JP2006166270A - Image reading device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an image at a level according to an original image without bringing down the entire machine even when one data system for reading a document is down.
A photoelectric conversion element having a plurality of light spectrum sensitivity characteristics and receiving a reflected light or transmitted light from a document to read a document image, and a read image from the photoelectric conversion element for each of the light spectrum sensitivity characteristics. Driving image data reading means for outputting as image data and one image data reading means of the plurality of image data reading means, and when one image data reading means goes down, other image data reading means Selecting means for selecting and driving.
[Selection] Figure 5

Description

  The present invention relates to an image reading apparatus for reading an image in a facsimile apparatus, a printer apparatus or the like.

  Patent Document 1 discloses an image reading apparatus that increases the amount of received light by focusing on a blue light receiving element array. Patent Document 2 discloses an image reading apparatus that retracts an ND filter from a reading optical path when an image is read by a photoelectric conversion element having low sensitivity. Patent Document 3 discloses an image reading apparatus that issues a warning when a density difference between adjacent areas is small when printing a single color after reading a multicolor original.

On the other hand, in an image reading apparatus having a plurality of light spectrum sensitivities and capable of color reading, reproduction in monochrome mode is performed by using one type of data from a plurality of light spectra. It has been broken. For example, in the three-color spectrum of red (R), green (G), and blue (B), the density is expressed by using green, which is close to the visibility characteristic, as monochrome read data. .
JP-A-9-172519 JP 2000-253207 A JP-A-6-225167

However, if there is an abnormality in the data system to be read, for example, the G reading system data, it is caused by hardware abnormalities such as CCD, analog amplifier circuit, AD converter, field memory, etc. Affects the acquisition of complex images. In other words, an abnormal image occurs and the machine is in an unusable down state, so that the unusable time (down time) of the machine until repair is prolonged, and the loss time for waiting for the arrival of maintenance workers and replacement parts is wasted. .
The present invention can acquire an image at a level according to the original image without bringing down the entire machine even when one data system for reading a document goes down. An object is to provide an image reading apparatus.

  An image reading apparatus according to a first aspect of the present invention has a plurality of optical spectrum sensitivity characteristics, receives a reflected light or transmitted light from a document and reads a document image, and a read image from the photoelectric conversion element. The image data reading means for outputting image data for each optical spectrum sensitivity characteristic and one image data reading means among the plurality of image data reading means, and when one image data reading means goes down And selecting means for selecting and driving other image data reading means.

  The invention described in claim 2 is the image reading apparatus according to claim 1, further comprising display means for displaying down when one image data reading means goes down.

  A third aspect of the present invention is the image reading apparatus according to the first or second aspect, wherein the selection means is capable of manually switching selection of other image data reading means.

  According to a fourth aspect of the present invention, in the image reading apparatus according to any one of the first to third aspects, the plurality of image data reading units read three types of image data of RGB, and normally G data. Is used to make a mono-color image, and when the G data is abnormal, either R data or B data is used as substitute data.

  According to the present invention, when one image data reading unit used for image reading goes down, the selection unit selects and drives the other image data reading unit, so that reading by the data of the other image data reading unit is performed. Is possible. For this reason, even in an abnormal state, an image can be acquired at a level according to the original image without bringing down the machine.

  FIG. 1 is an overall configuration diagram of a color image reading apparatus as an embodiment of the present invention. In the scanner main body 13 in FIG. 1, the document placed on the document table glass 8 is irradiated by the illumination lamp 2 configured integrally with the first mirror 3, and the reflected light is configured integrally with the first mirror 3. The second mirror 5 and the third mirror 4 are scanned. Thereafter, the reflected light is converged by the lens 1 and irradiated to a color CCD (see FIG. 5: 42) mounted on an SBU (Sensor Board Unit) 10 to be photoelectrically converted into RGB. In this case, the 1st mirror 3, the illumination lamp 2, the 2nd mirror 5, and the 3rd mirror 4 can move to the left-right direction by making the traveling body motor 9 into a drive source.

  When an automatic document feeder (ADF) 12 is used, when the document passes through the reading position between the DF document glass 6 and the reflection guide plate 20 for both reading of the front surface and the back surface, the reading position The reflected light is scanned by the first mirror 3 and the second mirror 5 and the third mirror 4 that are integrally formed. Thereafter, the reflected light is focused by the lens 1 and irradiated to the SBU 10 on which the color CCD is mounted, thereby being photoelectrically converted to RGB. The SCU 7 that controls the operation of the color image reading apparatus including the main body and the ADF is mounted on the scanner main body 13 described above.

