. The adder / subtracter 19 subtracts the output signal of the DAC 14 from the output signal of the line sensor 12 and inputs it to the analog / digital (ADC) converter 16.

  The black correction circuit 15 performs black correction on the digital signal digitized through the ADC 16. The DAC 14 subtracts and inputs to the adder / subtracter 19 an analog signal obtained by digital / analog (D / A) conversion of an offset amount appearing as an output signal of the black correction circuit 15. The adder / subtracter 19 subtracts the output signal of the DAC 14 from the output signal of the line sensor 12 and inputs the result to the ADC 16. The ADC 16 performs analog / digital (A / D) conversion on the output of the adder / subtractor 19 and inputs the result to a digital programmable gain amplifier (PGA) 18. The white correction circuit 17 performs white correction on the digital PGA 18 based on the digital output signal of the digital PGA 18. As a result, the digital PGA 18 is adjusted to an optimum gain based on the output signal of the white correction circuit 17.

  The shading correction apparatus 100 according to the first embodiment illustrated in FIG. 1 adjusts an offset amount by performing black correction on an image signal obtained by reading a document 11 to be subjected to image processing, and adjusts a gain by performing white correction. Perform shading correction. The shading correction apparatus 100 includes a black correction circuit 15 that black corrects an image signal, a variable gain amplifier 18 that receives a signal corrected by the black correction circuit 15, amplifies the signal with a desired gain, and outputs the amplified signal. And a white correction circuit 17 that is connected to the gain amplifier 18 and performs white correction based on the true white level peak value detected from the output signal of the variable gain amplifier 18. Then, the black correction circuit 15 performs black correction, and the white correction circuit 17 performs white correction on the signal that has passed through the variable gain amplifier 18. The variable gain amplifier 18 is the digital PGA 18 described above.

And the main operation | movement of this shading correction apparatus 100 is performed in the order shown to following (1)-(3).
(1) A black correction value detection operation is performed.
(2) A white correction value detection operation is performed.
(3) A document reading operation is performed.
That is, (1) black correction is required before (2) white correction value detection operation, and (3) original reading operation is completed with both (1) black correction and (2) white correction. Is an essential requirement.

  The shading correction method performed by the shading correction apparatus 100 according to the first embodiment shown in FIG. 1 is as shown in the flowchart below. The output of the line sensor 12 is black corrected to adjust the offset amount and Correct and adjust the gain.

  FIG. 2 is a flowchart for explaining the shading correction method according to the first embodiment of the present invention. That is, this shading correction method is a shading correction method in which an image signal obtained by reading an original 11 to be subjected to image processing is black-corrected to adjust an offset amount, and white is corrected to adjust a gain. The procedure of this shading correction method is as follows. First, the image signal is black corrected. (S101) The signal after the black correction in S101 is input to the variable gain amplifier 18 and amplified with a desired gain. (S102) The white correction circuit 17 completes the shading correction by performing white correction based on the true white level peak value detected from the output signal amplified in S102. (S103)

  FIG. 3 is a block diagram showing the configuration of the white correction circuit 17. The white correction circuit 17 shown in FIG. 3 includes a timing averaging circuit 30, 2 to 1 selectors 33 and 35, comparators 34 and 37, registers 36 and 38, and a control circuit 39. The timing average circuit 30 includes a timing 1 average circuit 31 and a timing 2 average circuit 32 connected in parallel so as to distribute the output of the digital PGA 18, and outputs of the timing 1 average circuit 31 and the timing 2 average circuit 32 respectively. Is input to the 2to1 selector 33.

  The control signal of the control circuit 39 is input as a selective load condition of the 2to1 selector 33. The control circuit 39 generates the timing 1 to control the timing 1 average circuit 31 and also generates the timing 2 to control the timing 2 average circuit 32. The selection output of the 2to1 selector 33 is input to the comparator 34 and input as an option of the 2to1 selector 35. The output of the comparator 34 is input to the 2to1 selector 35 as a control signal. The data stored in the register 36 is input to the comparators 34 and 37, and is input as an option of the 2to1 selector 35. The white correction target level stored in advance in the register 38 is used as a reference value by the comparator 37.

