JPH09107470A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide excellent shading correction data without lowering the output rate of image data. SOLUTION: This processor is provided with a CCD line sensor for reading an original image for respective lines, a circuit 810 for adding data obtained by reading a reference while board for plural times by the CCD line sensor for respective images, a memory 809 for storing the average value of the respective picture elements of the data added for the respective picture elements by the circuit 810 as the shading correction data and a shading correction arithmetic circuit 813 for correcting the shading distortion of the image data obtained by reading the original image by the CCD line sensor corresponding to the shading correction data stored in the memory 809. The data of different picture element positions are added by an adding means circuit 810 everytime the reference white board is read by the CCD line sensor.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for correcting shading distortion of read data of an original image.

[0002]

2. Description of the Related Art A digital copying machine is known in which an original image is photoelectrically read by a CCD line sensor or the like, and an image is recorded on the basis of image data obtained thereby.
In reading such an original image, shading correction is performed to eliminate variations in image signal level (shading distortion) due to uneven sensitivity of pixels of the CCD line sensor, uneven light amount of the light source for exposing the original, and the like. Processing is performed.

In order to perform this shading correction, first, a reference white plate is read by a CCD line sensor, a change state of its output level is recognized, and the read data of the original image is corrected using the recognition result. Is commonly used.

Further, in order to prevent erroneous detection of shading correction data due to contamination of the reference white plate or noise mixing in the read signal, a configuration is proposed in which the reference white plate is read a plurality of times and the average value thereof is used as the shading correction data. Has been done.

[0005]

However, in order to form the shading correction data by using the average value, it is necessary to add the data of the reference white plate reading a plurality of times for every pixel. Image data must be output from the CCD line sensor in accordance with the time required for the addition process, which may hinder the speeding up of the original image.

[0006]

The present invention has been made in view of the above points, and an object of the present invention is to increase the processing speed in a method of obtaining shading correction data using an average value. Is a reading means for reading the original image line by line, an adding means for adding the data obtained by reading the reference white plate a plurality of times by the reading means for each pixel, and a data added by the adding means for each pixel. A storage unit that stores the average value of each pixel as shading correction data, and a correction unit that corrects the shading distortion of the image data obtained by reading the document image by the reading unit according to the shading correction data stored in the storage unit. Image processing for adding data of different pixel positions by the adding means every time the reading means reads the reference white plate It is to provide a location.

[0007]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described below with reference to preferred embodiments.

FIG. 1 shows the structure of a digital copying machine to which the present invention is applied, and will be described below with reference to the drawing.

The exposure lamp 201 is composed of a fluorescent lamp, a halogen lamp or the like, and irradiates an original on the original placing glass (original table) 200 while moving in a direction perpendicular to the longitudinal direction thereof. The scattered light from the document due to the irradiation of the exposure lamp 201 is generated by the first, second and third mirrors 202, 204, 2
It is reflected by 05 and reaches the lens 207. At this time, the first lamp including the exposure lamp 201 and the first mirror 202
The second movable body 206 composed of the second mirror 204 and the third mirror 205 in response to the movement of the movable body 203.
Moves at 1/2 speed, and from the irradiated document surface,
The distance to the lens 207 is always kept constant. The image on the original is the mirror 202, 204, 205 and the lens 20.
CC in which thousands of light receiving elements are arranged in a line through
An image is formed on the light receiving portion of the D line sensor 208, and is sequentially photoelectrically converted line by line by the CCD line sensor 208. The photoelectrically converted signal is processed by a signal processing unit (not shown), PWM-modulated, and output.

The exposure controller 210 drives the semiconductor laser based on the PWM-modulated image signal output from the signal processor, and irradiates the surface of the photoconductor 240 rotating at a constant speed with a light beam. At this time, a polygon mirror is used to deflect and scan the light beam in parallel with the axial direction of the drum-shaped photoconductor 240. The exposure controller 210 is cooled by the cooling fan 209.

Before irradiating the light beam on the photoconductor 240, the remaining charge on the drum is removed by a pre-exposure lamp (not shown), and the surface of the photoconductor 240 is uniformly charged by the primary charger 228. . Therefore, the photoconductor 240 receives the light beam while rotating to form an electrostatic latent image on the drum surface. Then, the developing device 211 visualizes the electrostatic latent image on the drum surface with a developer (toner) of a predetermined color.

On the other hand, reference numerals 223 and 224 denote transfer sheet stacking sections, on which transfer sheets having a standard size are stacked and stored.

The lift-ups 225 and 226 operate to lift the transfer paper stored in the transfer paper stackers 223 and 224 to the position of the feed roller pair 229 and 232. The feeding roller pairs 229 and 232 are driven by the same motor (not shown), and by switching the roller rotation direction, the transfer paper is selectively fed from one of the transfer paper stacking units 223 or 224. Further, the pair of rollers of the feeding roller pairs 229 and 232 are torqued in the reverse rotation direction of the sheet feeding, thereby preventing the double feeding of the recording medium.

Feed rollers 230, 233, 234, 2
Reference numeral 35 feeds the transfer paper from the transfer paper stacking units 223 and 224 to the registration roller 238. In the present embodiment, it is possible to further expand the third and fourth transferred paper stacking parts by further connecting them to the lower side. It is a pair of feeding rollers that performs a paper feeding operation to lead to the feeding roller. When the manual feed mode is selected on the operation unit, the manual tray 23
When the sheet 7 is opened and the transfer-target paper is manually fed, the feeding roller 236 feeds the manually-transferred transfer paper to the registration roller 238.

