JP2005051566A - Image reading apparatus and image reading method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image reading apparatus capable of emitting light with a maximum luminous quantity in various lighting patterns of a plurality of light sources, using a small gain to suppress a noise signal and obtaining an image of an original with high gradation. <P>SOLUTION: The image reading apparatus is provided with: N-sets of light sources for emitting light to the original or a reference board and whose luminous quantity can be adjusted; an optical read means for converting a reflected light from the original or the reference board into an analog signal; an adjustment means for applying gain processing to the analog signal; an analog / digital converter for converting the analog signal subjected to gain processing into a digital signal; and a control means for setting a gain to the adjustment means, wherein the control means lights up one of the N-sets of the light sources to apply read scanning to the reference board, acquires N-sets of gains of each light source at which an input signal given to the analog / digital converter reaches a level that is between a maximum input level and a minimum input level of the analog / digital converter and is nearly a multiple of [1/N] of a level in the vicinity of the maximum input level, sets one selected among the N-sets of the gains to the adjustment means and reads the image of the original in a state that the luminous quantity of the light sources other than the light source related to the selected gain is adjusted. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image reading apparatus and an image reading method for optically reading an original, and in particular, a technique for reading an original with as high image quality (high gradation) as possible in an image reading apparatus capable of selectively lighting a plurality of light sources. About.

  In an image reading apparatus mounted on a copying machine or a facsimile machine, light is irradiated onto a document surface to be read, and the reflected light is converted into an analog electric signal by an optical reading means constituted by an optical conversion element such as a CCD. This analog signal is converted into a digital signal by an A / D converter, and this digital signal is subjected to various image data processing such as shading correction and γ correction, and then output as image data to a printing device or an external device. ing.

  Among such image reading apparatuses, there is a type of apparatus that scans an original by irradiating the original with light using a plurality of light sources. By irradiating light from different angles with respect to the document surface, the image data on the document can be read precisely. In addition, when reading a document by switching the monochrome reading mode or the color reading mode, it may be desirable to irradiate the document only with light of an appropriate color tone by selectively turning on the light source instead of turning it on completely. is there.

  Japanese Patent Application Laid-Open No. 10-304140 discloses an image reading apparatus that reads a document by irradiating the document with light using two light sources so that there is no shadow even if the document surface is uneven. . Here, first, one of the light sources is turned on, and the other light source is turned off to read the output value of the optical reading means comprising a CCD or the like. Next, one light source is turned off and the other light source is turned on to read the output value of the optical reading means. Then, the light amount is adjusted by changing the current of each light source or the high-speed blinking duty so that the output values of the two become predetermined values.

On the other hand, in Japanese Patent Laid-Open No. 10-233903, the output voltage of an analog signal output from the photoelectric conversion means can be decomposed by the A / D converter in order to obtain gradation according to a predetermined resolution of the A / D converter. An image input system that performs amplification processing and offset processing so as to be in a wide input voltage range is disclosed. Here, the process for converting the output voltage from the density range of the original (film) to be acquired in the photoelectric conversion device into the required digital data is performed in two stages, the output adjustment means of the photoelectric conversion device and the digital signal conversion means. Like to do.
JP-A-10-304140 JP-A-10-233903

  In a normal image reading apparatus, an analog signal output from an optical reading means is converted into a digital signal by an A / D converter, but at the previous stage, the input voltage range is decomposable by the A / D converter. Thus, the offset and gain processing for performing amplification after the offset processing of the analog signal output from the optical reading means is performed.

  However, when performing such offset and gain processing, if the light amount is adjusted by changing only the current of the light source or the fast blinking duty as in the invention disclosed in Patent Document 1, the gain value (amplification factor) is adjusted. Needs to be set to a relatively large value in consideration of individual differences in the amount of light emitted from the light source and changes over time in the amount of light. Therefore, although the light source has a sufficient light amount irradiation capability, the light amount irradiated to the document is suppressed, and the gain value is set to be large by the amount of the light amount suppression. If the image signal is amplified with a large gain value while suppressing the irradiation light amount, the noise signal is also amplified with the gain value, and high quality image data cannot be acquired.

  On the other hand, Patent Document 2 discloses that an analog signal output voltage output from a photoelectric conversion means can be decomposed by an A / D converter in order to obtain gradation according to a predetermined resolution of the A / D converter. Although amplification processing and offset processing are disclosed so as to be in the voltage range, amplification processing and offset processing when a plurality of light sources are selectively turned on and irradiated onto the document surface are not applicable.

  The present invention suppresses a noise signal by using a small gain value so as to irradiate the maximum amount of light in various lighting patterns of a plurality of light sources, and is affected by variations in characteristics of optical reading means such as a CCD. An object of the present invention is to provide an image reading apparatus and an image reading method capable of adjusting the light emission amount of each light source so that an original image can be read at the maximum resolution without any problem.

  For this reason, the present application relates to N light sources each capable of adjusting the amount of light for irradiating light on a document or a reference plate, optical reading means for converting reflected light from the document or reference plate into an analog signal, and gaining the gain of the analog signal. Adjustment means for processing, an A / D converter for converting the gain-processed analog signal into a digital signal, and control means for setting a gain value for the adjustment means, The N light sources are lit alone to read and scan the reference plate, and the input signal input to the A / D converter is between the maximum input level and the minimum input level of the A / D converter and N gain values for each light source that are approximately [1 / N] times the level near the maximum input level are acquired, and one of the N gain values selected from the N gain values for the adjustment unit is acquired. Before setting the gain value There is provided an image reading apparatus and performs image reading of the document while the light amount adjustment of the light source other than the light source according to the selected gain value.

  As a result, the image reading apparatus can irradiate the maximum amount of light in any lighting pattern having two or more light sources, and can output a signal at a predetermined level (A / D converter) with a smaller gain value. In this way, it is possible to obtain a high-quality image with less noise.

