JPH10341332A - Image-forming device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To form an appropriate image, even when an original protrudes from a read area by calculating the inclined angle of original and the position deviation of original from the coordinate of the area of detected original and operating an arbitrary angle rotating means based on these values, when the original protrudes from the read area. SOLUTION: When the original protrudes, an image signal processing part 20 detects the area of original placed in the read area on original glass and extracts plural coordinates showing the detected area of original. Then, an inclination angle θ of original and the deviation of original are calculated from the extracted coordinates of original area and based on these values, various parameters for rotating processing are set. Namely, even when the original is inclined and further placed while protruding from the read area, the area where the original placed on the original glass exists is detected from image data, respective sides are extracted from the coordinates of detected original area, four corners of original are estimated. Based on them, the various parameters required for arbitrary angle rotating processing are calculated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a digital copying machine,
The present invention relates to an image forming apparatus such as a facsimile and a multifunction peripheral.

[0002]

2. Description of the Related Art A function of rotating an arbitrary angle of two-dimensional image data to be output from an image forming apparatus has been proposed (for example, Japanese Patent Publication No. 63-9266). In the arbitrary angle rotation process, when the position of the original placed on the original glass for reading the original image is incorrect, the inclination of the original is detected, and the image data is rotated by an arbitrary angle to reduce the inclination. to correct. An image is formed from the image data corrected in this way. Therefore, even if the document placed on the platen glass by the user is skewed, the skew is automatically corrected and a copy can be taken.

[0003]

However, in the conventional image forming apparatus, when the original is out of the reading area on the original glass, it is impossible to set a correct parameter for the arbitrary angle rotation processing. Therefore, if an arbitrary angle rotation process is performed on such a document as it is, there is a disadvantage that an image originally required cannot be obtained.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an image forming apparatus having an arbitrary angle rotation function which can form an appropriate image even when a document is out of a reading area.

[0005]

According to the present invention, there is provided an image forming apparatus comprising: an original;
Image input means for inputting two-dimensional image data, detection means for detecting an area where the original placed on the original glass exists from the image data of the original input by the image input means, and detection by the detection means Arbitrary-angle rotating means for rotating the two-dimensional input image by an arbitrary angle when the area of the original is inclined, and an inclination angle of the original from the coordinates of the detected original area when the original is out of the reading area. And a control means for calculating the deviation of the position of the document and operating the arbitrary angle rotation means on the basis of them. That is, the area where the original placed on the original glass exists (original area)
Is detected from the image data, and even if the document is inclined and is placed out of the reading area, the four corners of the document area are estimated from the coordinates of the document area, and each side is extracted. Further, various parameters (paper size, tilt angle, deviation, etc.) necessary for the arbitrary angle rotation processing are calculated. In this way, even when the original is placed out of the reading area on the original glass, various parameters are calculated from the image data, and the image data is rotated by an arbitrary angle based on the calculated parameters, thereby automatically correcting the inclination of the original. Images can be output. Preferably, the control unit detects the length of the side of the original in the reading area from the coordinates of the original area obtained from the image data of the original, and determines the length of the side of the original in the reading area based on the inclination angle and the length of the side of the original in the reading area. Set the side length. The document is rectangular, and in the reading area, the inclination angle and the sides of the document are obtained from the image data, and the four corners of the document including those located outside the reading area are determined. Then, calculate the paper size based on this,
The arbitrarily rotating unit is operated based on the obtained tilt angle, paper size and deviation. Preferably, the control unit extracts a tilt angle by regarding a straight line obtained from the coordinates of the document area from the image data as each side of the document area. That is, the inclination angle is extracted by regarding the straight line that defines the boundary of the area where the document exists as each side. Preferably, the control means calculates the inclination angle from the longest side among the sides of the extracted document. Thus, the most reliable angle is obtained. Preferably, the control unit does not operate the arbitrary-angle rotating unit when the side indicating the document area cannot be extracted from the image data. This prohibits uncertain rotation correction of the document image. Also,
Preferably, the control unit does not operate the arbitrary-angle rotating unit when there is an angle that is significantly different from another calculation result among the respective inclination angles calculated from each side of the extracted document area. That is, if the shape of the manuscript is irregular,
Prohibits rotation of the original image. Preferably, the control means calculates the paper size from a rectangular area including the document area detected by the document detection means and having the extracted side as one of the sides. That is, a rectangular area including sides existing in the document area is obtained from the image data, and the size of the document is estimated based on this. Preferably, the control unit does not operate the arbitrary-angle rotating unit when the calculated paper size is larger than the maximum paper size. This prohibits rotation when a normal image cannot be obtained. More preferably, the image forming apparatus further includes a 90 ° rotation unit that rotates the image data by 90 °, and the control unit determines that the calculated paper size is larger than the maximum paper size, and that the maximum length is obtained when the length and width of the paper are switched. If the size is smaller than the paper size, the 90 ° rotation unit is operated together with the arbitrary angle rotation unit. That is, even if the obtained paper size is larger than the maximum paper size that can be handled by the image forming apparatus, if a normal image can be obtained by rotating the image by 90 °, the 90 ° rotation can be used together to tilt the image. Rotate the image to correct for. Preferably, the control unit is configured to control two adjacent sides of the extracted sides.
If there are the above sides, the amount of parallel movement of the image is calculated from those sides and the calculated paper size. That is, when there are two or more adjacent sides, at least one corner can be determined. Therefore, it can be used to calculate the parallel movement amount of the image data when the inclination of the image is corrected by the arbitrary angle rotation.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A digital copying machine as an embodiment of the image forming apparatus of the present invention will be described below with reference to the accompanying drawings. (1) Configuration of Copying Machine FIG. 1 schematically shows the overall configuration of a digital copying machine 1. The digital copying machine 1 includes a reading device 200, a printer device 300, a document transport unit 500, an operation panel (see FIG. 2), and a re-feed unit 600. The document transport unit 500 automatically transports the document set on the document feed tray 510 to a reading position on the document glass 18 and, after the reading device 200 reads the document, discharges the document to the discharge tray 5.
Discharge to 11 The size of the original is determined by sensors 551 and 55
2 detected. The reading device 200 includes a scanning system, the image processing unit 20, and the like. In the scanning system, the original glass 18
The upper document is read, and image data corresponding to each pixel of the image of the document is generated. That is, the original is irradiated by the exposure lamp 12 attached to the scanner 19 that moves below the original glass 18. Light reflected from the document passes through a first mirror 12, fixed mirrors 13a and 13b, and a condenser lens 14, and is incident on a line sensor (photoelectric conversion element) 16 using a CCD array or the like. The scanner position sensor 17 is provided to detect that the scanner 19 reaches the document reading area (image area). The line sensor 16 has a large number of photoelectric conversion elements arranged in a direction (main scanning direction) orthogonal to the paper surface of FIG. 1. The line sensor 16 reads an image at 400 DPI, for example, and outputs reflected light corresponding to each pixel of the image of the document. The signal is converted into an electric signal and output to the image signal processing unit 20. The image signal processing unit 20 processes the input electric signal and outputs image data to the memory unit 30. Here, the image signal processing unit 20 can detect the inclination of the document. The memory unit section 30 compresses and temporarily stores the image data input from the image signal processing section 20, and then performs decompression processing and transmits it to the printer device 300. At the time of decompression, if necessary, editing processing such as rotation of the inclined document by an arbitrary angle is performed. Further, the memory unit section 30 includes an external device interface section described later, and is connected to an external device through an external cable 51 via an external device connection connector 50.