  2 is an overall block diagram of the image reading apparatus of this embodiment, FIG. 3 is an image data flow of the image reading apparatus, and FIGS. 4 and 5 are block diagrams relating to image control. The operation of image reading will be described.

  The reflected light of the original entering the color CCD on the SBU 10 is converted into analog signals of RGB colors having voltage values corresponding to the light intensity in the color CCD. The analog signals for each color of RGB are divided into odd bits and even bits and output.

  The analog image signal of the SBU 10 is input to the A / D converter 33 after the dark potential portion is removed by the analog processing circuit 32 on the VIOB 31, the odd bits and the even bits are combined, and the gain is adjusted to a predetermined amplitude. Signaled.

  The digitized image signal is subjected to shading correction by the shading ASIC 34, subjected to image processing such as gamma correction and MTF correction by the RIPU 35 on the SCU from the VIOB through the SCU 7, and then output as a video signal together with the synchronization signal and the image clock. The

  The video signal output from the RIPU 35 is output to the OIPU 36 shown in FIGS. The video signal output to the OIPU 36 undergoes predetermined image processing in the OIPU 36 and is input to the SCU 7 again. The video signal input again to the SCU 7 is input to the VIDEO input switching circuit 37.

  The other input of the VIDEO input switching circuit 37 is a video signal output from the RIPU 35, and the OIPU 36 can select whether or not to execute image processing.

  The video signal output from the VIDEO input switching circuit 37 is input to SIBC 2:38 that manages the image data storage means (SDRAM), and is stored in an image memory composed of SDRAM. The image data stored in the image memory is sent to the SCSI controller 39 and transferred to an external device such as a personal computer or a printer.

  As shown in FIG. 4, a CPU 108, ROM 109, and RAM 110 are mounted on the SCU 7, and the CPU 108 communicates with an external device such as a personal computer by controlling the SCSI controller 36 of the external I / F 107 unit. . Similarly, the CPU 108 uses the video signal output from the VIDEO input switching circuit 37 as an external I / F function by the IEEE 1394 controller: ISIC 40 via the IEEE 1394 I / F and the network scanner controller: NIC 41 via the network I / F. And communication with external devices such as printers. The CPU 108 also performs motor control for moving the traveling body, and a traveling body motor (scanner motor) 9 which is a stepping motor of the scanner body, an ARDF paper feed motor (not shown), and a transport motor (not shown). The timing control is also performed.

  The ADU 42 has a function of relaying power supply of electrical components used for the ARDF unit. The input port connected to the CPU 108 on the SCU 7 is connected to the main body operation panel: SOP 43 via the VIOB 31.

  Main body operation panel: A start switch (not shown) and an abort switch (not shown) are mounted on the SOP 43, and a liquid crystal display (LCD: 117) is used as a display device. In addition, the liquid crystal display device 117 is provided with a matrix touch switch 116, and when the user presses an input symbol on the liquid crystal display, the touch switch reacts and the content input from the pressed coordinate information (corresponding to display information). Judging. The LCD controller 115 performs liquid crystal display control and touch switch input control. When the touch switch is pressed, the CPU 108 detects that the switch is turned on via the LCD controller 115.

  The image data flows from the SBU 10 to the RIPU 35, from the RIPU 35 to the memory control LSI (SIBC2), from the memory control LSI (SIBC2) to the SCSI controller, and from the SCSI controller to an external device (such as a personal computer). Further, ON / OFF of image output from the CPU is provided as a known technique in an image processing LSI (RIPU) (not shown) and is controlled by rewriting an internal register.

  Next, the flow of RGB image data will be described with reference to FIG. The analog video signal photoelectrically converted by the color CCD 42 on the SBU 10 for each of odd-numbered RGB (ODD) and even-numbered bit (EVEN) is input to the VIOB 31 via a buffer on the SBU 10. Here, each signal is represented as R-ODD (RO), R-EVEN (RE), G-ODD (GO), G-EVEN (GE), B-ODD (BO), and B-EVEN (BE).

  The analog video signal for each of the odd and even bits is input to the analog processing circuit 32 on the VIOB 31. The signal input to the analog processing circuit 32 is combined into an even number and an odd number via an internal gain amplifier (not shown) whose output level can be finely varied for every odd bit and even bit, and is output as an analog video signal.

  Two channels of D / A converters 44 are connected to the gain control terminal of each gain amplifier, and by varying the output voltage of the D / A converter 44 in an analog manner, the gain for the output analog video signal is set to an even number and an odd number. Can be varied. The CPU 45 on the SCU 7 sets the output voltage of the D / A converter 44. Since the D / A converter 44 has a reference voltage of 5V and a bit number of 8 bits, the CPU 45 can set the output voltage from 0 to 5V in 255 stages. The setting of the CPU 45 is digital (integer) from 0 to 255.