  As one aspect thereof, the selected one gain value is a maximum gain value among the N gain values.

  The image reading apparatus includes: a gain value supply register that supplies the gain value to the adjustment unit; and outputs of a plurality of photoelectric conversion elements that constitute the optical reading unit that is output from the A / D converter. A comparison means for comparing the respective values of the digital signal corresponding to the value with a predetermined threshold, a threshold supply register for supplying the predetermined threshold to the comparison means, and a comparison result in the comparison means Counting means for counting, and the control means obtains a gain value for each light source by performing various gain value settings in the gain value supply register in accordance with the count number of the counting means. It is possible.

  For this reason, the threshold value is set to a level between the maximum input level and the minimum input level of the A / D converter and approximately [1 / N] times the level near the maximum input level.

  The present application further includes N light sources each capable of adjusting the amount of light for irradiating light on the document, an optical reading unit that converts reflected light from the document into an analog signal, an adjustment unit that performs gain processing on the analog signal, An image reading method in an image reading apparatus comprising an A / D converter for converting a gain-processed analog signal into a digital signal, wherein (a) the N light sources are lit alone at a maximum output ratio. An input signal that scans the reference plate and is input to the A / D converter is approximately between a maximum input level and a minimum input level of the A / D converter and a level in the vicinity of the maximum input level. (1) acquiring N gain values of each light source that is 1 / N] (step S201, step S203), (b) selecting a maximum gain value from the acquired N gain values; (Steps S205) and (c) A light source other than the light source having the maximum gain value is individually turned on to read and scan the reference plate, and the value of the input signal of the A / D converter is converted to the A / D converter. Determining the lighting time of each light source between the maximum input level and the minimum input level of the device and approximately [1 / N] times the level near the maximum input level (step S1004), An image reading method is provided, in which a document reading process is performed in a state where the gain value of the gain adjusting means is set to the maximum gain value.

  Here, in the step (a), the gain value relating to each light source is acquired by the digital signal corresponding to the output values of the plurality of photoelectric conversion elements constituting the optical reading means output from the A / D converter. Comparison between each of the values and a threshold value set between the maximum input level and the minimum input level of the A / D converter and a level approximately [1 / N] times the level near the maximum input level. This is possible by performing.

  The present invention realizes suppression of a noise signal using a small gain value so as to perform maximum light amount irradiation in various lighting patterns of a plurality of light sources. Furthermore, the light emission amount of each light source can be adjusted so as to obtain a high gradation original image without being affected by variations in characteristics of optical reading means such as a CCD.

  FIG. 1 shows an example of a system that describes an image reading apparatus according to the present invention and a printer that receives and prints image data from the image reading apparatus.

  The image reading apparatus 1 is directly connected to the printer 2 and used. The printer 2 is an ink jet printer and includes an operation unit 200, a power switch 204, a copy start button 204a, and a sheet discharge unit 203 on the front surface. The operation unit 200 includes a display panel 201 and operation buttons 202a and 202b, and resolution (600 dpi / 300 dpi) and color / monochrome are selected by the operation buttons 202a and 202b.

  A signal corresponding to a reading condition such as a mode set by the operation unit 200 is transmitted to the image reading apparatus 1 via the cable 205, and the image reading apparatus 1 performs reading according to the signal. An image signal generated by the image reading apparatus 1 is transferred to the printer 2 via the cable 205, and a sheet such as a postcard on which an image is formed is output from the discharge unit 203.

  When the image reading apparatus 1 is supplied with power from the printer 2 and the printer 2 is activated by the power switch 204, the image reading apparatus 1 is also activated.

  FIG. 2 shows an example of the internal configuration of the image reading apparatus.

  The image reading apparatus 1 includes a platen glass 107 placed on the upper surface opening of the resin frame 100 and placing an original, and an image sensor unit 108 that can move along the platen glass 107. A pressure plate 101 that can be opened and closed is attached to hold the document placed on the upper surface. (See FIG. 1) Further, a white reference plate 109c serving as a reference when gain adjustment or shading correction is attached to the upper surface of the platen glass 107.

  The movable image sensor unit 108 can be moved in the left-right direction (sub-scanning direction) of the apparatus along the guide shaft 103c, and can be moved by a timing belt 103a, a drive pulley 103b, a motor 309 (see FIG. 5), and the like. It can be moved to a position. At this time, the image sensor unit 108 is supported by the guide shaft 103c via the carriage 103d and is urged upward by the spring 103e. A spacer 108 a is interposed between the image sensor unit 108 and the document table glass 107. The image sensor unit 108 performs scanning by moving a predetermined area for reading a document and the white reference plate 109c at a constant speed.

  FIG. 3 is a perspective view of an image sensor unit in which a light source and an optical reading unit constituting the image reading apparatus are integrally formed, and FIG. 4 is a longitudinal sectional view of the image sensor unit shown in FIG. .

  3 and 4, the image sensor unit 108 includes a LED 10 serving as a light emitting element as a light source, and a light guide 11 that guides light emitted from the LED 10 to a document, and includes a SELFOC lens that constitutes an imaging optical system. A pair of light guide light sources is provided along both sides of the array 12. An optical reading means 13 (sensor) in which about 2000 photoelectric conversion elements are arranged in a line is arranged directly under the SELFOC lens array 12, and each image sensor constituent member is arranged in the frame body 14. Composed.

  The LED 10 is fixed to one end of the light guide 11 in the longitudinal direction. In the illustrated example, three LEDs are respectively provided at one end of one light guide 11 and the other end of the other light guide 11. R, G, B, a total of six (R1, R2, G1, G2, B1, B2) are provided. Thus, it is provided on the opposite side between the two light guides 11 and has a point-symmetric arrangement with respect to the central axis C. The LED can be controlled to be turned on and off at an independent timing for each color.