The printer device 300 includes a print processing unit 40,
It comprises a laser optical system 60, an image forming system and the like. The print processing unit 40 controls the laser optical system 60 based on the image data received from the memory unit 30. In the laser optical system 60, the semiconductor laser 61 emits a laser beam modulated by a signal from the print processing unit 40, and the polygon mirror 65 deflects the laser beam and scans the laser beam on the photosensitive drum 71. That is, the deflected laser beam is guided to the exposure position on the photosensitive drum 71 via the f-θ lens 66 and the reflection mirrors 67 and 68, and the scanning of the original image on the photosensitive drum 71 is performed by this scanning. An image is formed.

Image formation is performed by an electrophotographic method in an image forming system, and a latent image formed on a photosensitive drum 71 is developed, transferred and fixed on paper to form an image on paper. Here, the photosensitive drum 71 rotated and driven in the counterclockwise direction in FIG. 1 is uniformly charged by the charging charger 72 and then exposed, and the latent image is developed by the developing unit 73.
(Only one developing device is shown for simplicity of illustration, but a color image can be developed by providing developing devices for four colors of cyan, magenta, yellow, and black.) Then, the toner image is guided from the cassettes 80a and 80b to the photosensitive drum 71 through the sheet guide 81 and the timing roller 82, and the toner image obtained by the development is transferred to the sheet by the transfer charger 74. Next, the sheet is separated by the separation charger 75 and is conveyed to the fixing roller 84. The fixing roller 84 fixes the image on the sheet by heat, and then the sheet is discharged through the re-feed unit 600.

FIG. 2 is a plan view of the operation panel of the copying machine. The operation panel includes a liquid crystal touch panel 91 for status display and various mode settings, a numeric keypad 92 for inputting numerical values (number of sheets, magnification, etc.), a clear key 93 for returning numerical values to standard values, and a copy mode. Panel reset key 94 for initializing, stop key 95 for instructing stop of copy, start key 96 for instructing start of copy, interrupt key 97 for designating interrupt start and return, tilt angle correction A key 102 for instructing a mode is arranged. By pressing the key 102, various choice settings (mode setting, etc.) related to the arbitrary angle rotation function can be made on the liquid crystal touch panel 91.
Can be performed. FIG. 3 shows an example of a system configuration in which a plurality of digital copying machines 1, 1 'are connected on a network. The digital copying machines 1 and 1 'are connected to controllers 2 and 2', which are an example of external equipment, via an external equipment interface connector 50 and an interface cable 51. The controller 2 is connected to the computer 3 via a general-purpose interface such as Ethernet, and is connected to another digital copying machine 1 'through the controller 2'. As an example, when various settings (conditions such as paper size, magnification, number of prints, staple, sorting, etc.) and an output command are input for printing a data file created by the computer 3, print processing on the computer 3 is performed. Is sent to the controller 2. The controller 2 generally converts image data (such as PostScript data) sent from the computer 3 into raster data. Therefore, the controller 2 is equipped with a memory for at least one screen. When converted into bitmap data, the image data is transmitted to the digital copying machine 1 (at the same time, various setting conditions are also transmitted via the controller 2). Processing is performed. It should be noted that the present invention is not limited to this system form, but also includes a system such as a facsimile apparatus connected to a modem through a telephone line.

(2) Copying Machine Control System FIGS. 4 and 5 show a control unit 100 for controlling the copying machine 1. The control unit 100 is mainly composed of eight CPUs 101 to 107. Each of the CPUs 101 to 107 has a ROM 111 to 117 storing a program and a RAM 121 to 117 serving as a work area for executing the program.
127 are provided. Note that the CPU 103 is provided in the memory unit 30.

The first CPU 101 controls input of signals from various operation keys on an operation panel (see FIG. 2) and display on a liquid crystal touch panel. The second CPU 102 controls driving of a scanning system of the reading device 200 and controls each unit of the image signal processing unit 20. Here, the inclination of the document is detected from the read image. The third CPU 103 compresses the read image data by controlling the memory unit 30, temporarily stores the read image data in the code memory 303, then reads out the image data and outputs it to the print processing unit 40. Here, at the time of reading, rotation editing of image data including image tilt correction is performed. The memory unit 30 has a function of interfacing with an external device, and transmits and receives image data and control data. The fourth CPU 104 controls the print processing unit 40, the optical system 60, and the image forming system of the printer device 300. The fifth CPU 105 controls the control unit 10
Processing for adjusting the overall timing of 0 and setting the operation mode is performed. The sixth CPU 106 controls the document conveying unit 50
0 is controlled. The seventh CPU 107 controls the sheet re-feed unit 600. These CPUs 101 to 10
Between 7, serial communication is performed by interruption, and data such as commands and reports are transmitted and received.

FIG. 6 is a block diagram of the controller 2 controlled by the CPU 105. The first external interface 700 exchanges signals with the external computer 3. The interpreter 702 translates data (for example, postscript data) sent from the computer 3 and develops the data into raster data,
2 stores raster data. When one page of image data is stored, the image data is output to the printing unit via the second external interface 704.

Next, processing of image data will be described. FIG. 7 illustrates an image signal processing unit 20. The image signal processing unit 20 includes a timing control unit 21, an amplifier 23, an A / D
Converter 25, shading correction unit 26, density conversion unit 2
7, an electric scaling section 28, an editing section 29, etc.
It is controlled by the PU 102. The image signal processing unit 20 converts the input signal from the line sensor 16 into an amplifier 23
And amplified by the A / D converter 25 for each pixel.
It is quantized to bit image data. The image data is then subjected to various processes such as shading correction, conversion to density data (gamma correction), electric scaling, and editing, and then sent as image data to the memory unit 30 or a printer device. The CPU 102 controls the image signal processing unit 20
Of the scanner unit of the laser scanning system, communication with the CPU 105, and the like, and controls the entire reading apparatus 200. The document size and the orientation of the document are detected as follows. In determining whether or not the read image is a document, for example, the document placed on the document glass 18 is scanned by setting the document cover of the document feeder 500 to a mirror-like surface, and a portion where the amount of reflected light is large is determined as the document. to decide. If the mirror surface is used, the amount of reflected light is very small, so that the determination is easy. Therefore, scanning may be performed with the document cover opened. C
When receiving an instruction of the document size detection operation from the CPU 105, the PU 102 performs a preliminary scan. CPU10
2 controls a scanner motor that drives the scanner 19 based on position information from the scanner position sensor 17;
The scanner 19 is caused to scan in the sub-scanning direction. At the timing corresponding to the sub-scanning position, the document size and the portrait / landscape orientation are detected from the content of the image data and the monitor position information, and the detection result is transmitted to the CPU 105. The CPU 102
Based on the magnification information transmitted from the PU 105, at the time of image reading, the speed of the scanner motor is controlled at a scan speed that matches the magnification information.