The registration roller 238 is a photosensitive member 240.
The transfer target paper is fed to the transfer position in synchronization with the timing of the front end of the transfer target paper formed on the image. A transfer charger 239 transfers the developed toner image on the photoconductor 240 to the fed transfer paper. After transfer, photoconductor 240
The residual toner is removed by the cleaner 227. The transferred paper that has been transferred is easily separated from the photoconductor 240 because the photoconductor 240 has a large curvature, but by applying a voltage to the charge removal needle 244,
The suction force between the transfer paper and the transfer paper is weakened to facilitate separation.

The separated transfer paper is conveyed by the conveyor belt 241.
Then, the toner is sent to the fixing units 212 and 213, and the toner is fixed. 212 is a ceramic heater and a film,
Composed of two rollers, the heat of the ceramic heater is efficiently transferred through the thin film. The cooling roller 214 radiates heat to the fixing unit roller 213. The feeding roller 215 is configured by one large roller and two small rollers, and feeds the transfer target paper from the fixing unit, and
Correct the curl of the paper to be transferred.

The direction flapper 222 switches the transfer direction of the transferred paper according to the operation mode. In the mode in which the transfer is performed once on one side of the transfer target paper (during one-side recording), the path from the feeding roller 215 to the sheet discharge port is selected. Reference numeral 216 denotes a pair of paper discharge rollers, which stack the transferred paper on which the image has been formed on the paper discharge tray 242 and discharge the paper.

During double-sided recording, the developing roller pair 216 is rotated in the reverse direction while leaving the trailing edge of the transferred sheet during the discharging operation by the discharging roller pair 216 after finishing the development on one side. At the same time, the direction of the direction flapper 222 is switched so that the direction flapper 222 passes below the direction flapper 222, and the transfer target paper is sent from the paper discharge port to the feeding roller 217. The feeding roller 217 has the same configuration as the feeding roller 215,
Correct the curl of the transfer paper and transfer the transfer paper to the intermediate tray 24.
Send to 3. After that, the transfer paper is transferred from the intermediate tray 243 to the feeding rollers 218, 219, 221, and 235 in this order.
The paper is fed to the above-mentioned transfer position, the toner is transferred to the back surface of the transfer target paper, and further discharged to the paper discharge tray 242.

Further, at the time of multiple recording, the feeding roller 215
The transfer target paper that has passed through is passed through the right side of the direction flapper 222 in the drawing by the switching of the direction of the direction flapper 222, and is sent to the feeding roller 217. The feeding roller 217 sends the transfer target paper to the intermediate tray 243. After that, the transfer paper is transferred from the intermediate tray 243 to the feeding roller 21.
8, 219, 221, and 235 are sequentially fed to the above-described transfer position, and toner transfer is performed on the same surface as the previous transfer,
Further, the paper is discharged to the paper discharge tray 242.

During double-sided recording or multiple recording,
When recording a plurality of sheets, the first sheet of transfer paper is fixed on the stopped feeding roller 218, and the intermediate tray 24
3 is loaded. The feeding roller 218 starts rotating when the second transfer target paper arrives, and sandwiches the two transfer target papers between the rollers. The two transfer target sheets are stacked on the intermediate tray 243 while being fixed by the stopped feeding roller 218. The third and subsequent transfer target sheets are similarly stacked on the intermediate tray 243. At this time, the front ends of the transfer papers that are stacked afterward are stacked so as to be shifted rearward with respect to the feeding direction as shown in FIG.

When a desired number of sheets are stacked on the intermediate tray 243, the feeding operation from the intermediate tray 243 is started. The feeding rollers 218 and 219 are the feeding rollers 2
While the transfer target paper is being sent to 21, the separation lever 220 is lowered between the leading ends of the first and second transfer target papers, and the first transfer target paper is sent to the feeding roller 221 as it is. The sheet is fed by the feed roller 235, guided to the transfer position, and transferred. For the second and subsequent sheets of transfer paper, the separation lever 2
20 and then feed rollers 218, 219
Is reversely rotated and returned to the intermediate tray 243. By repeating the same operation, all the transfer target papers in the intermediate tray 243 are fed.

A paper detection sensor is arranged in the conveyance path of the transfer paper and is used for detecting an error such as paper jam and measuring the operation timing of each part. The first sensor 250 is before the feeding roller 235, the second sensor 251 is before the registration roller 238, the third sensor 252 is before the feeding roller 215, and the fourth sensor 253 is for paper ejection. Between the roller pair 216 and the paper exit,
The fifth sensor 254 is provided immediately after the feeding roller 217,
The sixth sensor 255 is arranged in front of the separation lever 220.

Reference numeral 260 denotes a standard white plate, which is shading correction data for correcting variations in the output level of an image signal (shading distortion) due to uneven sensitivity of the CCD line sensor 208, uneven light source, uneven light quantity, and the like. Is to get. Prior to scanning the manuscript,
The standard white plate 260 is scanned several times by the CCD line sensor 208, and the white image data obtained thereby is used as shading correction data. Performs uniform correction (shading correction).

FIG. 3 shows a sectional view of an automatic document feeder (hereinafter referred to as ADF) of the present invention.

The ADF is detachable from the main body and has a structure like a lid that can be opened and closed with respect to a document table.
FIG. 3 shows a closed state, in which the ADF can operate at this time. Reference numeral 1515 is a platen glass (corresponding to the platen 200 of FIG. 1), and 1516 is a backing plate, which is a member belonging to the main body side. In addition, with a cable not shown,
The DF and the main body are electrically connected to each other to enable solidarity operation with the main body.

The side regulating member 1501 is a member that is movable in a direction perpendicular to the document feeding direction, in the front and back directions in the drawing, and aligns with the width of the document. When an operator places a document on the document stack tray portion 1517 with the document surface facing downward, sets the side regulating member 1501 and operates the copy start key 305 to start a copy operation or the like, a document feed start signal is output. It is transmitted from the main body via the cable and the document feeding operation starts.