  The light emitted from each LED 10 travels while being repeatedly reflected in each light guide 11, and is emitted from the entire length of the light guide 4. The light emitted from the light guide 11 is applied to the original 107 a on the original table glass 107 as shown in FIG. 4, and each reflected light enters the optical reading means 13 through the SELFOC lens array 12.

  In this way, the image sensor unit 108 reads the document image by irradiating the document with light and causing the reflected light from the document to enter the sensor 13 via the imaging optical system.

  FIG. 5 is a block diagram showing a circuit configuration of the image reading apparatus 1.

  A CPU 300 (control means) that controls the entire image reading apparatus 1 is connected to a sensor unit driving circuit 310, a motor driving circuit 304, and a signal processing unit that drive the image sensor unit 108 including the sensor 13 and the LED 10 through the bus line 301. 305 and the interface 306 are connected. A ROM 307 and a RAM 308 are connected to the CPU 300 via a bus line 301. The ROM 307 is approximated from various control programs such as a lighting control program for the LED 10 and a drive control program for the sensor 13, and gain values described later. An approximate expression for obtaining an offset value is stored, and the RAM 308 is used to temporarily store data when processing image data.

  In the signal processing unit 305, various images such as gain / offset processing of an analog image signal output from the sensor 13, conversion from an analog signal to a digital signal by an A / D converter, shading correction of a digitized image signal, and the like. Processing is performed. The image signal subjected to image processing is connected to the interface 306.

  The interface 306 exchanges data and signals with the printer 2 such as output of the image data subjected to image processing by the signal processing unit 305 to the outside and input of signals according to the reading mode from the printer 2. In accordance with a standard standard such as SCSI or USB, the printer 2 is connected.

  The motor driving circuit 304 drives the motor 309 and moves the image sensor unit 108 in the sub-scanning direction at a predetermined speed according to the reading mode output from the printer 2.

  FIG. 6 shows details of the sensor unit drive circuit 310 shown in FIG.

  The sensor unit drive circuit 310 includes an LED drive circuit 303 and a sensor drive circuit 304. Here, the sensor driving circuit 304 outputs a start pulse SP generated at a cycle corresponding to the clock CLK and the reading mode to the sensor 13 and the LED driving circuit 303. Further, the LED drive circuit 303 is lit by a start pulse SP, a counter 303a that counts the clock CLK simultaneously with the input of the start pulse SP, a counter 303b that counts the clock CLK counted by the counter 303a every eight cycles, and the like. And a register 303c for setting the lighting time of each LED (all LEDs R1 to B1).

  As shown in FIG. 7 described later, the LED is turned on simultaneously with the input of the start pulse SP, and at the same time, the counter 303a starts counting the clock CLK. The counter 303b counts the clock CLK counted by the counter 303a in units of 8 cycles, and the LED is turned off when the counted number reaches the lighting clock number Con set in the register 303c. In this embodiment, there are six LEDs, and the register 303c has six systems so that each can independently control the lighting time. For simplification of description, only one LED will be described.

  Here, (accumulation time T) / (clock CLK8 period) becomes the maximum number of lighting clocks (maximum lighting time) “Cmax” in which the LED can be lit within the accumulation time T, and the number of lighting clocks Con set in the register 303c is , 0 to Cmax can be set, whereby the lighting time can be changed.

  FIG. 7 is a diagram for explaining the lighting output ratio P of the light source (LED).

  The output ratio P of the LED lighting time Ton / accumulation time T is represented by the ratio of the number of lighting clocks Con set in the register 303c and the maximum number of lighting clocks Cmax, “number of lighting clocks Con / maximum number of lighting clocks Cmax”. . That is, in the case of “Con = Cmax”, the output ratio P is Con / Cmax = 1 (100%), the LED lighting time Ton = accumulation time T, and in the case of “lighting clock number Con = 1/2 Cmax”, The output ratio P is (Cmax / 2) /Cmax=0.5 (50%), and the LED lighting time Ton = 1/2 accumulation time T. When the lighting clock number Con is 0 (minimum lighting clock number), the LED is not lit.

  FIG. 8 is a block diagram showing an internal configuration of the signal processing unit 305 shown in FIG.

  Image light reflected from the document or the white reference plate 109c is converted into an analog signal by the sensor 13 and input to the signal processing unit 305 as an analog image signal.

  The signal processing unit 305 includes a sample hold circuit 400 that holds an analog image signal, an adjustment circuit 401 (adjustment unit) that performs gain processing (amplification) and offset processing (shift processing) on the analog image signal, and converts the analog image signal into a digital image signal. And an A / D converter 402 for converting the digital image signal, and an image processing unit 403 for performing image processing such as shading correction and γ correction on the digital image signal.

  The adjustment circuit is composed of, for example, an operational amplifier circuit, and is connected to an offset value supply first register 404 for setting an offset value and a gain value supply second register 405 for setting a gain value. The analog image signal input via 400 is subjected to offset processing according to the offset value and gain processing according to the gain value.

  The A / D converter has a resolvable input voltage range of, for example, DC 0 to 5 V, and a resolution of 10 bits (0 to 1023). The A / D converter 402 is connected to a comparison circuit 406 (comparison means), and the comparison circuit 406 is connected to a count circuit 407 (counting means).

  The comparison circuit 406 compares, for each pixel, digital image data output from the A / D converter 402 and a threshold described later set in the threshold supply third register 408 during offset adjustment and gain adjustment. . The count circuit 407 (counting unit) counts the number of pixels that are equal to or greater than or less than the threshold value in the comparison circuit 406.