FIG. 8 is a block diagram of the memory unit 30. The switching unit 301 switches the route of image data to the image signal processing unit 20, the print processing unit, and the external interface 310. The area determining unit 303 determines whether the input image data is simple binary data or halftone data. The binarization processing unit 302
Based on the parameter settings from the CPU 103, the image data is converted into binary data within a range that can be restored to multi-valued image data by error diffusion, dithering, or the like. The image memory has a capacity for two pages. The binary data is stored in the image memory. The binary data read from the image memory 304 is then sent to the compressor 3 in the code processing unit 305.
11 and stored in the code memory 306. The code memory 306 stores, for example, A at 400 DPI.
This is a multi-port memory having a capacity of one page of four sizes. The code memory 306 temporarily stores read data of a document image of a plurality of pages. The data in the code memory 306 is managed by a code management table provided in the RAM 126.

At the time of printing, the compressed image data in the code memory 306 is expanded by an expander 312. If image editing is required, the editing control unit 307 performs editing processing (rotation, scaling, movement, etc.) at the time of decompression, and performs the editing processing and decompression processing simultaneously. If necessary, a tilt angle rotation is also performed at the time of extension. The decompressed data is transferred to the image memory 304. When the data of one page is decompressed, the binary data read from the image memory 304 is converted into multi-value data by the multi-value processing unit 308.
If necessary, after the smoothing processing is performed in the smoothing processing unit 309, the data is transferred to the print processing unit 40 or the external device via the switching unit 301. CPU103
Sends control parameters to the multi-value processing section 302 and the smoothing processing section 309.

Next, the operation sequence of the copying machine 1 in image reading and printing will be described with reference to the CPUs 101 to 106.
Command (Q), report (A) exchanged between
And the flow of data. FIG. 9 shows a schematic sequence of the memory write operation. First, the fifth CPU 105, which manages the entire sequence, requests the third CPU 103, which controls the memory unit 30, to prepare for memory. In response, the CPU 103
Setting of a bus connection state for transferring image data from the image signal processing unit 20 to the image memory 304 to the internal hardware, setting of a mode for binarization processing, start address of a writing area in the image memory 304 , X
Make settings such as Y-length information. When these settings are completed and the preparation is completed, the CPU 103
05 is notified of the completion of memory preparation. Next, CP
U105 requests the CPU 103 and CPU 102 to read. In response, the CPU 102 requests the document scanning unit of the reading device 200 to scan. In this way, when the scanning of the original is started and the scanner 19 reaches the original reading area on the original glass, the read data is read according to the image processing mode set by the CPU 102.
(Image data D2) is transferred from the image signal processing unit 20 to the memory unit unit 30. When the scan is completed, a scan end signal is sent to the CPU 102, and the CPU 102
103 notifies the CPU 105 of the completion of the reading. Next, the CPU 105 requests the CPU 103 to compress the data. In response to this, the CPU 103 sets the read address from the image memory 304, the XY length information, the write address to the code memory 306,
Eleven modes (for example, arithmetic coding system and MH system) are set, and each unit is started. This allows the compressor 3
11 is performed, and the code data is stored in the code memory 306. When the compression process is completed, the CPU 1
03 notifies the CPU 105 of the completion of compression. At this time, if the code memory 306 is full of data, a compression completion report to which a parameter indicating that compression is impossible is added is sent to the CPU 105. Thus, the CPU 105 can know that the code memory 306 has become full.

FIG. 10 shows a schematic sequence of a memory read operation. In the memory read operation, image data is read from the image memory 304, and a copy image is printed on a sheet based on the image data. CPU10
5 requests the CPU 103 to decompress data. In response, the CPU 103 reads the address from the code memory 306, the amount of data, the write address to the image memory 304, the XY length information, the mode of the decompressor 312 (for example, the arithmetic coding method and the MH method), Set the edit processing mode such as angular rotation and start each unit. As a result, decompression (editing) processing is performed, and the image data is written to the image memory 304.
When the decompression process is completed, the CPU 103 notifies the CPU 105 of the completion of the decompression. Next, the CPU 105
The PU 103 is requested to prepare a memory for reading an image from the image memory 304. In response, CP
U103 is the image memory 3 for the internal hardware.
04, the setting of the bus connection state for outputting the image data to the print processing unit 40, the setting of the start address of the read area of the image memory 304, the XY length information, and the like are performed, and the CPU 105 is notified of the completion of the memory preparation. next,
The CPU 105 controls the CPU 103 and the printing unit 40
Request a print. From the print processing unit 40 to the CPU 105
Then, a paper feed report notifying the paper transport state is sent to the printer, and thereafter, the image data read from the image memory 304 is output to the print processing unit 40, and printing is performed. When printing is completed, the CPU 103 and the print processing unit 40
A print completion report and an ejection completion report are sent to the PU 105. C who received these reports
The PU 105 gives a memory clear request to the CPU 103 as needed.