During paper feeding, the lever 1518 pushes down the roller presser 1502, lifts the stopper 1519 upward, and rotates the paper feed roller 1503 to feed paper. Separation roller 1505 and roller holder 150
Reference numeral 4 separates the originals received from the paper feed roller 1503 one by one and sends them. Registration roller pair 1508, 1509
Sends the document received from the separation roller 1505 between the document holder 1510 and the document table glass 1515. The document retainer 1510 includes a pair of registration rollers 1508 and 150.
The original fed from 9 is guided to the original glass. Further, a document is appropriately pressed by a spring member (not shown) so that the document comes into close contact with the document table glass 1515.

At this time, the first movable body 203 of the main body of the digital copying machine is moved to a position below the position 1520 where the original comes into close contact with the original glass 1515 and is stopped. As a result, the document sent by the ADF is scanned (CC
Photoelectric conversion is performed by the D line sensor 208). The pair of paper ejection rollers 1511 and 1512 eject the scanned document onto the paper ejection tray 1514.

Note that 1506, 1507, and 1513 are
A lever switch that detects the passing state of the document during the document feeding operation determines the error such as paper jam at the ON / OFF timing.

FIG. 4 shows the key arrangement of the operation section of the digital copying machine, which will be described below with reference to the drawing.

Reference numeral 315 is a main power lamp, which lights up when the power is turned on. A power switch (not shown) is arranged on the side surface of the main body and controls energization to the main body. Reference numeral 301 denotes a preheating key, which is used to turn ON / OFF a preheating mode.

A copy mode key 304 is used to select a copy mode from a plurality of functions. A fax mode key 303 is used when selecting a fax mode from a plurality of functions. An option mode key 302 is used when an optional device such as a printer is attached and an optional mode is selected from a plurality of functions.
317 to 322 are status display lamps, respectively.
The lamps 317 and 318 are used during the copying operation.
And 320 are lamps 321 and 3 during the facsimile operation.
Reference numeral 22 indicates a state during option operation. Also, lamp 3
Reference numerals 17, 319 and 321 indicate normal operation states, and lamps 318, 320 and 322 indicate error states. Further, the lamp 317 indicates that copying is in progress if it is blinking, and that the image memory is in use if it is on. The lamp 319
Blinking indicates that fax transmission / reception is in progress, and lighting indicates that the image memory is being used. The lamp 321 indicates that data is being received if it is blinking, and that data is being sent if it is on. The lamps 318, 320, and 322 in each mode indicate a paper jam, no paper, and no toner, if blinking, and indicate a failure if lighting.

A copy start key 305 is used when instructing the start of copying. 306
Is a stop key, and is a key used for interrupting or canceling copying.

A reset key 308 operates as a key for returning to the standard mode during standby. 309
Is a guide key, and is a key used when it is desired to know each function. Reference numeral 311 is an interrupt key, which is used when interrupting during copying and copying is desired. 312
Is a numeric keypad and is used when entering numerical values. A clear key 313 is used to clear a numerical value. A user mode key 310 is used when the user changes the basic settings of the system.

Reference numeral 314 denotes 20 one-touch dial keys, which are used for one-touch dialing in facsimile transmission. Reference numeral 316 denotes a two-piece lid, and each key portion of the one-touch dial key 314 is hollowed to form a double lid. The first with the two lids closed by a sensor switch (not shown)
State, the second state in which only the first lid is opened, and 2
The third state in which the lids are opened is detected. Since the operation of the one-touch dial key 314 is determined by the combination of the open / closed states of these three types of lids, this embodiment has the same effect as the presence of 20 × 3 = 60 keys. .

Reference numeral 307 denotes a touch panel composed of a combination of a liquid crystal screen and a touch sensor, and an individual setting screen is displayed for each mode. Further, various detailed settings can be made by touching the drawn keys. Is possible.

FIG. 5 shows a block diagram of signal processing of the digital copying machine, which will be described below with reference to the drawing.

A CCD line sensor 401 (corresponding to the CCD line sensor 208 in FIG. 1) consisting of several thousand light receiving elements outputs an electric signal for one line by dividing it into two systems of an odd pixel and an even pixel. The A / D converter 402 has a C
It receives an analog signal output from the CD line sensor 401, converts it into a digital signal, and outputs it.

FIG. 6 shows the details of the A / D conversion unit 402. The analog processing circuit 901 inputs signals from two systems of the CCD line sensor, performs clamping, gain adjustment, and sample hold for each system, and then outputs the signals integrated into one system by switching processing. The A / D converter 902 inputs the output signal of the analog processing circuit 901 to the signal input, and outputs an 8-bit digital signal with the voltage of the reference voltage input from the analog switch 903 as a reference.

The AE circuit 904 controls the reference voltage of the A / D converter 902 so as to bring the output of the A / D converter in the background portion of the document close to the white level (FFhex). The analog switch 903 selects one of the constant reference voltage from the analog processing circuit 901 and the reference voltage from the AE circuit 904 according to a control signal from the CPU 423, and outputs it to the A / D converter 902.

A drive signal generating circuit (not shown) is provided to supply a line clock signal such as a synchronizing signal or a drive clock to the CCD line sensor, the analog processing circuit 901 and the A / D converter 902.

The AE circuit 904 controls the reference voltage value of the A / D converter based on the output of the A / D converter 902. That is, if the output of the A / D converter 902 is FFhex, the AE circuit 904 determines the first capacitor and the first resistor (not shown) connected to the AE circuit 904.
According to the time constant of, the reference voltage output is increased as shown in (2) of FIG. The output of the A / D converter 902 is FF.
If it is not hex, the second (not shown) connected to the AE circuit
In accordance with the second time constant determined by the capacitor and the second resistance, the reference voltage output is decreased as shown in (1) of FIG. Each of the above time constants is set to several tens of lines so that the reference voltage output does not suddenly change due to changes in the image signal.