  The analog video signal output by combining the even and odd numbers from the analog processing circuit 32 is input to the A / D converter 33 for each RGB, and the A / D converter 33 converts the analog video signal into an 8-bit digital video signal. To do. A D / A converter 44 is connected to the reference (ref) setting terminal of the A / D converter 33 in the same manner as described above. By changing the output voltage of the D / A converter 44 in an analog manner, the A / D converter 33 is connected. The digital output value can be varied.

  Digital video data digitally converted by the A / D converter 33 is input to the shading ASIC 34. The shading ASIC 34 mainly performs shading correction, but also has a peak detector (not shown) that can detect a peak value of one line in the main scanning direction. This peak detector performs the function of storing the peak value of one line. The CPU 45 on the SCU 7 can read the stored peak value of one line.

Assuming that the input analog video signal of the analog processing circuit 32 is Vin and the output analog video signal (output signal of the analog processing circuit) amplified by a certain gain value G that is an analog processing circuit is Vout, Vout = G × Vin.
It becomes. The A / D converter 33 performs digital conversion with the reference voltage (reference voltage) as the maximum output. When the reference voltage is Vref, the 8-bit digital output value D is D = (Vout / Vref) × 255.
It becomes.

  In the above configuration, for example, how to deal with a case where an abnormality occurs in the shading data acquired by the shading ASIC 34 due to a problem before the A / D converter 33 will be described.

  In the image reading apparatus, a white plate with high reflectivity and good balance is used for each RGB wavelength spectrum immediately before image acquisition, and shading processing is performed for each RGB. Here, the shading correction is correction of the sensitivity variation of the CCD and the illuminance distribution variation of the light source, and is an indispensable matter for the digital optical system.

In implementation, shading data is generated first. The shedding data is obtained by reading the white reference plate and calculating the average of 16 lines of image data. That is,
Shading data = (first line data + second line data +... + 16th line data) / 16
Ask for.

In general shading correction, corresponding pixels are calculated from the shading data. The shading calculation formula when the maximum subtraction amount is 255 (8 bits) is shown below. The basic formula for this shading is
Data after shading correction = (read data / shading data) × 225
In the above equation, read data is D (n), shading data is S_ (n),
If the data after shading correction is D_ (n),
D_ (n) = (D (n) / S_ (n)) × 255
It becomes. (N is an arbitrary effective pixel number, for example, n = 1 to 7300.)

If the shading data data is D, the pre-shading image data is Din, and the image data after the shading data is Ds, each RGB data is individually executed. Therefore, each RGB data is expressed with a lowercase rgb.
D_ (n) g = (D (n) g / S_ (n) g) × 255
It becomes.

  When the RGB main scanning peak data of the white reference plate is usually about 220 digits and the analog gain is adjusted, but the shading data S_ (n) g created at the time of shading is extremely low (for example, S_ (n) When r and S_ (n) b are 208 digits and 209 digits, when only this S_ (n) g data is 32 digits), it can be clearly determined that the G system is abnormal.

  FIG. 8 shows the state of the registers in the ASIC as an example. The address (hex) of the shading data starts from 8000H and is written in 8 bits from b7 to b0.

  That is, the specific pixel value (8 bits) 208 of the R shading data is described at the address of 8000H. A specific pixel value (8 bits) 32 of G shading data is described at an address 8001H. A specific pixel value (8 bits) 209 of the B shading data is described in the address 8002H. The address of 8004H is an error flag (3 bits) of RGB shading data, “b2” is R, “b1” is G, and “b0” is B (“b7” to “b4” are Not used), “b1” is set to “1”, indicating an error state.

  In this case, even if the value of D_ (n) g is small (dark side), S_ (n) g has a large correction coefficient, so it skips to the bright side, and in some cases, the output saturation state (255 digit) It becomes. At this time, the abnormal pixel portion of the image looks like a white streak. If the abnormality is not a pixel unit of the CCD but a circuit, the abnormal range extends from the streak to the entire surface.

  As described above, shading abnormality data flows after the shading ASIC 34, but the value of the specific pixel (EX. Central several pixels) for main scanning is written in the G internal register in the shading ASIC 34. A normal setting range (for example, 64 digits or more) is designated in advance for this value, and a register in which an error flag is set is provided for other values (for example, 32 digits).

  At this time, the CPU 45 can read the data and make a determination in the CPU processing, but the load on the CPU side read via the CPU bus is less likely to be implemented on the ASIC 34 side. Similarly, R and B have the same configuration, and the same can be implemented for the three data. By providing such a function in the image processing portion of each RGB image data as described above, it is possible for the CPU to easily determine which RGB data system has an abnormality.