(3) Reading of an original when the original is skewed (including a case where the original protrudes) Next, image reading when the original to be read on the original glass 18 is inclined will be described. As shown in FIG. 11, the document 10 is placed with its copy surface facing downward on a document glass 18 with reference to the upper right end (indicated by a triangle). The longitudinal direction of the original glass 18 is the sub-scanning direction of scan reading, and the direction perpendicular thereto is the main scanning direction. In the example shown in FIG. 11, the document is placed away from the image reference, and its position is not parallel to the sub-scanning direction. In this example, the document does not protrude from the reading area. FIGS. 12 and 13 show images of a read image when the document 10 (shown by a broken line) does not protrude from a reading area (a rectangle shown by a solid line). (Since the document 10 is viewed from the copy surface side, the image reference is at the upper left corner.) The image signal processing unit 20 processes image data of a rectangular area including at least the range of the document and processes the image data of the document area where the document exists. Perform detection,
From all the coordinates around the original (that is, the edge),
As shown in the figure, the coordinates of the four corners of the rectangular document are detected. Here, the main scanning direction is the X axis, and the sub scanning direction is the Y axis. X _max and X _min are the maximum and minimum X coordinates. Of the remaining two X coordinates, the larger _one is X ₁ and the smaller _one is X ₂ . Y _max and Y _min are the maximum and minimum Y coordinates. Of the remaining two Y coordinates, the larger _one is Y ₁ and the smaller _one is Y ₂ . In the example of FIG. 12, the X and Y coordinates of the four corners of the document are (X ₁ , Y _min ),
(X _max, Y _2), a _{(X min, Y 1),} (X 2, Y max). The lengths a, b, c, and d of the four sides of the document 10 can be calculated as follows from the coordinates of the document. a = {(X _max −X ₁ ) ² + (Y ₂ −Y _min ) ² } b = {(X ₁ −X _min ) ² + (Y ₁ −Y _min ) ² } (1) c = √ {(X ₂ −X _min ) ² + (Y _max −Y ₁ ) ² } d = {(X _max −X ₂ ) ² + (Y _max −Y ₂ ) ² } In the example of FIG. Unlike the case of FIG. 12, the coordinates of the four edges of the document are (X _min , Y ₁ ),
(X ₁ , Y _min ), (X ₂ , Y _max ) and (X _max , Y ₂ ). Also in this case, the lengths a, b, c, and d of the four sides can be calculated by the same formula. Here, the editing processing parameters for correcting the tilted image are as follows. First, X ₁ −X _min <
In the case of Y ₁ −Y _min (FIG. 12), rotation coordinates: (X ₁ , Y _min ) Rotation angle θ: tan ⁻¹ {(X ₁ −X _min ) / (Y ₁ −Y _min )} (2) Destination Nation address (pmmax, pmday): (−X ₁ , −Y _min ) When X ₁ −X _min > Y ₁ −Y _min (FIG. 13), rotation coordinates: (X _min , Y ₁ ) Rotation angle θ : Tan ^-1 {(Y ₁ −Y _min ) / (X ₁ −X _min )} (3) Destination address (pmdx, pmday): (−X _min , −Y ₁ ) The rotation coordinates are the upper left corner of the figure. Are the coordinates of the corner near. The rotation angle θ is a rotation angle for rotating the original with reference to the rotational coordinate position to make the original parallel to the reading area. The destination address is the coordinates of the transfer destination to the memory, and corresponds to the length from the upper left position to the image reference.

When the original 10 is out of the reading area (solid line rectangular portion), for example, even if the read image (broken line) is an image as shown in FIG. 14, the above coordinates are detected by detecting the original area. I do. The upper left end of the figure is the reference position of the reading area. Now, the reading range of the X axis is X _min0 ~
It is assumed that X _max0 and the reading range of the Y axis is Y _min0 to Y _max0 . Since the corner of the document is outside the reading area, the extreme value X
Each of _max , X _min , Y _max , and Y _min is detected at two points that intersect the edge of the reading area. That is, when the extreme values of the X axis and the Y axis match the maximum values X _max0 , Y _max0 or the minimum values X _min0 , Y _{min0 of the} reading range, two points are detected. Thus, the coordinates of the eight points at the edge of the document are (X _min0 , Y
_{10), (X min0, Y} 11), (X 10, Y min0), (X 11,
_{Y min0), (X 20,} Y max0), (X 21, Y max0), (X
_max0, Y _20), a (X _max0, Y _21). Thus, sides a, b, c, and d existing in the reading range can be extracted. Here, the lengths a, b, c, and d of the four sides of the document and the lengths e, f, g, and h of the protruding portion can be calculated from these coordinates as follows. a = √ ｛(X _max0 −X ₁₁ ) ² + (Y ₂₀ −Y _min0 ) ² ｝ b = √ ｛(X ₁₀ −X _min0 ) ² + (Y ₁₀ −Y _min0 ) ² ｃ c = √ ｛(X ₂₀ −X _min0 ) ² + (Y _max0 −Y ₁₁ ) ² ｝ d = √ ｛(X _max0 −X ₂₁ ) ² + (Y _max0 −Y ₂₁ ) ² ｝ (4) e = X ₁₁ −X ₁₀ f = Y ₁₁ −Y ₁₀ g = X ₂₁ −X ₂₀ h = Y ₂₁ −Y _{20 Further} , as shown in the figure, the lengths of the four sides when the document is rectangular can be calculated as follows. e * cosθ + a + h * sinθ, f * cosθ + b + e * sinθ, (5) g * cosθ + c + f * sinθ, h * cosθ + d + g * sinθ. Here, the inclination angle θ can be calculated for each side from the coordinates described above.

On the basis of the above-described concept, it is possible to obtain the editing processing parameters corresponding to every manner of placing the original. If at least one of the four sides can be extracted, the inclination angle θ can be calculated. In addition, the length and coordinates of each side can be calculated when the extracted side is regarded as a part of the side and the rectangular area including the read original area is defined as the effective original area. FIG. 15 shows an example of an image of a read image in a case where the placed document protrudes from the reading area but all four sides can be extracted. Although the original protrudes from the left and upper sides of the reading area, two points (X _min0 , Y ₁₀ ) and (X _min0 ,
_{Y 11), (X 10,} Y min0), can be detected (X _11, Y _min0). Although the length of the side d can be extracted, complete sides cannot be extracted for the sides a, b, and c. However, the length of each side and the coordinates of the four corners can be calculated based on the sides e and f and the inclination angle θ.
In this example, since g = h = 0, the length of the four sides can be calculated as g = h = 0 in equation (5). e * cosθ + a, f * cosθ + b + e * sinθ, (6) c + f * sinθ, d + g * sinθ.

FIG. 16 is a diagram showing an example of a read image when only three sides can be extracted. in this case,
Position α = (X _min0 ,
Y ₁₀ ) = (X ₁₀ , Y _min0 ). In this example, b = g
= H = 0, the length of the four sides can be calculated as b = g = h = 0 in equation (5). e * cosθ + a, f * cosθ + e * sinθ, (7) c + f * sinθ, d. FIG. 17 is a diagram illustrating an example of an image of a read image when only two sides c and d can be extracted. In this example, the upper left corner position α = (X _min0 , Y ₁₀ ) = (X ₁₀ , Y _min0 ), and the upper right corner position β = (X ₁₁ , Y _min0 ) = (X _max0 ,
Y ₂ ). In this example, since a = b = g = h = 0, the length of the four sides is a = b = g = h in equation (5).
= 0. e * cosθ, f * cosθ + e * sinθ, (8) c + f * sinθ, d. FIG. 18 is a diagram illustrating an example of an image of a read image when only one side can be extracted. In this example, the position α = (X _min0 , Y ₁₀ ) of the upper left corner of the readable document portion
(X ₁₀ , Y _min0 ), and the position β = (X ₁₁ , Y
_{min0) = (X max0, Y} 2), the position of the lower right corner gamma =
(X _max0 , Y ₂₁ ) = (X ₂₁ , Y _max0 ). In this example, since a = b = d = g = 0, the length of the four sides can be calculated as a = b = d = g = 0 in equation (5). e * cosθ + h * sinθ, f * cosθ + e * sinθ, (9) c + f * sinθ, h * cosθ.