In FIG. 5, the shading correction unit 40
An A / D-converted digital signal 3 is input to mainly digitally correct variations in output values between pixels of the optical system and the sensor with respect to black level and gain.

The pattern generator 404 generates various image patterns such as vertical ruled lines, horizontal ruled lines, checkered patterns, and gray scales for checking the functions after the scanner unit. The selector A 405 has a shading correction unit 40.
3 or one of the outputs of the pattern generator 404 is selected and output according to the control signal of the CPU 423.

The connector A406 includes image input signals, image output signals, pixel clock signals, image enable signals, and synchronizing signal terminals. The output of the selector A405 is connected to the terminal of the image output signal, and the function can be expanded by connecting a new signal processing circuit to the connector A406. The function can be expanded by connecting the selector B circuit to the connector A 406. Selector B407
Is the output signal of the selector A405 and the connector A406
One of the output signals from is selected and output according to the control signal of the CPU 423.

The scaling processing unit 408 has functions of thinning processing in the main scanning direction, linear interpolation processing, thinning processing in the sub scanning direction, italic processing, mirror image processing, repeat processing, and folding processing.

The scaling in the main scanning direction is realized by inputting a signal from the selector B407 and calculating a pixel value from two adjacent pixels by linear interpolation processing. However, when the scaling factor in the main scanning direction is 50% or less, the main scanning direction thinning process is performed as a pre-process in order to prevent moire and breaks in thin lines. In the pre-processing, adjacent n pixels (n = 2, 4,
Regarding 8), (a) maximum value or (b) average value is output. The CPU 4 selects the value of n and the processing (a) or (b).
23.

The magnification in the sub-scanning direction is changed by changing the scanning speed of an optical system such as an exposure lamp or a mirror, or by changing the speed of feeding the document to the document table 200 when using an automatic document feeder. Realize zooming. However, in the case of a small scaling ratio that exceeds the limit of the scanning speed of the optical system or the speed of feeding, the scaling processing unit 408c realizes the scaling with the electrical thinning process in the sub-scanning direction. To do. In the thinning process in the sub-scanning direction, a line buffer including an SRAM memory A409 is provided, and (a) the maximum value or (a) for n pixels (n = 2, 4, 8) adjacent between lines
(B) Output the average value. n and the processing (a),
The selection of (b) is performed by the CPU 423.

Further, each function of italic processing, mirror image processing, repeat processing, and folding processing is performed by the SRAM memory A40.
This is realized by the reading control of the line buffer consisting of 9.

The histogram is created by receiving the output of the selector B 407 and receiving the output of the selector B 407 from the SRAM memory A of the scaling unit 408.
409 is used to create a histogram. A signal that determines a sampling interval and a sampling range in creating a histogram is controlled by a timing signal generator 424. The timing signal generator 424 is connected to the CPU 4
Controlled by 23. The created histogram data is used for determining the luminance density conversion table in the AE mode when the prescan is used.

The filter processing unit 410 has a line buffer 411, which receives the signal after the scaling process and receives 5 × 5 signals.
Filter with the mask size of. As shown in FIG. 8, there are six types of coefficients constituting the filter of the filter processing unit 410, a to f, and the coefficients at the point-symmetrical positions are set to the same value. Further, a to f have the following relational expressions. a + 4 * (b + c + d + f) + 8 * e = 1

By adjusting the coefficient of the filter processing unit 410, the optical system and the output system are corrected, and the sharpness is adjusted by the user.

The image processing unit 412 is the filter processing unit 4
10 signals are input and masking processing or inversion processing is performed. The connector B413 includes terminals for an image input signal, an image output signal, an image clock signal, an image enable signal, and a synchronization signal. The output of the filter processing unit 410 is connected to the terminal of the image output signal, and the function can be expanded by connecting a new signal processing circuit to the connector B413.

The selector C414 selects one of the output signal of the image processing unit 412 and the output signal of the connector B413 according to the control signal of the CPU 423 and outputs it.

The connector C415 includes image input signal, image output signal, pixel clock signal, image enable signal, synchronization signal, address bus, data bus, and interrupt signal terminals. Selector C4 is connected to the image output signal terminal.
14 outputs are connected to connect the new system to connector C4
By connecting to 15, it is possible to expand the function.

The synthesizing section 416 inputs the output signal of the selector C414 into one input A, inputs the output signal from the connector C415 into the other input B, synthesizes and outputs. The output types are (1) inset synthesis, (2) multiplex synthesis, (3) halftone overlay, (4) watermark synthesis, (5) input A through, and (6) input B through, and the CPU 42
3 control signal.

Each composition will be described with reference to FIG. The inset composition is a process of extracting a part of an image and fitting it into another image. (D) is an example thereof, and a rectangular area around the rectangle in (b) is extracted and fitted in (a). It is an image. The multiple composition is a process of selecting one of the two images having a higher density and combining the two images. (E) is an example thereof, and (a)
And (b) is a composite image. In the watermark synthesis, a pixel whose density of one image is equal to or lower than a predetermined threshold value is replaced with another image. At this time, the image to be replaced is multiplied by a predetermined coefficient to reduce the density and has an effect like a watermark. (F) is an example thereof, in which the image of (b) is combined with (a) as a watermark. The halftone overlay replaces pixels in which the density of one image is equal to or higher than a predetermined threshold value with the other image. (G) is an example thereof, in which the dark circle of the image of (a) is replaced by the pattern of (c).

The table conversion processing unit 417 inputs the output signal of the synthesis processing unit 416, performs table conversion processing according to the data of the connected SRAM memory B 418, and outputs it.

The binarization processing unit 419 inputs the output signal of the table conversion processing unit 417, and selects whether to perform processing by a predetermined binarization method or through output according to the control signal of the CPU 423. Output.