  The offset value can be set with 8 bits (0 to 255), the gain value can be set with 6 bits (0 to 63), and when the gain setting value is 0, the amplification factor is about 1 time and the gain When the set value is 255, the amplification factor is about 10 times.

  FIG. 9 shows various reading modes and a timing chart of each reading mode.

  As described above, resolution (600/300 dpi) and color / monochrome are set by the printer 2. Accordingly, the reading mode has the following five reading modes at a resolution of 600 and a resolution of 300 dpi, for a total of ten reading modes. (1) Color / LED sequential lighting, (2) Monochrome / LED full color lighting, (3) Monochrome / red LED lighting (R1, R2), (4) Monochrome / green LED (G1, G2) lighting, ( 5) Reading mode for each lighting pattern of monochrome / blue LED (B1, B2) lighting.

  Here, as in (2) above, the “monochrome / LED all-color lighting mode” is for preventing a so-called dropout color in which a color that is not read is generated depending on the color of the light source.

  In addition, as described in the above (3) to (5), the “monochrome / monochromatic LED lighting mode” is, for example, for reading a document in which black characters are written on a white background and partially marked with a red highlighter. When the original is scanned with only the red LEDs (R1, R2) turned on, the mark portion has an output value (high output value) equivalent to that of the white background portion, and only the black character portion is not detected. Is detected as a low output value. Thus, the “monochrome / LED single color lighting mode” is intended to prevent a predetermined color from being intentionally detected as a dropout color.

  As shown in FIG. 9A, in the above-described (1) color / LED sequential lighting mode, each color (R1, R2), (G1, G2) of each LED 10 during one line scanning indicated by the accumulation time T. , (B1, B2) are sequentially turned on, the sensor 13 is driven in synchronization with this, and the image sensor unit 108 is moved in the sub-scanning direction to perform scanning.

  As shown in FIG. 9B, in the case of the above-described (2) monochrome / LED all color lighting mode, accumulation is performed by the sensor 13 with the LEDs 10, R1, R2, G1, G2, B1, B2 being lit simultaneously. One line indicated by time T is scanned.

  As shown in FIG. 9C, in the case of the above-described resolution monochrome / color single color mode of (3) to (5), only the single color LEDs are lit (for example, in the case of red single color, only R1 and R2 are lit). 1 line is scanned by the sensor 13 as indicated by the accumulation time T.

  FIG. 10 shows the relationship between the offset value and gain value obtained when the image reading apparatus is powered on.

  FIG. 11 illustrates offset processing and gain processing when performing offset processing for uniformly adding a constant value to the output signal of each pixel constituting the optical reading means and gain processing for amplification.

  As shown in FIG. 10, when the white reference plate 109c is scanned with the LED 10 turned on, the digital signal of each pixel output from the A / D converter 402 is output from the A / D converter. Each gain value that is a value in the vicinity of the highest level and the digital signal of each pixel output from the A / D converter 402 when reading scanning is performed with the LED 10 turned off are output from the A / D converter. Each offset value that is a value near the lowest level 0 has a substantially linear relationship, and the offset value is represented by a linear function with the gain value as a variable. Therefore, if the offset values for the two gain values are known, the offset value for an arbitrary gain value can be calculated using the following approximate expression.

  Offset value (approximate value) Of '= ((Of1-Of2) / (Ga1-Ga2)) * Ga + (Ga1 * Of2-Ga2 * Of1) / (Ga1-Ga2), where Of1 is the first offset value. Of2 indicates the second offset value.

  The approximate expression F is stored in the ROM 307 (FIG. 5). Further, when the approximate expression F is represented by an inclination A and an intercept B, Of ′ = A * Ga + B, and A = (Of1-Of2) / (Ga1-Ga2), B = (Ga1 * Of2-Ga2 * Of1) / (Ga1-Ga2). Since the constants A and B of the approximate expression F change depending on the state of the light source and the like, by obtaining each constant for each reading mode before reading the document, the offset value for each gain value can be approximated in a constantly stable state. Can be requested.

  In the present embodiment, the offset value Of1 when the maximum gain value Ga (max) 63 (the amplification factor at this time is about 10 times) is the first gain value Ga1, and the minimum gain value Ga (min) 0 (at this time) Since the slope is obtained using the offset value Of2 when the gain is about 1), the first gain value Ga1-the second gain value Ga2 is set to a fixed value (63). Stored in the ROM 307 in advance. At this time, the intercept B is equal to Of2.

  FIG. 12 is an example of a flowchart illustrating the entire pre-reading operation of the image reading apparatus.

  When the operator starts up the printer 2 by pressing the power switch 204, an initial signal is transmitted from the printer 2 and the image reading apparatus 1 starts up. After that, the user designates the reading mode using the operation buttons 202a and 202b, and finally presses the copy button 204a to determine the gain value Ga and the offset value Of in the designated mode before reading the document and use them. The pre-reading process for adjusting the light quantity of the LED to be started is started.

  As shown in FIG. 12, in the pre-reading process, offset adjustment is first performed to obtain an offset value Of1 at the maximum gain value Ga (max) and an offset value Of2 at the minimum gain value Ga (min) ( Steps S1000 and S1001), and the difference between them “Of1−Of2” is obtained (step S1002). Thereby, the slope A of the approximate expression F described above is determined. At the same time, by obtaining the intercept B “Of2”, the approximate expression F is established, and the approximate offset value corresponding to the gain value can be obtained from each gain value using this approximate expression.

  Next, the gain is adjusted while obtaining the offset value Of with respect to the gain value Ga using the determined approximate expression F (step S1003), and finally the light amount of each LED is adjusted (step S1004). finish. In this embodiment, only the red LEDs (R1, R2) are lit to perform reading, and (3) monochrome / red LED (R1, R2) mode will be described as an example.