(4) Tilt Angle Correction by Arbitrary Angle Rotation When the tilt of the read document is detected, the image is automatically rotated by the tilt angle to obtain an image that does not tilt. Therefore, when the user places the document on the document glass and erroneously places the document on the document glass, the image is automatically rotated by the tilt angle, so that a normal image can be obtained. Hereinafter, the arbitrary angle rotation processing will be described. The arbitrary angle rotation is performed in the editing processing unit 307 in the memory unit 30 based on the editing parameters detected by the image reading. First, the editing processing unit 307 will be described.
FIG. 19 shows a portion related to the arbitrary angle rotation of the edit processing unit 307 in the memory unit 30 and the image memory 304.
Editing such as rotation by the editing processing unit 307 is performed by using the image memory 304 and combining shift processing, XY conversion processing, and bit swap processing (FIG. 20,
See FIG. 21). When the rotation process is not performed (0 ° rotation), only the bit swap process is performed, and only the signals W2 and R2 are accessed, as indicated by the broken line.

When the rotation processing is performed, the input image data transferred from the decompression unit 312 is 16 bits, and is first converted into 32 bits by the 16 → 32 conversion unit 3070 in the editing processing unit 307. In order to process the internal operation of the edit processing unit 307 and access to the image memory 304 at high speed, all data processing is performed in 32 bits. First
The shift unit 3071 performs a shift process if data after bit conversion is necessary, and
The data is written to the slice area 3040 and the 1-2 slice area 3041. The 1-1X-Y conversion section 3072 and the 1-2X-Y conversion section 3073 respectively include a 1-1 slice area section 3040 and a 1-2 slice area section 30.
An XY conversion process is performed on the data R1 read from 41. The second shift unit 3074 performs a shift process on the data after the XY conversion, if necessary. The first bit swap unit 3075 performs bit swap processing on the data from the 16 → 32 conversion unit 3070 or the data from the second shift unit 3074 if necessary, and edits the virtual paper area 3042 (maximum) in the image memory 304. To write to A3 size capacity). The second XY conversion unit 3076 outputs the second bit swap unit 3 if necessary at the time of printout.
077, the X of the R2 data read from the print output virtual paper area 3042 in the image memory 304
After performing the -Y conversion process, the data is written to the second slice area 3043 in the image memory 304. The R2 data read from the virtual paper area 3042 or the R3 data read from the second slice area 3043 during the print output operation is converted into one bit by the 32 → 1 conversion unit 3078 and sent to the multi-value processing unit 308. Will be transferred. During continuous operation, access to the image memory 304 (signals W1 to W3,
R1 to R3) are processed in parallel on a 32-bit bus line in a time division manner.

Next, referring to FIG.
The 0 ° rotation processing will be described. The image of the input image transferred from the decompression unit 312 is received with the upper left end as the image reference, as shown in FIG. The input image data is, as shown in (b) at the upper right, the first to the first in the line unit.
The data is read from the slice area unit 3040 in units of 32 bits × 32 bits from data of 32 bits × 32 lines, and is edited by the 1-1X-Y conversion unit 3072. Next, for 90 ° rotation, the center (c)
Is read out as shown in FIG.
The data is written in the virtual paper area 3042. At this time, image reference (destination address (pmdax, pmday))
Expand the data so that is at the lower left. 1-1 slice area section 3040 and 1-2 slice area section 3
No. 041 is a pair, and the above processing is performed by a double buffer operation. Next, referring to FIG. 21, first bit swap unit 3075 and second bit swap unit 3077
A bit swap process of image data according to will be described. In this process, as shown in the upper part of the figure, the arrangement of the 32-bit data is reversed. Then, as shown in the lower part of the figure, the inverted data is written in the virtual paper area 3042 such that the image reference (destination address) is at the lower right corner. In addition, by combining the 90 ° rotation and the bit swap processing, it is also possible to rotate 270 °. At this time, the data is developed so that the image reference is at the upper right end. The editing processing unit 307 can simultaneously process the parallel movement of the image by changing the transfer destination (destination address) to the virtual paper area based on the image. Although not described, processing such as erasing, copying, and pasting of an image in a specific area is also possible.

Next, the calculation for the arbitrary angle rotation will be described. The following equation shows a rotation process in the affine transformation. As shown in Expression (10), the coordinates (X, Y) are rotated by the angle θ to become coordinates (U, V). This operation is decomposed as shown in Expression (11), and the first shift, 9
It is a combination of 0 ° rotation, second shift, and -90 ° rotation. The above-described editing processing unit 307 performs an operation by combining these processes. In the first shift and the second shift, only the X-axis parameter is processed with the Y-axis parameter fixed.

(Equation 1)

Next, the inverse operation of the above-described rotation address operation will be described. Equations (12) and (13) are obtained from equations (10) and (11).
This shows the inverse transformation of.

(Equation 2)

In the arbitrary angle rotation processing of the image data, the image data is processed according to the above-described calculation for the arbitrary angle rotation. An arbitrary angle rotation process (shift shift method) of image data will be described with reference to FIG. In the editing processing unit 307, as shown in the upper left of FIG. 22, the image of the input image transferred from the decompression unit 312 is received on the basis of the upper left corner. The input image data is subjected to a first shift process in the above-described first shift unit 3071 in accordance with the set rotation angle θ, as shown in the upper part of the center of FIG.
1 slice area 3041 and 1-2 slice area 3
042 is written in line units. Next, a 1-1 slice area 3041 and a 1-2 slice area 304
The data is read out in units of 2 to 32 × 32 bit blocks, and processed by the 1-1X-Y conversion unit 3072 and the 1-2X-Y conversion unit 3073. Next, based on the data for the two blocks, the second shift unit 3074 performs a second shift process according to the set rotation angle θ, and writes the data to the virtual paper area 3042. The two slice areas 3040 and 3041 form a pair, and the above operation is performed by a double buffer operation. With the processing so far, the image is rotated by (90 ° + θ). When printing out from the virtual paper area 3042, -90
The rotation by 270 ° (270 ° rotation) is processed using the second XY conversion unit 3076 and the second bit swap unit 3077, buffered (bit converted) to the second slice area unit 3043, and then multi-valued. Output to the unit 308. In the rotation process by the shift method, θ is desirably as small as possible in consideration of the image quality after the process.
<Θ <45 °. 90 ° rotation for more angles
Process by combining with unit rotation.

Next, a logical expression for simultaneously processing the arbitrary angle rotation and the scaling process of the image data is shown. Equation (14)
Indicates rotation processing and scaling processing in the affine transformation,
The coordinates (X, Y) are rotated by an angle θ and multiplied by α to become coordinates (U, V). This is decomposed as shown in Expression (16), and the first shift, the 90 ° rotation, the second shift, and -9
The rotation is 0 °. In the first and second shifts, only the X-axis parameter is processed while the Y-axis parameter is fixed.