In the binarization of this embodiment, one pixel is divided into two in the main scanning direction, and the divided small pixels are represented by a binary value to improve the resolution in the main scanning direction. Buffer 4
20 receives the output signal of the binarization processing unit 419 and adjusts the processing speed before and after the buffer 420 and the read start time of the image signal.

The PWM circuit 421 inputs the digital signal from the buffer 420, modulates the pulse width, and outputs it. In addition, the PWM circuit 421 has three types of modulation methods, and an appropriate modulation method according to the mode of the system is the CPU.
It is selected by 423 and processing is performed.

The first modulation method is the resolution priority mode, and as shown in FIG. 10A, pixel data represented by 8 bits is D / A converted, sampled and held,
The signal is compared with a triangular wave signal of one pixel cycle to output PWM.

The second modulation method is the gradation priority mode, and as shown in FIG. 10B, pixel data represented by 8 bits is D / A converted, sampled and held,
It is PWM-compared with a triangular wave signal having a 3-pixel cycle. Note that FIG. 10E shows a modulation method in which both the sample and hold cycle and the triangular wave signal are set to 3 pixel cycles, but a low frequency modulation signal is obtained and stable gradation characteristics are obtained. In the second modulation method, the same characteristics as in FIG. 10 (e) can be obtained for low-frequency image data, but as shown in FIG. 10 (d), for large-frequency components and high-frequency components, It also has the feature of modulating while preserving the components.

The third modulation method is a binary image mode, and as shown in FIG. 10C, one pixel is divided into four in the main scanning direction and a binary modulation signal is output. The values of the four divided small pixels are calculated by using the rising and falling edges of the first pixel clock and the rising and falling edges of the second pixel clock that is delayed by 1/4 phase from the first pixel clock. Are obtained by latching bit 7, bit 5, bit 6, and bit 4, respectively. In the binarization method of this embodiment, one small pixel is formed by combining bits 7 and 6 and bits 5 and 4 respectively, and one pixel is used in a form divided into two in the main scanning direction.

The laser section 422 receives the output of the PWM circuit 420 and controls the lighting operation of the laser.

The timing generator 424 and the control signal generator 425 are operated by the CP generator before starting the scanner operation.
It is set by U423. Timing generator 424
Generates timing signals for various parts of the system. Also,
The control signal generation unit 425 operates as an expansion port of the CPU 423 and is used as a control signal for each unit of the system. The system address bus and the system data bus, which are the buses of the CPU 423, are connected to the respective connectors, and the CPU 423 controls the system connected to the connectors.

Reference numeral 426 is the operation unit shown in FIG.
Reference numeral 423 also controls key input and display. Also, CPU
Reference numeral 423 controls the ADF 42 of FIG. 3 and the drive control of the driver 42 that drives the polygon mirror.

11 and 12 show a system configuration, which will be described below with reference to the drawings.

FIG. 11 shows a system in which a facsimile function, a printer function and an electronic sorter function are added to the basic copy function, and FIG. 12 shows a system in which a facsimile function is added.

In FIG. 11 and FIG. 12, the signal processor 5
01 executes the processing relating to the basic copy function shown in FIG. 5, and the facsimile unit 503 and the printer unit 50
5, the image storage unit 504 has a facsimile function,
Performs processing related to the printer function and electronic sorter function. The image storage unit 504 further has an external connector and can be added with functions such as an external scanner and an electronic file.

The resolution / gradation number conversion unit 502 has different resolutions and gradation numbers for each function.
And the number of gradations are matched, and an operation for converting to an optimum image quality is performed. The resolution / gradation number conversion unit 502 is used as the signal processing unit 5.
By connecting to 01, it is possible to expand the system as needed. The connection between the signal processing unit 501 and the resolution / gradation number conversion unit 502 is performed by the connector C415 in FIG.
Done through.

Each connector connecting the resolution / gradation number conversion unit 502, the signal processing unit 501, the facsimile unit 503, the image storage unit 504, and the printer unit 505 has a system address bus, data bus, and synchronization signal. , Pixel clock signals, interrupt signals, and other terminals are included.

Of these, the address bus and the data bus are the buses of the CPU 423 in FIG. Further, the synchronization signal is generated from the timing generator 424. The pixel clock signal is generated by an oscillator (not shown) of the signal processing unit. The interrupt signal is prepared for transmitting the request for processing, the end of processing, the occurrence of an error, etc. from the facsimile unit 503, the image storage unit 504, and the printer unit 505 to the CPU 423 in FIG.

FIG. 13 shows a block diagram of the inside of the resolution / gradation number conversion unit 502. Multiplexer A 601
Receives output signals from the signal processing unit 501, the facsimile unit 503, the printer unit 505, and the image storage unit 504,
Allocate to the first, second, third, and fourth signal paths. First
The signal path of is a through path, and the second signal path is a path for receiving a binary image signal and performing processing for displaying the contour of the curve smoothly, and outputting the contour.
A multiplexer 602 receives a binary image signal from the multiplexer A 601, and allocates 1 bit to each small pixel obtained by dividing one pixel into four in the main scanning direction, and outputs 4 bits per pixel.

The third signal path is a path for binarizing a multi-valued signal. The table conversion processing unit 603 receives the multi-valued signal from the multiplexer A 601 and performs table conversion, and the binarization processing unit 604 further. The output signal of the table conversion processing unit 603 is received and binarized using the average density method. At this time, 1 bit is assigned to each small pixel obtained by dividing one pixel into two in the main scanning direction, and 2 bits are output per pixel.

The fourth signal path is a path for receiving binary images of various resolutions, processing for displaying the contour of the curve smoothly and outputting multi-valued data of 8 bits, and the contour smoothing processing section B6.
05 receives the binary signal from the multiplexer A 601, performs a process for making the contour of the curve look smooth, and outputs an 8-bit multi-valued signal. Further, the scaling processing unit 606 causes the contour smoothing processing unit. Upon receiving the multi-valued output 605, linear interpolation is performed and an 8-bit multi-valued signal is output.