  FIG. 13 shows an example of a flowchart for obtaining the offset value Of1 at the maximum gain value Ga (max) and the offset value Of2 at the minimum gain value Ga (min).

  In step S100, the maximum gain value Ga (max) is set in the second register 405. In this embodiment, since the gain value is 6 bits (0 to 63), 63 is set, and the amplification factor at this time is 10. In step S101, a maximum offset value Of (max) and a minimum offset value Of (min) are set. In this embodiment, since the offset value is 8 bits (0 to 255), Of (max) = 255 and Of (min) = 0 are set. In step S102, an offset adjustment threshold L1 is set in the third register 408.

  As shown in FIG. 11, this threshold value L1 considers the output variation of each pixel of the sensor 13 and performs scanning with the light source turned off, and the signal output from each pixel of the A / D converter 402 A value set so as not to be below the minimum output value 0 of the A / D converter 402 and close to the minimum output value 0 (the voltage of each pixel input to the A / D converter 402 is A / D In this embodiment, the A / D converter 402 (10) is set to a value that is not lower than the minimum input voltage (minimum input level) 0 V of the D converter and is close to the minimum input value 0. Bit) output value 50, which corresponds to about 245 mV near the minimum input level 0 V of the A / D converter 402.

  In step S103, an intermediate offset value Of (mid) between the maximum offset value Of (max) and the minimum offset value Of (min) is set in the first register 404.

  Next, in step S104, the motor 309 is driven via the motor drive circuit 304 to move the image sensor unit 108 to below the white reference plate 109c, and then the sensor 13 is driven with the LED 10 turned off. The white reference plate 109c is scanned for one line. The analog image signal output from the sensor 13 by this reading scanning is input to the adjustment circuit 401 through the sample hold circuit 400 for each pixel. The analog image signal is subjected to gain processing (amplification) using the maximum gain value Ga (max) and offset processing (level shift) using the offset value Of (mid) in the adjustment circuit 401, and then converted into a digital signal in the A / D conversion circuit 402. And sequentially input to the comparison circuit 406.

  In the comparison circuit 406, the comparison with the offset adjustment threshold L1 set in the third register 408 is performed for each pixel, and the number of pixels that are equal to or less than the offset adjustment threshold L1 is counted by the count circuit 407 (step S105). ).

  Next, a difference [Of (max) −Of (min)] between the maximum offset value Of (max) and the minimum offset value Of (min) is obtained, and it is determined whether or not it is 1 or less (step S106). In the case of 1 or less, the offset value Of (mid) is stored in the predetermined area of the RAM 308 as the first offset value Of1 (step S110).

  If [Of (max) −Of (min)] is larger than 1 in step S106, the number of pixels M obtained in step S105 and the predetermined value D stored in the ROM 307 (in the present embodiment, about 2000 photoelectric conversion elements). (Step S107), and if the number of pixels M is equal to or greater than the predetermined value D, the minimum offset value Of (min) is rewritten to the offset value Of (mid), and the process returns to step S103 ( Step S108).

  When the number of pixels M is less than the predetermined value D in step S107, the maximum offset value Of (max) is rewritten to the offset value Of (mid), and the process returns to step S103 (step S109). In this way, steps S103 to S109 are repeated until [Of (max) −Of (min)] becomes 1 or less, and the offset value Of1 at the maximum gain value is obtained.

  As described above, the offset value is changed by the binary search method, and when 40 pixels among the 2000 photoelectric conversion elements become the threshold value L1 or more, that is, most of the analog signals output from the photoelectric conversion elements are the threshold value L1 or more. The offset value when the value is in the vicinity of the threshold value L1 is set as the first offset value Of1.

  Next, an offset value Of2 at the time of the minimum gain value Ga (min) is obtained (step S1001). Similarly to the case where the offset value Of1 at the time of the maximum gain value Ga (max) is obtained, the offset value Of2 at the time of the minimum gain value Ga (min) is obtained. That is, the minimum gain value Ga (min) (0 in this embodiment) is set in the second register 405 (step S100), and the second offset value Of2 is stored in the RAM 308 by performing steps S102 to S109. (Step S110).

  Next, a difference [Of1−Of2] between the first offset value Of1 and the second offset value Of2 is obtained and stored in a predetermined area of the RAM 308, and the first offset value Of2 is determined as a intercept B of the approximate expression. It is stored in the area (step S1002).

Thus, the approximate expression F in the main reading mode is determined.
Next, a description will be given of the content of performing the light amount adjustment of the light source using the gain value Ga, which is the main part of the invention of the present application.

  First, in step S1003 of FIG. 12, the gain value Ga and the offset value Of in this reading mode are obtained using the determined approximate expression F.

  FIG. 14 shows a flowchart for the method of acquiring the gain value Ga and the offset value Of.

  In order to obtain the gain value Ga and the offset value Of, first, a gain value Gar1 when only one of the red LEDs R1 is lit at an output ratio of 100% (maximum output ratio) is obtained (step S201). An offset value Ofr1 when the value is Gar1 is obtained (step 202).

  Next, the gain value Gar2 is obtained when only the red LED R2 is lit at an output ratio of 100% (maximum output ratio) (step S203), and the offset value Ofr2 at the gain value Gar2 is obtained. The gain values Gar1 and Gar2 obtained in step 201 and step 203 are compared (step S205), the larger gain value is set as the gain value Ga, and the offset value is set as the offset value Of (step 206, step). 206 ′).

  In the case of monochrome / LED all lighting (6 lights) mode, the gain value when each LED (R1 to B2) is turned on at 100% is obtained, and the largest gain value among the six gain values is obtained. Set as gain value Ga.

  FIG. 15 shows a flowchart for obtaining the gain value Gar1 (Gar2) when the LEDR1 (R2) output ratio is 100% illuminated.