  Hereinafter, the operation of the white correction circuit 17 will be briefly described. In the white correction period, a signal corresponding to the white level is input from the digital PGA 18, and the average value is calculated by the timing 1 averaging circuit 31 and the timing 2 averaging circuit 32, respectively. Further, in the control circuit 39, timing 1 and timing 2, which will be described later along FIGS. 8C to 8F, are generated. This control circuit 39 gives a control signal as a selective load condition to the 2to1 selector 33. Therefore, the 2to1 selector 33 operates so as to select and fetch the calculation result of the timing 1 average circuit 31 based on the timing 1 and the calculation result of the timing 2 average circuit 32 based on the timing 2 in a half-phase delay relationship. To do.

  As a result, the moving average values of the timing 1 averaging circuit 31 and the timing 2 averaging circuit 32 are obtained at the output terminal of the 2to1 selector 33. The 2to1 selector 33 uses the comparator 34 to compare the magnitudes of the timing 1 and the timing 2 for each moving average section. The 2to1 selector 33 has a large calculation result of the timing 1 average circuit 31 and the timing 2 average circuit 32 each time the moving average interval of the timing 1 and the timing 2 is ended by the signal generated by the control circuit 39. Is selected and input to the comparator 34. In addition, the size comparison for each moving average section between the timing 1 and the timing 2 will be described later with reference to FIGS. 8C and 8D, 1-1 to 2-1, 2-1 to 1-2,. , 2-6 vs. 1-7, and the comparison is continued.

  Next, the comparator 34 compares the calculation result selected by the 2to1 selector 33 with the data of the previous moving average section stored in the register 36 in the above-described moving average section, and the 2to1 selector 35. As a control signal. The 2to1 selector 35 selects a value updated for each moving average section as the white level peak value stored in the register 36. Thus, the moving average is calculated by updating the white level peak value at timing 1 and timing 2 for each moving average section. When the white correction period ends, the comparator 37 compares the output of the register 36 using the white correction target level stored in the register 38 as a reference value, and outputs the comparison result. The white correction circuit 17 controls the gain of the digital PGA 18 to match the white correction target level based on the comparison result output from the comparator 37.

Next, the white correction operation after the black correction is performed will be described below.
FIG. 4 is a flowchart showing the entire white correction operation by moving average in the first and second embodiments of the shading correction apparatus according to the present invention. As shown in FIG. 4, when the white correction period is started, the number of set lines is compared with the number of executions. (Step S1) In step S1, if “the number of set lines> the number of executions” is Yes, that is, if the number of executions is less than the number of set lines, the moving average value of the white correction period is calculated for each set line to obtain the peak value Is detected. (Step S2) The peak value detected in Step S2 is compared with the target value to determine whether or not the fine gain has been determined. (Step S4) Here, the operation proceeds from step S2 to step S4.

FIG. 4 is a diagram combining the first embodiment and the second embodiment. In FIG. 4, the difference between the first embodiment and the second embodiment is the presence or absence of steps S3 and S5. In the second embodiment, steps S3 and S5 executed in the second embodiment are omitted.
If No in step S4, that is, if the fine gain has not been determined, the gain of the digital PGA 18 is changed. (Step S6) If the gain of the digital PGA 18 is changed in step S6, the process returns to step S1 and the comparison of the number of set lines and the number of times is repeated. If YES in step S4, that is, if the fine gain determination is completed, the white correction ends. (Step S7)

  FIG. 5 is an explanatory diagram relating to variations in offset and gain of the line sensor. As indicated by the hatched portion in FIG. 5, the offset is, for example, that each pixel of the line sensor should originally output in the state where the irradiation light to the line sensor is eliminated when scanning a reference black plate (not shown). It is the difference between the reference signal and the actual line sensor output value. Also, the gain variation is, for example, when a light amount irradiated to the line sensor is set to a predetermined amount by scanning a reference white plate (not shown), and the output signal of each pixel of the line sensor is in one scanning line. What is supposed to be flat means that the peak value increases or decreases due to distortion or scratches of the optical system. In order to correct this gain variation, for example, the pixel gain of the output signal 0.5 times the peak value is set to “2 (= 1 / 0.5)” with the peak value as a reference, and each pixel The output characteristics in a predetermined light quantity irradiation state are flattened.

  In this specification, the offset for each pixel is referred to as “black distortion”. Correcting this black distortion is called “black correction”. For this black correction, the offset correction value is detected by A / D converting the output of the line sensor in a state where the irradiation light to the line sensor is eliminated. The gain variation for each pixel is called “white distortion”. Correcting this white distortion is called “white correction”. For this white correction, the gain variation correction value is detected by A / D converting the line sensor output when the reference white plate is scanned so that the amount of light applied to the line sensor is a predetermined amount.