The multiplexer B 607 receives the signals from the first, second, third and fourth signal paths, and receives the signal processing section 501, the facsimile section 503, the printer section 505,
Allocate to the image storage unit 504.

In the facsimile transmission operation, the image signal is sent from the signal processing section 501 to the facsimile section 503 via the resolution / gradation number conversion section 502. At this time, the multiplexer A 601 of FIG. 13 selects 1 ...> C, the image signal passes through the third signal path, and the multiplexer B 607 selects 3 ...> D so that CP is selected.
U controls. Also, in the facsimile reception operation,
Facsimile unit 503 to resolution / gradation number conversion unit 502
The image signal is sent to the signal processing unit 501 via. At this time, the multiplexer A 601 of FIG.
D is selected, passes through the fourth signal path, and the multiplexer B 607 sets the CPU so that 4 ...> A is selected.
Controls.

In the printer operation, the printer unit 50
The image signal is sent from 5 to the signal processing unit 501 via the resolution / gradation number conversion unit 502. At this time, 2 ...> B is selected in the multiplexer A 601 of FIG. 13, the image signal passes through the second signal path, and the multiplexer B
In 607, the CPU controls so that 2 ...> A is selected.

Further, in the printer operation, when performing fixed-size scaling for different paper sizes, the image signal is sent from the printer unit 505 to the signal processing unit 501 via the resolution / gradation number conversion unit 502. At this time, 2 ...> D is selected in the multiplexer A 601 of FIG. 13, the image signal passes through the fourth signal path, and the multiplexer B 60
7, CPU controls so that 4 ...> A is selected.

When using the electronic sorter function or the like, the image signal is sent from the signal processing unit 501 to the image storage unit 504 via the resolution / gradation number conversion unit 502. At this time, FIG.
The multiplexer A 601 of the CPU selects 1 ...> A, the image signal passes through the first signal path, and the multiplexer B 607 selects the CPU of 1 ...> C.
Controls.

In the printer function, when the image data is sorted and output in the reverse order, the image signal is once sent from the printer unit 505 to the image storage unit 504 via the resolution / gradation number conversion unit 502. At this time, 1 ...> A is selected in the multiplexer A 601 of FIG. 13, the image signal passes through the first signal path, and the multiplexer B 60
CPU 7 controls 7 so that 1 ...> C is selected.
The image storage unit 504 writes the image to the hard disk 1909, and when reading the image, reads it in the reverse order. The read image signal is sent from the image storage unit 504 to the signal processing unit 501 via the resolution / gradation number conversion unit 502. At this time, the multiplexer A 601 of FIG.
B is selected, the image signal passes through the second signal path, and the multiplexer B 607 is controlled by the CPU so that 2 ...> A is selected.

FIG. 14 is a block diagram of the facsimile unit 503, which will be described below with reference to the drawing.

Connector C 701 includes terminals for the system address bus, data bus, video input signal, video output signal, sync signal, pixel clock signal, interrupt signal, and the like. The connector C 701 connects the facsimile unit 503 and the resolution / gradation number conversion unit 502.

In the facsimile transmission operation, the memory control unit 703 receives the video input signal from the connector C 701 and stores the image data in the image memory 704.

The encode / decode processing unit 702 is connected to the memory control unit 703 via an image address bus and a data bus. The encoding / decoding processing unit 702 has a built-in DMA controller, communicates with the memory control unit 703, captures image data at high speed from the image memory 704 by DMA transfer, and encodes the captured image data.
Code memory 7 by DMA transfer of the encoded code
Accumulate in 05. The code memory 705 is provided with a data protection measure by a backup power source 706 in order to cope with a system power source trouble such as a power failure.

The encode / decode processing unit 702, the memory control unit 703, the modem unit 710, and the voice synthesis unit 711 connected to the system address bus and data bus are connected by the CPU 423 connected via the connector C 701. Controlled.

When the encoding / decoding processing unit 702 finishes encoding the image in the image memory 704,
An interrupt signal to the CPU 423 for notifying the end is generated. Upon receiving the interrupt signal, the CPU 423 receives the code memory 705 via the memory control unit 703.
The encoded code is read from the modem unit 710.
Write as transmission data to. The transmission data is modulated into an analog signal by the modem unit 710 and transmitted from the external connector 709 via the NCU (network control unit) unit 708. Speaker unit 707
Is connected to the modem unit 710, and the communication status of the modem can be monitored by voice.

In the facsimile receiving operation, when the analog section received from the external connector 709 is received by the modem section 710 via the NCU section 708, an interrupt signal to the CPU 423 for notifying the data reception is generated. Upon receiving the interrupt signal, the CPU 423 reads the data from the modem unit 710 and writes the code in the code memory 705 via the memory control unit 703. When the CPU 423 finishes writing to the code memory 705, it sends a write end signal to the encode / decode processing unit 702 via the system bus.
The encoding / decoding processing unit 702 takes in data at high speed from the code memory 705 by DMA transfer through communication with the memory control unit 703, decodes the taken code data, and decodes the decoded image data by DMA transfer. It is stored in the memory 704 for use.
When the decoding of the code in the code memory 705 is completed, the encoding / decoding processing unit 702 generates an interrupt signal to the CPU 423 for notifying the completion.

The memory control section 703 outputs a video signal from the image memory 704 to the connector C 701 according to the timing signal from the connector C 701. In addition, the memory control unit 703 uses an internal buffer to store images in the image memory 704.
It has a function of rotating by 180 degrees, 180 degrees, and 270 degrees, and rotates a received or transmitted image as necessary and outputs it.