  In FIG. 15, only R1 is turned on at an output ratio of 100%, the reference plate 109c is read and scanned, gain adjustment is performed while changing the gain value by the binary search method to obtain the gain value Gar1, and then offset adjustment is performed. To obtain the offset value Ofr1.

  First, in step S300, a value 1/2 that is 1/2 of the gain adjustment threshold L2 is set in the third register 408. As shown in FIG. 11, the threshold value L2 is set to 973 between the maximum output value (1023) and the minimum output value (0) of the A / D converter 402 and in the vicinity of the maximum output value. Yes. The threshold L2 corresponds to a value between the maximum input level 5V and the minimum input level 0V of the A / D converter 402 and in the vicinity of the maximum input level 5V, 4.8V.

  This gain adjustment is a gain value such that the analog signal input to the A / D converter 402 is at a level near the threshold L2 when R1 and R2 are turned on simultaneously and scanning of the reference plate 109c is performed. This reading mode is monochrome / monochromatic mode (red), since two lights R1 and R2 are turned on at the same time. Therefore, the input voltage and output value of the A / D converter 402 for each of R1 and R2 are turned on twice. Since it is half of the hour, the threshold is set to ½ L2 which is half of L2. In the case of monochrome / LED all lighting (6 lights) mode, the input voltage and output value of the A / D converter 402 of each LED (R1 to B2) are 1/6 of all lighting, so the threshold value is Set to 1/6 and 1 / 6L2 of L2. In the color / LED sequential lighting mode, the LEDs of the respective colors are sequentially turned on, so that two lights are turned on at the same time, and the threshold value is set to 1 / 2L2, which is half of L2.

  In step S301, a maximum gain value Ga (max) and a minimum gain value Ga (min) are set. In this embodiment, since the gain value is 6 bits (0 to 63), Ga (max) = 63 and Ga (min) = 0 are set. In step S302, an intermediate gain value Ga (mid) between the maximum gain value Ga (max) and the minimum gain value Ga (min) is set in the second register 405. In step S 303, an offset value Of ′ for the gain value Ga (mid) is obtained from the approximate expression F, and this is set in the first register 404.

  Next, in step S304, the motor 309 is driven via the motor drive circuit 304, and the image sensor unit 108 is moved below the white reference plate 109c. The “maximum lighting clock number Cmax” is set in the register 303c of the LED driving circuit 303, and the sensor 13 is driven in a state where only R1 is turned on at a 100% output ratio, and the reading scanning of the white reference plate 109c is performed for one line. Do.

  The analog image signal output from the sensor 13 by this reading scanning is input to the adjustment circuit 401 through the sample hold circuit 400 for each pixel. The adjustment circuit 401 performs gain processing (amplification) using the gain value Ga (mid) and offset processing (level shift) using the offset value Of ′, and then converts the digital signal to the comparison circuit 406 by the A / D conversion circuit 402. Input sequentially.

  In the comparison circuit 406, the comparison with the threshold value 1 / 2L2 set in the third register 408 is performed for each pixel, and the number T of pixels having the threshold value 1 / 2L2 or more is counted by the count circuit 407 (step S305). .

  Next, a difference [Ga (max) −Ga (min)] between the maximum gain value Ga (max) and the minimum gain value Ga (min) is obtained, and it is determined whether or not it is 1 or less (step S306). In the case of 1 or less, the gain value Ga (mid) is stored in the predetermined area of the RAM 308 as the gain value Ga1 (step S111).

  If [Ga (max) −Ga (min)] is greater than 1 in step S307, the number T of pixels obtained in step S306 is compared with the predetermined value D (40 in this embodiment) stored in the ROM 307. (Step S307), and when the number of pixels T is equal to or greater than the predetermined value D, the maximum gain value Ga (max) is rewritten to the gain value Ga (mid), and the process returns to step S302 (step S309).

  When the number of pixels T is less than the predetermined value D in step S307, the minimum gain value Ga (min) is rewritten to the gain value Ga (mid), and the process returns to step S302 (step S308). In this way, steps S302 to S309 are repeated until [Ga (max) −Ga (min)] becomes 1 or less to obtain the gain value Gar1.

  As described above, the gain value is changed by the binary search method, and the analog signal of 40 or more pixels in 2000 photoelectric conversion elements becomes the level of the threshold value 1 / 2L2 or more, that is, output from the optical reading unit. Gain value when the analog signal is at a level in the vicinity of the threshold value 1 / 2L2 (when most of the signals output from the photoelectric conversion elements are equal to or less than the threshold value 1 / 2L2 and in the vicinity of the threshold value 1 / 2L2). Is the first gain value Gar1.

  Next, offset adjustment for obtaining the offset value Ofr1 at the gain value Gar1 is performed (step S202). This is performed by the method shown in FIG. 13, and the offset value Ofr1 can be obtained by setting the gain value Gar1 obtained in step S206 in the second register 405 in step S100 and performing steps S102 to S110. .

  Next, the gain value Gar2 is obtained when only the other red LED, R2, is lit at an output ratio of 100%. In this case as well, the gain value Gar2 is obtained in the same manner as described above, and only R2 is obtained with an output ratio of 100%. This is done by turning on and performing steps S300 to S310. (Step S203 in FIG. 14) The offset value Ofr2 of the gain value Gar2 is also obtained by the same method as the offset value Ofr1 (Step S204 in FIG. 14).

  In step S205, the difference between the gain values Gar1 and Gar2 thus obtained, "Gar1-Gar2" is obtained. If this is greater than or equal to 0, Gar1 is the gain value Ga for the main reading mode, and Ofr1 is the offset value for the main reading mode. If stored in the RAM 307 as Of, if less than 0 (negative), Gar2 is stored in the RAM 307 as the gain value Ga in the main reading mode and Ofr1 is stored in the RAM 307 as the offset value Of in the main reading mode (step S206, step S206 ').