  Note that it is necessary to perform white correction after performing black correction. That is, the shading correction is completed by performing black correction and white correction on the output signal of the line sensor obtained when the image to be read such as a document is scanned by the line sensor.

Next, a white level peak value necessary for white correction, a calculation timing for detecting the peak value, and the like will be described below.
FIG. 6 is an explanatory diagram in the case where the peak value is obtained accurately because one line sensor output has one peak in one scanning line. FIG. 6A shows the white level peak value. FIG. 6B shows a white level peak detection section. FIG. 6C shows a white level peak detection value. When a white level peak value exists at the center position in one line of scanning shown in FIG. 6A, as shown in FIG. 6B, a white level peak detection section corresponding to one line of scanning is obtained. And the average value is calculated in the white level peak detection section 1H, and the peak value required for the white correction value can be determined. By using the peak value of the white correction value determined in this way, the gain of the digital PGA 18 of the shading correction apparatus 100 described above with reference to FIG. Can do.

  That is, when detecting the white level peak value, the white level peak detection section is set, the average value of the section is calculated, the PGA is operated with respect to the white target level, and white correction is performed. It is. Here, in the white correction period, when it is known where the peak is in one line scanning the line sensor, that is, the white level peak value is at the center as shown in FIG. In such a case, the white level peak detection section 1H is set as shown in FIG. The average value can be calculated in the white level peak detection section 1H to determine the peak value of the white correction value, and can be converged to the target white correction value.

  However, in the above-described method, if it is not known where the peak position of the white level output from the line sensor is in one line that is scanning the line sensor, the white level peak value to be detected for white correction is determined. , Can not be detected accurately. Such a problem will be described below.

  FIG. 7 is an explanatory diagram when the white level peak value becomes inaccurate because one line of scanning has two peaks in the line sensor output. FIG. 7A shows a white level peak value. FIG. 7B shows a white level peak detection section. FIG. 7C shows white level peak detection values. The detection interval 1H shown in FIG. 7B and the average value of the detection values shown in FIG. 7C will be described later with reference to other figures.

  As shown in FIG. 7, when detecting the white level peak value, the white level peak detection section 1H is set, and the average value of the detection values in that section is calculated. Using the average value shown in FIG. 7C, it is possible to perform white correction by operating the PGA so that the white target level, which is the original ideal value, is referred to as the target value. Specifically, the white correction circuit 17 shown in FIG. 3 appropriately controls the gain of the digital PGA 18.

  For example, when the white level peak value is at a position slightly shifted to the right as shown in FIG. 7, the peak value of the white correction value cannot be detected correctly and does not converge to the target white correction value. The peak value appearing at the position shifted to the right is the original white level peak value. However, if the average value of the entire white level peak detection section 1H (hereinafter referred to as section 1H) is calculated, the second peak located on the left shoulder of the first peak of the original white level is also added. . Therefore, the average value is smaller than the original first peak value alone. In this case, since the gain of the PGA is controlled using a correction value that is unnecessarily larger than the original ideal for the white target level, the result of white correction is inaccurate.

  FIGS. 8A to 8F are explanatory diagrams for detecting the moving average value because one line of scanning has two peaks in the line sensor output. FIG. 8A shows the white level peak value. FIG. 8B is a detection section. FIG. 8C shows a moving average section based on the detection timing 1. FIG. 8D shows a moving average section based on the detection timing 2. FIG. 8E shows the white level peak value at the detection timing 1. FIG. 8F shows the white level peak value at the detection timing 2. Then, when detecting the white level peak value in one scanning line, the moving average section 1-1... 1-7 shown in FIG. 8C is included in the white level peak detection section 1H shown in FIG. Set and calculate the moving average of each section. If the calculated moving average value for each section is larger than the white target level, the gain of the digital PGA 18 (FIG. 1) is decreased, and if the moving average value is smaller than the white target level, the gain of the digital PGA 18 is increased. Control.