  In an ordinary text document, the document occupies a larger proportion than color graphics, and the color is currently black and white (gray) in most cases. There is no significant difference in reflectance. Therefore, as a substitute when G cannot be used, it is often possible to substitute B or R data.

  Actually, even in the case of a gray original, the original reflectivity with respect to the wavelength spectrum of the R or B component varies depending on each printed matter, ink, dye, and the like. Since the device side performs RGB shading with a white reference plate, there is a possibility that a subtle difference will occur in the density of a monochrome image obtained by using R and B data in the case of a document having a different RGB reflectance ratio of the white reference. It can be. Therefore, the monochrome image income data selection means is useful for the user as a fine adjustment.

  As an alternative to G, if R or B can be selected by looking at the image, for example, if there are many red documents in a gray document, select R if you want to brighten the red density, To make it darker, B can be selected. In the block diagram of FIG. 4, such selection can be performed by input to the touch switch 116 on the SOP 43. Since the LCD display is overlaid on the lower layer of the switch, when the graphic (R, G, B key display) displayed on the LCD is pressed, the upper layer touch switch reacts and R or The CPU can read that B is selected via the LCD controller.

  FIG. 6 shows an input structure in the SOP 43, and FIG. 7 shows a display example thereof. In FIG. 6, the upper sheet 116a of the switch is the x coordinate and the lower sheet 116b is the y coordinate, and the LCD 117 is disposed directly below.

  FIG. 7 shows the state of monochrome image usage data, where (a) is a normal state (white is expressed during use; G data is used), and (b) is an abnormality (G abnormality: It is expressed in dark. Abnormality is shown at the bottom), indicating that substitute data is being used (R selected, white change). At this time, the B display does not change.

  The touch switch of this embodiment is an electrode joint type between two sheets, but may have a structure that can obtain a pressed coordinate, a sensor type, or a hard key. good. As for the display device, a TN, STN, or TFT method may be used for the LCD, but a light source such as an LED or a light bulb may be provided for display under a printed picture.

  When R is selected as in this example, the CPU 45 writes the R selection into the monochrome selection register of the RIPU 35.

  The example shown in FIG. 9 shows a case where a selection register is installed at 9000H of the RIPU register. As shown in FIG. 9, the address of 9000H is an RGB selection data flag (3 bits), “b2” is R, “b1” is G, “b0” is B (“b7” to “b4” are not used) “B1” is set to “1” to indicate the R selection state.

  The RIPU 35 selects R data from the RGB data fetched by the RIPU 35 and sends it to the circuits after the memory controller on the subsequent stage through the image data bus. The same applies when another color is selected. In this embodiment, the address 9001H is the output destination selection flag (3 bits), “b2” is the OPIU, “b1” is the memory controller, “b0” is the other (such as an image evaluation apparatus), and “b7” to “b7” “b4” is not used. “B1” is set to 1 to indicate a memory controller selection state.

1 is a cross-sectional view illustrating an overall configuration of a color image reading apparatus. 2 is a block diagram showing the inside of a color image reading apparatus. FIG. It is a flowchart which shows the flow of image data. It is a block diagram which performs image control. It is a block diagram which shows the flow of RGB image data. It is sectional drawing and a perspective view which show a touch switch. It is a front view which shows the example of a display of a liquid crystal display device. It is a table which shows the internal register of shading ASIC. It is a table which shows the internal register of RIPU.

Explanation of symbols

1 Lens 2 Illumination lamp 3, 4, 5 Mirror 8 Glass platen 10 SBU
31 VIOB
32 Analog processing circuit 33 A / D converter 34 Shading ASIC
35 RIPU
36 OIPU

Claims (4)

  A photoelectric conversion element that has a plurality of light spectrum sensitivity characteristics, receives reflected light or transmitted light from the document, and reads the document image, and reads the image from the photoelectric conversion element into image data for each light spectrum sensitivity characteristic. An image data reading unit that outputs the image data, and one image data reading unit of the plurality of image data reading units. When the one image data reading unit is down, the other image data reading unit is An image reading apparatus comprising selection means for selecting and driving.   2. The image reading apparatus according to claim 1, further comprising display means for displaying the down when the one image data reading means is down.   3. The image reading apparatus according to claim 1, wherein the selection means is capable of manually switching the selection of the other image data reading means.   The plurality of image data reading means read three types of image data of RGB, and normally use G data to make a monocolor image, and when the G data is abnormal, one of R data and B data is used as substitute data The image reading apparatus according to claim 1, wherein the image reading apparatus is used.

Images