  In other words, each light source is lit alone at an output ratio of 100% and the white reference plate is scanned to obtain the gain value of each light source, and the maximum gain value among the gain values for each light source is selected as the gain value Ga. is there. In other words, each light source is individually turned on at an output ratio of 100%, the reference plate is scanned, and the gain value of the light source having the smallest light emission amount is selected as the gain value Ga.

  In the present embodiment, the description will be continued assuming that Gar1 is set as the gain value Ga and Ofr1 is set as the offset value Of. In the case of monochrome / LED all lighting (6 lights) mode, each LED is turned on at a 100% output ratio to obtain each gain value, and the largest gain value among these six gain values is gained. Set as value Ga. That is, the gain value at the time of the LED having the smallest light emission amount with 100% output ratio lighting among the six is adopted as the gain value Ga.

  FIG. 16 shows a flowchart for light amount adjustment.

  The light amount adjustment is performed by turning on the light source (light source other than the light source for which the gain value is selected) R2 having the larger light emission amount, scanning the white reference plate, and performing offset / gain processing using the offset value Of and the gain value Ga. The R2 lighting time (the number of lighting clocks Con) is obtained so that the input signal to the A / D converter at the time of performing the above operation is at a level in the vicinity of the threshold value 1 / 2L2. This is done by changing the method.

  First, the threshold value 1 / 2L2 is set in the third register 408 (step S500), the gain value Ga in the main reading mode obtained in step S206 is set in the second register 405 (step S501), and the main value obtained in step S206. The offset value Of in the reading mode is set in the first register 404 (step S502). In step S503, the maximum lighting clock number Con (max) and the minimum lighting clock number Con (min) are set.

  Next, in step S504, an intermediate clock number Con (mid) between the minimum lighting clock number Con (min) and the maximum lighting clock number Con (max) is set in the register 303c of the LED drive circuit 303. In step 505, R2 is turned on at an output ratio (Con (mid) / Cmax) × 100% (lighting time Con (mid)), and the sensor 13 scans the white reference plate 109c.

  The analog image signal output from the sensor 13 by this reading scanning is input to the adjustment circuit 401 through the sample hold circuit 400 for each pixel. The adjustment circuit 401 performs gain processing (amplification) using the gain value Ga and offset processing (level shift) using the offset value Of, and then converts the digital signal to the A / D converter 402 and sequentially inputs the digital signal to the comparison circuit 406. .

  In the comparison circuit 406, the comparison with the threshold value 1 / 2L2 set in the third register 408 is performed for each pixel, and the number of pixels T that is equal to or greater than the threshold value 1 / 2L2 is counted by the count circuit 407 (step S507). .

  Next, a difference [Con (max)] − Con (min)] between the maximum lighting clock number Con (max) and the minimum lighting clock number Con (min) is obtained, and it is determined whether or not this is 1 or less ( Step S508) If it is 1 or less, Con (mid) is stored in the predetermined area of the RAM 308 as the number of lighting clocks Con of R2 (Step S512).

  If [Con (max) −Con (min)] is greater than 1 in step S508, the number of pixels T obtained in step S507 and a predetermined value D (40 in this embodiment) stored in the ROM 307 are obtained. (Step S509), and if the number of pixels T is equal to or greater than the predetermined value D, Con (max) is rewritten to Con (mid), and the process returns to Step S504 (Step S510).

  If the number of pixels T is less than the predetermined value D in step S509, Con (min) is rewritten to Con (mid), and the process returns to step S504 (step S511).

  In this way, Steps S504 to S511 are repeated until [Con (max) −Con (min)] becomes 1 or less, and the number of lighting clocks (lighting time) of R2 is obtained.

  In the case of the monochrome / full LED (6 lights) mode, the same light amount adjustment is performed for the five LEDs other than the LED having the smallest light emission amount, but the threshold at this time is set to 1 / 6L2. This completes the pre-reading process in the main reading mode and the monochrome / red lighting mode.

  FIG. 17 shows an example of a timing chart during document reading. Based on the conditions set by the pre-reading process, the original is read in the following procedure.

  By placing an original on the platen 107 and pressing the operation button 202 of the operation unit 200 of the printer 2, a reading start command is transmitted from the printer 2 to the image reading apparatus 1. The offset value Of is set in the first register 404, the gain value Ga is set in the second register 405, and as shown in FIG. 17, R1 is a 100% output ratio, and R2 is the output ratio (Con / The lighting clock number Cmax (for R1) and the lighting clock number Con (for R2) are set in the register 303c so that the lighting is performed at (Cmax) × 100%. Then, each of the image sensor units 108 is moved by the length of the document in the sub-scanning direction (left-right direction in FIG. 2) while R1 and R2 are turned on at the lighting time corresponding to the number of lighting clocks and scanning is performed line by line. Then, the reading scan is finished. In this way, the light quantity of R2 is adjusted by turning on R2 with the lighting time Con determined in step S1004.

  The analog image signal output pixel by pixel from the sensor 13 is input to the adjustment circuit 401 via the sample hold circuit 400 and is gain / level shifted by the gain value Ga and the offset value Of. Next, it is converted into a digital signal by the A / D conversion circuit 402, subjected to various image processing such as shading correction and γ correction in the image processing unit 403, and then output to the printer 2 as image data.

  In this embodiment, the lighting time of the light source (R2) other than the light source related to the maximum gain value is obtained by the method shown in step S1004. However, the gain value (maximum gain value) of R1 obtained in step S201 is obtained. The lighting time Con of R2 may be obtained from the ratio with the gain value of R2 obtained in step S203. In this case as well, it is possible to obtain a lighting time equivalent to the lighting time of R2 obtained in step S1004, and it is possible to further shorten the time.