The voice synthesizing unit 711 synthesizes a voice response message when an incoming call arrives based on the data set via the system bus, and transmits it via the NCU unit 708.

The connector D 712 is connected to the system bus and the image bus, and is used to add lines or
Preparing for system expansion to improve performance.

Next, the shading correction section 403 of FIG.
Will be described in detail.

FIG. 15 is a block diagram of the shading correction unit 403 in this embodiment. In FIG.
Reference numerals 801, 802, 706 and 812 denote selectors, and the A side is selected when the select terminal is “L”. Reference numeral 803 denotes a counter, the count value of which is cleared by a horizontal synchronizing signal HSYNC signal which is synchronized with the reading of each line of the CCD line sensor 401, and which is counted up in synchronization with the RCLK signal corresponding to each pixel of the image signal. This count value becomes an address of a memory for storing the shading correction data. Reference numerals 804 and 805 denote JK flip-flops, which are the RCLK signal and HSYN, respectively.
Divide the C signal. Reference numerals 807 and 808 denote D flip props that latch the data input to the D terminal at the rising edge of the input clock (here, RCLK2).

Digital image data from the A / D converter 402 is input to the flip-flop 807 and the selector 806. Reference numeral 809 denotes a memory for storing the shading correction data, which stores the data input to the Di terminal at the address designated by the address (VA) input to the A terminal. Further, the data stored in the address specified by the A terminal is output from the Do terminal. Here, the writing and reading to the memory 809 and the selector switching instruction are performed using the control signals from the CPU 423, respectively. 810 is an adder circuit,
The result of adding two input data is output. Reference numeral 811 denotes a division circuit, which calculates an average value by dividing the input data by the number of times of addition. Reference numeral 813 denotes an arithmetic circuit that performs shading correction calculation.

In FIG. 15, the CCD line sensor 4
When the shading data obtained by reading the reference white plate by 01 is loaded into the memory 809, the input signals of the select terminals of the selectors 802 and 806 are set to “H”.
And all the data in the memory 809 is set to 0. Further, the select terminal of the selector 812 is set to “H”, and the output of the adder circuit 810 is connected to the Di terminal of the memory.

When the next rising edge of the HSYNC signal after the clear signal of the flip-flop 805 becomes "H" is the 0th line, the Q output of the flip-flop 805 becomes "H", so that the B side of the selector 801 is selected. Therefore, the Q output of the flip-flop 804 is the selector 801.
Output, that is, RCLK2. At this time, the memory 8
The address IVA of 09 is the flip-flop 80 at the rising edge of RCLK2 in the output value of the counter 803.
There will only be even values that are latched at 8. At this time, since the B side is selected in the selector 806, only the even-numbered data latched by the flip-flop 807 at the rising edge of RCLK2 is transmitted to the next stage as image data. At the same time, the data (here, 0) stored in the memory address latched by the flip-flop 808 here is read from the memory 809,
This data and the even-numbered image data from the selector 806 are added by the adder circuit 810 and written again at the same memory address. Therefore, in the 0th line, the memory 80
The even-numbered image data for one line is stored in the even-numbered address 9.

In the next first line, the flip-flop 80
Since the Q output of 5 becomes "L" and the A side is selected by the selector 801, the RCLK2 is flip-flop 804.
Inverted output of Q, that is, a clock having a phase opposite to that of RCLK2 of the 0th line, the address VA of the memory 809 latched by the flip-flop 808 has only an odd value. At this time, only the odd-numbered image data latched at the rising edge of RCLK2 by the flip-flop 807 is transmitted to the next stage as the image data.
At the same time, the data (here, 0) stored in the memory address latched here is read, and this read data and the odd-numbered image data from the selector 806 are added by the addition circuit 810, The addition result is written in the memory address latched by the flip-flop 808. Therefore, in the first line, the odd-numbered image data for one line is stored in the odd-numbered address of the memory 809.

In the next second line, the same operation as the above-mentioned 0th line is performed, but since the image data of the 0th line is stored in the even numbered address of the memory 809, the value rewritten in the memory 809 is 0. It is the sum of the even-numbered image data on the line and the (even-numbered) image data at the same position on the second line.

When the same processing is performed up to the 15th line (for 16 lines), the addition result of 8 lines of the even-numbered image data is stored in the even address of the memory 809, and the odd-numbered address is added to the odd-numbered image data. The addition result of eight lines of image data is stored. (The 16 lines are read while moving the reading position of the standard white plate.) After the above addition processing is completed, the select terminal of the selector 812 is set to “L”, and the image data is read from the memory 809. The dividing circuit 811 divides the data at all addresses in the memory 809 by 8 to obtain the average value of the data for 8 lines for each pixel, and stores the average value in the memory 809 again as shading correction data. This completes the shading data averaging process.

In calculating the average value, the average value of the even-numbered image data is first calculated, and then the average value of the odd-numbered image data is calculated, whereby the calculation by the division circuit 811 and the memory 809 are performed. Can be read and written in 2 clocks.

After the shading correction data is created,
A / D conversion unit 4 when the selection terminal of the selector 806 is set to "L" and the original is read by the CCD line sensor 401
The image data from No. 02 are continuously input to the shading correction calculation circuit 813 (image data Vi
And). Then, the shading correction data Dn stored in the memory 809 is read out in accordance with the input of the image data, and the calculation circuit 913 calculates Vi and Dn (Vo = C / Dn · Vo). As a result, the shading correction data Vo is output. However, C is a constant. This shading-corrected image data is input to the next scaling processing unit 408.

In this embodiment, even-numbered pixels are used for even-numbered lines,
Although odd-numbered pixels are sampled in the odd-numbered lines, the present invention is not limited to this combination, and similar effects can be obtained if the pixels to be sampled are alternated between the even-numbered lines and the odd-numbered lines. Further, it is possible to easily increase the calculation time per pixel by increasing the interval of pixels to be sampled to, for example, 2 pixels, 3 pixels, or the like.