  As shown in FIGS. 8A, 8E, and 8F, even when the line sensor output in one scanning line has two peaks, the peak value is estimated to some extent by calculating the moving average value. can do. However, only with the white level detection timing 1, there is a possibility that the peak detection may fail due to a reason that a small peak and valley are mixed in the same section and no difference occurs. Therefore, as shown in FIGS. 8C and 8D, during the white level peak detection section 1H, the white level detection timing 1 and the half time of the moving average section 1-1 from the white level detection timing 1 are shown. A white level detection timing 2 shifted by (hereinafter referred to as half phase) is set. As described above, the time setting (hereinafter referred to as phase) at the start and end of the moving average section of one cycle at detection timing 1 sets two timings having a phase difference. The peak value is estimated based on the detection timing 1 and the detection timing 2 and the moving average value calculated for each moving average section at each timing. By doing so, even if the above-described peak detection failure occurs from the moving average value calculated at one detection timing 1, it is possible to compensate with the moving average value calculated at the other detection timing 2. As a result, an accurate peak value of the white level can be detected.

  Even when the peak is sharp, the peak can be detected more finely and accurately by adjusting the length of the moving average section (hereinafter referred to as the period). Further, the timing for updating the white level peak value is updated every time the moving average sections 1-1... 1-7, 2-1. For example, the size comparison is continued by the combination of 1-1 to 2-1, 2-1 to 1-2,..., 2-6 to 1-7 shown in FIGS. For example, if 2-3, 1-4, and 2-4 are continuously compared in magnitude, it is estimated that 1-4 or 2-4 indicating the maximum value will be the peak value. Then, even if the line sensor output in one scanning line has two peaks, if a larger first peak is detected after the second peak is detected, it is assumed that there is a true peak value. Updated. As a result, the white level peak value to be detected for white correction can be accurately detected. Here, as a result of comparing the magnitude relation of the moving average values in the moving average section 1-4 and the moving average section 2-4, the maximum value that is a true peak value in the moving average section 1-4 is shown here. Is detected.

  Note that, regarding the selection of the moving average value, the 2to1 selector 33 indicates the calculation result of the timing 1 averaging circuit 31 every time the moving average interval of the timing 1 and the timing 2 is ended by the signal generated by the control circuit 39. The first moving average value is compared with the second moving average value that is the calculation result of the timing 2 averaging circuit 32, and the larger one is selected and output to the next processing. The white correction circuit 17 compares the first moving average value and the second moving average value, selects the larger moving average value, and based on the selected larger moving average value, gain of the analog PGA 13 The gain of the digital PGA 18 is controlled to match the white correction target level. In some cases, the method is not necessarily limited to the method of selecting the larger moving average value.

  According to the shading correction apparatus according to the first embodiment, even if it is not known where the peak position of the white level output from the line sensor is in one line scanned by the line sensor, it is detected for white correction. It is possible to provide a shading correction apparatus and a shading correction method capable of accurately detecting a white level peak value to be detected.

  The white correction circuit 17 in the shading correction apparatus shown in FIG. 1 includes a first average circuit 31, a second average circuit 32, a selection circuit 33, and programmable gain amplifiers 34 and 37. The first average circuit 31 calculates a moving average value from the white level value for each of the plurality of moving average sections 1-1... 1-7 provided in the detection section 1H for detecting the white level peak value. The second average circuit 32 determines the white level value for each of a plurality of moving average sections 2-1... 2-6 different from the plurality of moving average sections 1-1... 1-7 provided in the detection section 1H. Calculate the moving average value. The selection circuit 33 compares the first moving average value calculated from the first average circuit 31 with the second moving average value calculated from the second average circuit 32 and selects one of the moving average values. The programmable gain amplifiers 34 and 37 perform white correction based on the moving average value selected by the selection circuit 33.

  FIG. 9 is a flowchart illustrating the white correction operation based on the moving average in the shading correction method according to the first embodiment. The step of performing white correction (S103) in the shading correction method shown in FIG. 2 includes steps S111 to S114 shown in FIG. That is, a plurality of moving average sections 1-1... 1-7 are provided in the detection section (1H) for detecting the white level peak value, and the first is determined from the white level value for each moving average section 1-1. A step of calculating a moving average value (S111), and a plurality of moving average sections 2-1 ... 2-6 different from the plurality of moving average sections 1-1 ... 1-7 are provided in the detection section (1H). The step of calculating the second moving average value from the value of the white level for each of a plurality of other moving average sections 2-1 ... 2-6 (S112) is compared with the first moving average value and the second moving average value. Step (S113) and Step (S114) of performing white correction based on one of the compared first moving average value and second moving average value are included.

  In the step of performing white correction based on one of the compared first moving average value and second moving average value (S114), a plurality of moving average sections 1-1... 1-7, 2-1 ... It is preferable to select one of the moving average values at the end of each moving average section in 2-6. At this time, it is more preferable to select the larger moving average value.