(Equation 3) Next, an inverse operation of the above-described rotation address operation will be described. Equation (1
6) and (17) show inverse transformations of equations (14) and (15).

(Equation 4)

FIG. 23 shows the above-described arbitrary angle rotation processing of image data. This processing includes shift processing, 90 ° rotation, shift processing, and −90 ° rotation. The scaling is performed in the editing processing unit 307 shown in FIG. 19 by associating the scaling parameters in the first and second shift units. However, depending on the magnification, it is necessary to increase the number of slice areas. At this time, only the scaling is processed,
If not, the rotation angle may be set to 0 °.

(5) Setting of Skew Correction When a Document Protrudes As described above, the copying machine reads two-dimensional input image data of a document placed on a reading area of a document glass. Then, the area where the document exists is detected, and when the area of the document is inclined, the two-dimensional input image is rotated by an arbitrary angle to correct the inclination. If the document is out of the reading area, the originally required corrected image cannot be obtained without taking the out of consideration into account. Therefore, when the original protrudes, the area of the original placed in the reading area on the original glass is detected, a plurality of coordinates indicating the detected area of the original are extracted, and the original is extracted from the coordinates of the extracted original area. And the deviation of the document are calculated, and various parameters for the rotation process are set based on the calculated angle. In other words, even when the original is inclined and protrudes from the reading area, the area where the original is placed on the original glass (original area) is detected from the image data, and the detected original area is detected. Each side is extracted from the coordinates to estimate the four corners of the document (see FIGS. 14 to 18), and various parameters (paper size, tilt angle, shift amount, and the like) necessary for the arbitrary angle rotation processing are calculated based thereon. In the inclination correction, the image data is developed on the virtual paper.
Then, the image data is moved so that the inclination is eliminated, and the position of the corner where the two adjacent sides can be extracted matches the corner (origin) of the virtual paper, thereby correcting the original displacement. In this way, even when the original is placed outside the reading area on the original glass, various parameters can be calculated from the image data, and an image in which the inclination of the original is automatically corrected based on the parameters can be output.

For example, the length of the side of the original in the reading area (a to d in the example of FIG. 14) is detected from the coordinates of the original area from the image data of the original, and the inclination angle θ and the side of the original in the reading area are detected. The length of the side outside the reading area is set based on the lengths a to d (see FIG. 40). The document is rectangular, and in the reading area, the inclination angle and the sides of the document are obtained from the image data, and the four corners of the document including those located outside the reading area are determined. Then, the paper size is calculated based on this, and the arbitrary angle rotation is operated based on the obtained tilt angle, paper size, and deviation. For example, a straight line obtained from the coordinates of the document area from the image data is regarded as each side of the document area, and the inclination angle θ is extracted (see FIGS. 35 to 39). That is, the inclination angle is extracted by regarding the straight line that defines the boundary of the area where the document exists as each side. Further, the inclination angle θ is calculated from the longest side of each side of the extracted document (see FIG. 35, steps S2437 to S2437d). Thus, the most reliable angle is obtained. When the side indicating the document area cannot be extracted from the image data, the arbitrary angle rotation is not performed (FIG. 35, step S24).
32, S2433). This prohibits uncertain rotation correction of the document image. If there is an angle that is significantly different from the other calculation results among the respective inclination angles calculated from each side of the extracted document area, the rotation is not performed at an arbitrary angle (see FIG. 35, steps S2435 and S2433). That is, when the shape of the document is irregular, rotation of the document image is prohibited.

The paper size is set from a rectangular area including the detected original area and the extracted side as a part of the side (see FIG. 41). That is, a rectangular area including sides existing in the document area is obtained from the image data, and the size of the document is estimated based on this. If the calculated paper size is larger than the maximum paper size, the rotation is not performed at an arbitrary angle (see FIG. 35, step S2459). This prohibits rotation when a normal image cannot be obtained. Further, the image forming apparatus further includes a 90 ° rotation unit that rotates the image data by 90 °, and the calculated paper size is larger than the maximum paper size.
If the paper size is smaller than the maximum paper size, the image is rotated 90 degrees and the image is rotated 90 °.
Rotate (Steps S2451, S2454, S24)
55-S2457). That is, even if the obtained paper size is larger than the maximum paper size that can be handled by the image forming apparatus, if a normal image can be obtained by rotating the image by 90 °, the 90 ° rotation can be used together to tilt the image. Rotate the image to correct for. If there are two or more adjacent sides of the extracted sides, the parallel movement amount of the image is calculated from the two sides and the calculated paper size (see FIG. 45). That is, when there are two or more adjacent sides, at least one corner can be determined. Therefore, it is used to calculate the amount of parallel movement of the image data when correcting the inclination of the image.

(6) Control of Copying Machine The control flow of the copying machine will be described below. FIG.
5 is a main flowchart of a first CPU 101 that controls an operation panel. When the power is turned on, first, an initialization for initializing a RAM, a register, and the like is performed (step S11). Thereafter, an internal timer that defines the length of one routine is set (step S12), key input processing for accepting key operation is performed (step S13), and panel display processing for performing display according to the key operation is performed (step S14). ). Then, after performing other processing (step S
15) Wait for the end of the internal timer (step S16)
YES), returning to step S12 and repeating the above processing. In addition, communication with another CPU is performed as interrupt processing at an appropriate time.

FIG. 25 is a main flowchart of the second CPU 102 for controlling the image signal processing section 20. RA
After initial settings for initializing M and registers are performed (step S21), an internal timer that defines the length of one routine is set (step S22). Next, image data is input (step S23), document detection processing is performed (step S24, see FIG. 26), image processing is performed (step S25, see FIGS. 27 to 39), and image data is output (step S23). S26). Then, other processing is performed (step S27), and after the end of the internal timer (YES in step S28), the process returns to step S22, and the above-described processing is repeated. FIG. 26 shows the document detection (step S2).
It is a flowchart of 4). Here, a straight line composed of a plurality of coordinates extracted in the document area is regarded as each side of the document area. First, the edge of the document is detected (step S24).
1) Extract coordinates of each edge point (step S24)
2). Then, from the extracted coordinates, X _min , X _max , Y
_min, extracting coordinate points indicating the Y _max (step S24
3) Extract each side of the document from the extracted coordinate points (step S244). Here, if it is determined that the side of the document could not be extracted (NO in step S245), the arbitrary angle rotation is prohibited (step S246), and a warning message is displayed on the display device 91 of the operation panel (step S247). ). FIG. 27 shows the image processing (FIG. 25,
It is a flowchart of step S25). Detection of presence or absence of protrusion (Step S251), angle detection (Step S252, see FIGS. 28 to 32), size detection (Step S253, FIGS. 33 to 35), size setting (Steps S254, FIG. 36), rotation Coordinate setting (step S255, see FIG. 37), destination setting (step S256, see FIG. 38), 90 ° rotation destination setting (step S257, see FIG. 39), and other processing (step S258) are sequentially performed.