According to this embodiment, since two clocks of the reference clock can be used per pixel in the addition process for averaging the shading correction data, a slow addition circuit or memory is used. Even
Shading correction data in high-speed image processing can be created.

Next, the scaling processing section 408 and the filter processing section 410 of FIG. 5 will be described in detail.

In the image processing described above, the sub-scanning decimation processing is performed by the scaling processing unit 408 at the same time as the reduction processing in the main scanning direction.

FIG. 16 is a block diagram of the scaling processing unit 408 and the filter processing unit 410 of this embodiment. Reference numeral 408 denotes a scaling processing unit, and 410 denotes a filter processing unit. 1
Reference numeral 01 is a circuit for performing reduction processing in the main scanning direction. 102
Is a sub-scanning decimation circuit, which produces effective data for only one of the predetermined lines while referring to the image data subjected to the reduction processing in the main scanning direction according to the decimation rate such as the resolution and the scaling factor. Here, it is reduced by 25% for simplicity, that is, 4
The case where valid data for one line is created will be described as an example. Line memories 103, 104, 105 and 106 are elements for storing one line of image data input for each pixel in synchronization with a pixel clock. The address in each line memory is reset by the RST signal. In the case of this embodiment, the HSYNC signal indicating the beginning of one line is connected to the RST terminal, and the image data for one line can be stored from the beginning of one line. Writing and reading of this line memory are controlled by the WE and RE signals, respectively, and the respective operations are performed only when the WE and RE signals are "L". When the WE and RE signals are "H", new data is not overwritten and the previous state is retained.

Reference numeral 107 denotes an arithmetic operation circuit for the filter processing, which outputs the image data for one line of the sub-scanning thinning circuit 102 and the line memories 103, 104, 105, 106.
A pixel matrix of 5 × 5 as shown in FIG. 8 is created from the image data of four lines stored in, and filter processing such as edge enhancement processing is performed by performing a predetermined calculation.

When the sub-scanning thinning circuit 102 performs the sub-scanning thinning processing of 25% in the sub-scanning direction, only one of the four lines becomes a valid image, and the remaining three lines are output as invalid images. At this time, the sub-scanning thinning circuit 102
Therefore, the signal VE which becomes “L” during the output of the valid image line and becomes “H” during the output of the invalid image line is simultaneously output. By inputting this VE signal to the WE and RE terminals of the line memories 103, 104, 105 and 106 in the filter processing unit 410, image data is written in each line memory only for valid image lines. Therefore, since the invalid image is not transmitted after this line memory, the image data used for the filter processing in the arithmetic circuit 107 is only the valid image.

In the present embodiment, only one of the four lines is an effective line. However, even when the sub-scanning thinning interval is changed, the arithmetic circuit 107 is operated by using the VE signal indicating the effective image line. It is possible to easily apply the idea of selecting the image data of. For example, at the time of 50% reduction, the VE signal is output once for inputting the image data of two lines.

As described above, the sub-scanning thinning circuit 1
02 is a VE signal indicating an effective image line,
The line memories 103 and 1 of the filter processing unit 410 in the next stage
Since the write enable control of 04, 105, and 106 can be performed, the filtering process can be performed only on the effective image line.

Although the digital copying machine is used in the above description, it is needless to say that the present invention is not limited to this and can be applied to a facsimile, an electronic file and the like.

[0113]

As described above, according to the present invention, data of different pixel positions are added every time the reference white plate is read, and the added data is used to form shading correction data. It is possible to lengthen the time, and thus it is possible to obtain the shading correction data at high speed without delaying the output of the image reading data.

[Brief description of the drawings]

FIG. 1 is a block diagram of a digital copying machine to which the present invention is applied.

FIG. 2 is a diagram showing a transfer target paper on an intermediate tray.

FIG. 3 is a diagram showing a configuration of an ADF.

FIG. 4 is a diagram showing an appearance of an operation unit.

FIG. 5 is a block diagram of a signal processing unit.

FIG. 6 is a block diagram of an A / D conversion unit.

FIG. 7 is a diagram showing an AE operation.

FIG. 8 is a diagram showing a filter matrix.

FIG. 9 is a diagram showing an example of image combining processing.

FIG. 10 is a diagram showing types of PWM modulation.

FIG. 11 is a diagram showing a system configuration.

FIG. 12 is a diagram showing a system configuration.

FIG. 13 is a block diagram of a resolution / gradation number conversion unit.

FIG. 14 is a block diagram of a facsimile unit.

FIG. 15 is a block diagram of a shading correction unit.

FIG. 16 is a block diagram of a scaling processing unit and a filter processing unit.

[Explanation of symbols]

 102 sub-scanning thinning circuit 401 CCD line sensor 403 shading correction unit 408 scaling processing unit 410 filter processing unit 809 memory

Claims (2)

[Claims] 1. A reading unit that reads an original image line by line, an adding unit that adds data obtained by reading the reference white plate a plurality of times by the reading unit for each pixel, and an addition unit that adds the data by each pixel by the adding unit. Storage means for storing, as shading correction data, an average value of the data for each pixel, and correction for correcting shading distortion of image data obtained by reading the original image by the reading means according to the shading correction data stored in the storage means. An image processing apparatus comprising: means for adding data of different pixel positions by the adding means every time the reference white plate is read by the reading means. 2. When the reading means reads the reference white plate at an even number of times, the adding means adds data of even-numbered (or odd-numbered) pixel positions, and the reading means reads the reference white plate at an odd-numbered time. The image processing apparatus according to claim 1, wherein the addition means adds the odd-numbered (or even-numbered) pixel position data.