  FIG. 10 is a block diagram illustrating a configuration of a shading correction apparatus using the shading correction apparatus according to the second embodiment of the present invention. The shading correction apparatus 200 shown in FIG. 10 is obtained by inserting an analog PGA 13 between the adder / subtractor 19 and the ADC 16 in the shading correction apparatus 100 shown in FIG. A signal output from the white correction circuit 17 is added to the feedback path of the analog PGA 13 so that the analog PGA 13 is also subjected to white correction. That is, the shading correction apparatus 200 includes a line sensor 12, an analog PGA 13, an A / D converter (ADC) 16, a black correction circuit 15, a D / A converter (DAC) 14, an adder / subtractor 19, A digital PGA 18 and a white correction circuit 17 are provided. The line sensor 12 is constituted by a CCD or the like, and outputs a signal including a true signal component based on an analog value corresponding to reading of the document 11 and variations in offset and gain. The analog PGA 13 amplifies the output signal of the line sensor 12. The adder / subtracter 19 subtracts the output signal of the DAC 14 from the output signal of the line sensor 12 and inputs the result to the analog PGA 13. The analog PGA 13 amplifies the output of the adder / subtractor 19 and inputs the amplified output to the ADC 16.

  The black correction circuit 15 performs black correction on the digital signal digitized through the ADC 16. The DAC 14 adjusts an offset amount appearing as an output signal of the black correction circuit 15 and subtracts and inputs an analog signal obtained by D / A converting the offset amount to the adder / subtractor 19. The adder / subtracter 19 subtracts the output signal of the DAC 14 from the output signal of the line sensor 12 and inputs the result to the analog PGA 13. The ADC 16 A / D converts the output of the analog PGA 13 and inputs it to the digital PGA 18. The white correction circuit 17 performs white correction on the analog PGA 13 and the digital PGA 18 based on the digital output signal of the digital PGA 18. That is, the analog PGA 13 and the digital PGA 18 are each adjusted to an optimum gain based on the output signal of the white correction circuit 17.

  The second embodiment will be described below. The shading correction apparatus 200 is equipped with an analog PGA 13 in addition to the digital PGA 18 with respect to the shading correction apparatus 100 of the first embodiment shown in FIG. 1, and performs white correction by gain control with both PGAs. Except for this difference, the basic signal processing principles for white correction and black correction are the same. More specifically, in the shading correction apparatus 200 shown in FIG. 10, the analog PGA 13 is inserted between the adder / subtractor 19 and the ADC 16 in the shading correction apparatus 100 shown in FIG. In addition to the signal output from the correction circuit 17, the analog PGA 13 is also subjected to white correction.

In the shading correction apparatus according to the second embodiment, the white correction operation after the black correction is performed will be described below.
FIG. 4 is a flowchart showing the entire white correction operation by moving average in the first and second embodiments of the shading correction apparatus according to the present invention. As shown in FIG. 4, when the white correction period is started, the number of set lines is compared with the number of executions. (Step S1) In step S1, if “the number of set lines> the number of executions” is Yes, that is, if the number of executions is less than the number of set lines, the moving average value of the white correction period is calculated for each set line and the peak value is calculated. Is detected. (Step S2) The peak value detected in Step S2 is compared with the target value to determine whether or not the rough gain has been determined. (Step S3) No in Step S3, that is, if the rough gain is not yet determined, the value of the analog PGA 13 is sequentially compared for each line, and the gain of the analog PGA 13 is changed. (Step S5) If the gain of the analog PGA 13 is changed in step S5, the process returns to step S1 and the comparison of the number of set lines and the number of times is repeated. In step S1, if “the number of set lines> the number of executions” is No, that is, if the number of executions is not smaller than the number of set lines, the white correction is terminated. (Step S7)
The difference between the first embodiment and the second embodiment in FIG. 4 is the presence or absence of steps S3 and S5. In the second embodiment, steps S3 and S5 that were not the first embodiment are executed.

  The present invention provides an output of a line sensor obtained by reading a document to be image-processed by a scanning operation of a line sensor such as a CCD in an image processing system constituting an information processing apparatus such as a facsimile, a digital copying machine, and a scanner. It can be used for a shading correction device and a shading correction method for correcting an offset and gain variation included in the output signal of the line sensor with respect to a signal and performing the image processing using the corrected signal as a signal representing a true image processing target. It is.