In this process, even if the document is placed outside the reading area, the area where the document exists (document area) is detected from the image data, and each side is extracted from the coordinates of the detected document area to form a rectangle. Assuming that there is, the four corners of the document are estimated. Next, various parameters for the arbitrary angle rotation are determined. The parameters include: The tilt angle (rotation angle) θ (step S252) indicates the tilt of the effective document area. Effective document area size spx, s
py (step S253) is a parameter representing the size of the detected document (x direction and y direction when the document is not inclined). In addition, size cpx, cpy
Is a size parameter (x direction and y direction) for setting a virtual paper area for developing image data. The paper size (step S254) is usually the minimum standard paper size including the effective document area. Rotation coordinates rot_
x, rot_y (step S255) are coordinates (x-direction and y-direction) serving as a rotation center with respect to the origin of the reading area
Represents For example, in the example of FIG. 14, since the upper left corner is the origin of the reading area, the rotation coordinates rot_x, rot_y
Represents the corner of the upper left effective document area 10 (represented by a broken line) closest to the origin, and is an amount representing the displacement of the document. The destinations (shift amounts) pmmax and pmday (step S256) represent the coordinates with respect to the origin of the virtual paper area from the rotational coordinates. When the image data is developed in the virtual paper area, the coordinates are moved only by the destination.
The image data is developed from the origin of the virtual paper area. The 90 ° rotation parameter (step S247) is a parameter used when rotating an image by 90 °. The obtained various parameters are set in the editing control for the image memory (see FIGS. 40 and 41).

FIG. 28 shows the angle detection (FIG. 27, step S).
It is a flowchart of 252). First, angles θ _a , θ _b , θ _c , θ _d are obtained from the sides a, b, c, d of the document area, respectively.
Is detected (step S2521, see FIGS. 29 to 32). When the angle detection is completed, the angle used for the actual rotation processing is selected. If all the angles θ _a , θ _b , θ _c , and θ _d are incorrect (YE in step S2522)
S) or when it is not θ _a ≒ θ _b ≒ θ _c ≒ θ _d (excluding an incorrect angle) (NO in step S2525), that is, among the angles detected from the four sides, there is one that is significantly different from the others. If, ie not rectangular, size
with prohibiting any angle rotation as 1 _err (step S2523), the theta and theta ₀ (step S25
23). Although θ _a aθ _b ≒ θ _c ≒ θ _d (excluding an incorrect angle) (YES in step S2525), θ _a ≒
If not _{θ b ≒ θ c ≒ θ d} > 0 ( excluding wrong angle) (NO in step S2526), the theta and theta ₀ (step S2524). If the above conditions are not satisfied (YES in step S2526), the lengths of the sides a to d are compared (step S2527), and the length is calculated from the longest side among the sides a, b, c, and d extracted from the reading area. angle θ _a, θ _b, a theta _c or theta _d adopted as an angle theta (step S2527a~S247d). Then, other settings (step S2528) are performed, and the process returns.

[0037] Figure 29, in step S2521 of FIG. 28 is a flowchart of an angle theta _a detection based on the edge a. If not sides a> 0 (NO in step S25211a), the angle theta _a as err (step S2521
9a), and return. If the side a> 0 (YES in step S25211a), it is determined whether the original protrudes next. If Y _min = Y _min0 , ie,
If the side e exists (step S25212a), X ₁₁
Is referred to as X ₁ (step S25213a). Also, X _max
= If an X _max0, i.e., if the edge g is present (step S25214a), the Y ₂₀ and Y ₂ (step S25215a). Then, (Y ₂ −Y _min ) and (X
_max- X ₁ ) are compared (step S25216)
a), the rotation direction is determined so that the rotation angle θ becomes smaller. That is, when (Y ₂ −Y _min ) <(X _max −X ₁ ), tan ⁻¹ {(X _max −X ₁ ) / (Y ₂ −Y _min )} is changed to the angle θ
(Step S25217a), (Y ₂ −Y _min ) <
When (X _max −X ₁ ) is not satisfied, tan ⁻¹ ｛(Y ₂ −Y _min ) / (X
_max− X ₁ )｝ is set to the angle θ (step S25218)
a).

30 to 32, the angles θ _b , θ _c and θ _d are detected as in FIG. FIG.
It is a flowchart of an angle theta _b detection based on side b in step S2521. If not sides b> 0 (NO in step S25211b), an angle theta _b as err (step S25219b), the process returns. Side b> 0
Is satisfied (YES in step S25211b), it is determined whether the original is protruding next. If Y _min = Y _min0, that is, if the side e exists (step S
NO in 25212b), the X ₁₀ and X ₁ (step S
25213b). If X _min = X _min0, that is, if the side f exists (step S25214)
b), using Y ₁₀ and Y ₁ (step S25215b).
Then, the lengths of (Y ₁ -Y _min ) and (X ₁ -X _min ) are compared (step S25216b), and the rotation direction is determined so as to reduce the rotation angle. That is, (Y ₁ −Y _min )
When <(X ₁ −X _min ), tan ⁻¹ ｛(X ₁ −X _min ) /
(Y ₁ −Y _min )｝ is set to the angle θ _b (step S25217)
b) When (Y ₁ −Y _min ) <(X ₁ −X _min ), ta
^{_{n -1 {(Y 1 -Y min}} ) / (X 1 -X min)} is referred to as angle theta _b (step S25218b).

FIG. 31 is a flowchart for detecting the angle θ _c based on the side c in step S2521 of FIG. If edge c> 0 is not satisfied (N in step S25211c
O), the angle θ _{c is set} to err (step S25219)
c), return. If edge c> 0 (step S
Next, it is determined whether or not the original protrudes. If a Y _max = Y _max0, i.e., if the edge g is present (YE in step S25212c
S), the X ₂₀ and X ₂ (step S25213c).
Also, if it is X _min = X _min0, i.e., if the edge f exists (step S25214c), the Y ₁₁ and Y ₁ (step S25215c). Then, (Y _max −
Y ₁₎ and to compare the length of the (X ₂ -X _min) (step S2
5216c), the rotation direction is determined such that the rotation angle becomes smaller. That is, (Y _max −Y ₁ ) <(X ₂ −X _min )
When it _{^{is, tan -1 {(X 2 -X}} min) / (Y max -Y 1)} and the angle theta _c (step S25217c), (Y _max -Y
₁ ) When not (X ₂ −X _min ), tan ⁻¹ ｛(Y _max −Y ₁ )
/ (X ₂ −X _min )｝ is set to the angle θ _c (step S252)
18c).