  In the above-described embodiment of the present invention, the image reading apparatus and the printer that receives and prints image data from the image reading apparatus have been described in detail. However, the present invention is not limited to an image mounted on a copying machine or a facsimile apparatus. Needless to say, the present invention is applicable to reading devices in general.

  The present invention relates to an image reading apparatus and an image reading method for optically reading a document, and has industrial applicability.

1 shows an example of a system that describes an image reading apparatus according to the present invention and a printer that receives and prints image data from the image reading apparatus. 2 shows an example of the internal configuration of the image reading apparatus. FIG. 3 is a perspective view of an optical reading unit in which a light source and an optical reading unit constituting the image reading apparatus are integrally formed. FIG. 4 is a longitudinal sectional view of the optical reading unit shown in FIG. 3. 2 is a block diagram showing a circuit configuration of the image reading apparatus 1. FIG. 6 shows details of the sensor unit drive circuit 310 shown in FIG. It is a figure explaining the lighting output ratio P of a light source (LED). FIG. 6 is a block diagram illustrating an internal configuration of a signal processing unit 305 illustrated in FIG. 5. The timing chart of various reading modes and each reading mode is shown. The relationship between the offset value and the gain value obtained when the image reading apparatus is powered on is shown. It is a figure explaining the offset process and gain process in the case of performing the offset process which adds a fixed value uniformly to the output signal of each pixel which comprises a CCD optical reading means. 3 is an example of a flowchart illustrating an entire read preprocessing operation when the image reading apparatus is powered on. The example of the flowchart which calculates | requires offset value OF1 in the case of the maximum gain value GA (max) and offset value OF2 in the case of the minimum gain value GA (min) is shown. 5 shows a flowchart for a method for obtaining a gain value GA and an offset value OF. The flowchart for gain value GA1 (GA2) acquisition at the time of LEDR1 (R2) 100% lighting is shown. The flowchart for light quantity adjustment is shown. An example of a timing chart at the time of document reading is shown.

Explanation of symbols

1: Image reading device 2: Printer 10: Light source (LED)
12: photoelectric conversion element group constituting optical reading means 13: optical reading means 108: optical reading means unit 109c: white reference plate (reading reference plate)
300: Control means (CPU)
305: Signal processing unit (signal processing circuit)
306: Interface circuit 308: Storage means (RAM)
400: Sample hold circuit 401: Adjustment means (adjustment circuit)
402: A / D converter 403: Image processing unit 404: First register (offset value supply register)
405: Second register (gain value supply register)
406: Comparison means (comparison circuit)
407: Count means (count circuit)
408: Third register (threshold supply register)

Claims (8)

N light sources each capable of adjusting the amount of light for irradiating the original or reference plate with light
Optical reading means for converting reflected light from a document or a reference plate into an analog signal;
Adjusting means for gain processing the analog signal;
An A / D converter for converting the gain-processed analog signal into a digital signal;
Control means for setting a gain value for the adjustment means,
The control means turns on the N light sources independently, scans the reference plate, and inputs signals to the A / D converter are the maximum input level and the minimum input of the A / D converter. N gain values of each light source that are approximately [1 / N] times between the levels and near the maximum input level are acquired, and the adjustment means is selected from among the N gain values. An image reading apparatus configured to read an image of a document while setting one selected gain value and adjusting a light amount of a light source other than the light source related to the selected gain value. The image reading apparatus according to claim 1, wherein the selected one gain value is a maximum gain value among the N gain values. The light amount of a light source other than the light source related to the maximum gain value is adjusted to decrease based on a ratio between the individual gain value and the maximum gain value related to each light source. Image reading device. The image reading apparatus according to claim 1, wherein the light amount of a light source other than the light source related to the maximum gain value is adjusted by changing a lighting time. A gain value supply register for supplying the gain value to the adjustment means;
Comparing means for comparing the magnitudes of respective values of digital signals corresponding to the output values of the plurality of photoelectric conversion elements constituting the optical reading means output from the A / D converter and a predetermined threshold value;
A threshold supply register for supplying the predetermined threshold to the comparison means;
Counting means for counting the comparison results in the comparison means,
The said control means acquires the gain value for every said light source by performing various gain value setting to the said gain value supply register | resistor according to the count number of the said counting means. Image reading device. The threshold value is set to a level between a maximum input level and a minimum input level of the A / D converter and approximately [1 / N] times a level near the maximum input level. Item 6. The image reading apparatus according to Item 5. N light sources each capable of adjusting the amount of light for irradiating light on the original, optical reading means for converting reflected light from the original into an analog signal, adjustment means for gain processing of the analog signal, and the analog processed by gain processing An image reading method in an image reading apparatus comprising an A / D converter that converts a signal into a digital signal,
(A) The N light sources are lit alone at the maximum output ratio, the reference plate is read and scanned, and the input signal input to the A / D converter is the maximum input level of the A / D converter. Obtaining N gain values for each light source that are approximately [1 / N] times between the minimum input level and near the maximum input level;
(B) selecting a maximum gain value from the acquired N gain values;
(C) A light source other than the light source having the maximum gain value is individually turned on to read and scan the reference plate, and the value of the input signal of the A / D converter is the maximum input of the A / D converter. Obtaining a lighting time of each light source between a level and a minimum input level and approximately [1 / N] times a level near the maximum input level;
And reading the original in a state where the gain value of the gain adjusting means is set to the maximum gain value. In step (a), the gain value associated with each light source is obtained by outputting each of the digital signals corresponding to the output values of the plurality of photoelectric conversion elements constituting the optical reading means output from the A / D converter. A comparison is made between a value and a threshold value set between a maximum input level and a minimum input level of the A / D converter and a level approximately [1 / N] times a level near the maximum input level. The image reading method according to claim 7, wherein the image reading method is performed.

Images