1-1 ... 1-7, 2-1 ... 2-6 Moving average section 11 Document 12 Line sensor 13 Analog PGA (Analog variable gain amplifier)
14 DAC (digital / analog converter)
15 Black correction circuit 16 ADC (analog / digital converter)
17, 27 White correction circuit 18 Digital PGA (Digital variable gain amplifier)
19 Adder / Subtractor 30 Timing averaging circuit 31 Timing 1 averaging circuit 32 Timing 2 averaging circuit 33, 35 2to1 selectors 34, 37 Comparator (programmable gain amplifier)
36, 38, 92, 94 Register 39 Control circuit 91 Average circuit 93 Comparator 95 Control circuit 100, 200, 300 Shading correction device

Claims (11)

In a shading correction apparatus that adjusts an offset amount by black correction of an image signal obtained by reading a document to be subjected to image processing, and adjusts a gain by white correction,
A black correction circuit for black correction of the image signal;
A variable gain amplifier that inputs the signal corrected by the black correction circuit and outputs the signal after amplification by a desired gain;
A white correction circuit connected to the variable gain amplifier and performing white correction based on a true white level peak value detected from an output signal of the variable gain amplifier;
While performing black correction by the black correction circuit,
A shading correction apparatus, wherein white correction is performed by the white correction circuit on a signal that has passed through the variable gain amplifier. The white correction circuit is
A first average circuit for calculating a moving average value from white level values for each of a plurality of moving average sections provided in a detection section for detecting a white level peak value;
A second averaging circuit that calculates a moving average value from a value of the white level for each of a plurality of moving average sections different from the plurality of moving average sections provided in the detection section;
A selection circuit that compares the first moving average value calculated from the first average circuit and the second moving average value calculated from the second average circuit to select one of the moving average values;
The shading correction apparatus according to claim 1, further comprising: a variable gain amplifier that performs white correction based on the moving average value selected by the selection circuit.   3. A section of the plurality of moving average sections in the detection section and a section of the other plurality of moving average sections corresponding to the one section have a time lag. The shading correction apparatus described in 1.   4. The shading correction apparatus according to claim 2, wherein the selection circuit selects the larger moving average value by comparing the first moving average value and the second moving average value. 5. . The variable gain amplifier is a digital programmable gain amplifier, and an analog programmable gain amplifier is inserted before or after the digital programmable gain amplifier,
5. The shading correction apparatus according to claim 1, wherein the analog programmable gain amplifier performs white correction by adding a white correction signal output from the white correction circuit to the analog programmable gain amplifier. 6. . A shading correction method in which an image signal obtained by reading a document to be subjected to image processing is black-corrected to adjust an offset amount and white-corrected to adjust a gain.
Black correcting the image signal;
Inputting the black-corrected signal to a variable gain amplifier and amplifying it with a desired gain;
A white correction circuit performing white correction based on a true white level peak value detected from an output signal of the variable gain amplifier;
A shading correction method comprising: The step of performing the white correction includes
Providing a plurality of moving average sections in a detection section for detecting a white level peak value, and calculating a first moving average value from the white level value for each moving average section;
Providing a plurality of moving average sections different from the plurality of moving average sections in the detection section), and calculating a second moving average value from the white level value for each of the plurality of other moving average sections; ,
Comparing the first moving average value and the second moving average value;
The shading correction method according to claim 6, further comprising a step of performing white correction based on one of the compared first moving average value and second moving average value.   One section of the plurality of moving average sections and one section of the other plurality of moving average sections corresponding to the section in the detection section (1H) have a time lag. Item 8. The shading correction method according to Item 7.   The step of performing white correction based on one of the compared first moving average value and second moving average value is larger than the step of comparing the first moving average value and the second moving average value. 9. The shading correction method according to claim 7, wherein the moving average value of the determined one is used.   The variable gain amplifier is a digital programmable gain amplifier, and the analog programmable gain amplifier is configured by adding a white correction signal output from the white correction circuit 17 to an analog programmable gain amplifier inserted in a preceding stage or a subsequent stage of the digital programmable gain amplifier. 10. The shading method according to claim 6, wherein white correction is performed.   In the step of performing white correction based on one of the compared first moving average value and second moving average value, at the end of each moving average section in the plurality of moving average sections, one of the movements The shading correction method according to claim 7, wherein an average value is selected.

Images