FIG. 32 is a flowchart for detecting the angle θ _d based on the side d in step S2521 in FIG. If the side d is not 0 (N in step S25211d)
O), the angle θ _d is set as err (step S25219)
d), return. If edge d> 0 (step S
Next, it is determined whether or not the original protrudes. If a Y _max = Y _max0, i.e., if the edge g is present (YE in step S25212c
S), the X ₂₁ and X ₂ (step S252133
d). Also, if it is X _max = X _max0, i.e., if the edge h is present (step S25214d), the Y ₂₁ and Y ₂ (step S25215d). And (Y
The length of ( _max− Y ₂ ) is compared with the length of (X ₂ −X _min ) (step S25216d), and the rotation direction is determined so as to reduce the rotation angle. That is, (Y _max −Y ₂ ) <(X ₂ −X
_min ), tan ^-1 ｛(X ₂ −X _min ) / (Y _max −
Y ₂ )｝ is set to the angle θ _d (step S25217d),
When (Y _max −Y ₂ ) <(X ₂ −X _min ), tan
^{_{-1 {(X 2 -X min)}} / (Y max -Y 2)} is referred to as angle theta _d (step S25218d).

FIG. 33 is a flowchart of the size detection (FIG. 27, step S253). The size of the document sheet is determined based on the inclination angle and the length of the side of the document sheet in the reading area, and the length of the side outside the reading area is set. Here, spx and spy are the sizes of the original paper to be set in the x and y directions. size_er
If it is r = 1 (NO in step S2531), since the side of the document can not be detected correctly, the angle theta as theta ₀ (step S2535), usually to detect the document size (step S2536). And spx as s
px3, spy is set to spy3 (step S2537), and other settings are made (step S253).
8). If size_err = 1 is not satisfied (step S
If YES at 2531), the skewed document size is detected (step S2532, see FIG. 34). And spx
1 and spx2 and spy1 and spy2 are compared in length (step S2533). If the lengths are significantly different (NO in step S2533), the flow advances to step S2535 to detect a normal document size (see FIG. 35). Otherwise (YE in step S2533)
S), spx1 is set to spx, spy1 is set to spy (step S2534), and other settings are made (step S2538).

FIG. 34 shows the detection of the skewed original size (FIG. 33,
It is a flowchart of step S2532). In this flow, the inclination angle θ and the length a of the side of the original in the reading area,
The length of the side outside the reading area is set from b, c, and d. First, the rotation direction is determined by comparing (Y ₂ −Y _min ) and (X _max −X ₁ ) (step S 25321). (Y ₂ −
Y _min) <(X _max if -X ₁₎ is the (step S2
5321), e * cosθ + a + f * sinθ is sp
x1 (step S25322); otherwise, substitute for spy1 (step S25323). Similarly, to determine the direction of rotation by comparing the (Y ₁ -Y _max) and (X ₂ -X _min) (step S25324). (Y ₁ −
Y _max) <(the case is X ₂ -X _min) (step S2
YES at 5324), g * cosθ + c + f * sinθ is sp
x2 (step S25325); otherwise, substitute for spy2 (step S25326). The rotation direction is determined by comparing (Y ₁ −Y _min ) and (X ₁ −X _min ) (step S25327). (Y ₁ −
If Y _min ) <(X ₁ -X _min ) (step S2)
YES at 5327), f * cosθ + b + e * sinθ is sp
y1 (step S25328), otherwise, substitutes spx1 (step S25329). Further, to determine the direction of rotation by comparing the (Y _max -Y ₂₎ and (X _max -X ₂₎ (step S2532A). (Y _max −
_{Y 2) <(X max -X} 2) If it is the (step S25
YES at 32A), h * cosθ + d + g * sinθ is spy
2 (step S2532B), otherwise substitute into spx2 (step S2532C). FIG.
Is normal document size detection (FIG. 33, step S253)
It is a flowchart of 6). By substituting X _max to SPX3 (step S25361), and substitutes the Y _max on Spy3 (step S25362).

FIG. 36 is a flowchart of the size setting (FIG. 26, step S254). Where spx,
spy is the size of the detected document, and cpx, c
py represents the set size of the virtual paper area 3042, and c
px_max and cpy_max represent the maximum size that can be set. If cpx <spx and cpy <spy (YES in steps S2541 and S2542), cp
A minimum fixed size satisfying x <spx and cpy <spy is set (step S2543). If the size of the detected document is larger than the maximum paper size but smaller than the maximum size that can be set by changing the length and width of the effective document size, that is, spx <cpx_max, spy ≧ cp
If y_max, spy <cpx_max (YES in step S2541, NO in step S2542, YES in step S2545), or spx ≧ cpx_
max, spx <cpy_max, spy <cpx_ma
x (NO in step S2541 and step S254
(YES at 2544, YES at step S2545), c
A minimum standard size satisfying px <spy and cpy <spx is set (step S2546), and 90 ° rotation is set (step S2547). In other cases, that is, when the effective document area sizes spx and spy are larger than the settable maximum sizes cpx_max and cpy_max (NO in step S2541 and N in step S2544).
O), set the maximum fixed size, ie, sx_ma
x is substituted for cpx, and sy_max is substituted for cpy (step S2548). Further, the angle θ is set to θ ₀ and the rotation is not performed at an arbitrary angle (step S2549). Finally, make other settings (step S254A), and return.

FIG. 37 is a flowchart for setting the rotational coordinates (FIG. 27, step S255). Where rot
_x and rot_y are coordinates representing the center of rotation with respect to the origin of the reading area. When not rotating (θ = θ ₀ )
(YES in step S2551), rot_x and rot_
Both y are set to 0 (step S2552). If it rotates (NO in step S2551), the flow branches depending on whether the detected rotation angle θ is positive or negative (step S2553). If the rotation angle θ is positive, if e> 0 (step S
(YES in 2554), X ₁₀ + e * sin ² θ is substituted for rot_x, and e * sin θ * cos θ is substituted for rot_y (step S2555). If e> 0, that is, e
= If 0 (NO in step S2554), the X ₁ r
ot_x and Y _min into rot_y (step S2556). If the rotation angle θ is negative, f> 0
If (YES in step S2557), f * sin θ
* Cos θ is substituted for rot_x, and Y ₁₀ + f * cos ² θ is substituted for ro
Substitute for t_y (step S2558). If f> 0, that is, if f = 0 (step S255)
7), Y _min is substituted for rot_x, and Y ₁ is substituted for rot_x.
y (step S2559). Then, other processing (step S255A) is performed, and the process returns.

FIG. 38 shows a destination setting (FIG. 2).
7 is a flowchart of step S256). The destination is coordinates pmdx and pmday (based on the virtual paper origin) corresponding to the read area origin of the input image to be developed on the virtual paper. While the virtual paper area is set to the standard size, the input image size determined by extracting each side from document detection does not always match it. Therefore, when there are two or more extracted sides with priority given to the reading reference side (the rotation center side), the two sides are extracted so that the corner from which two adjacent sides can be extracted matches the corner of the virtual paper area. The destination (parallel movement amount) is set based on the set paper size. here,
spx and spy are the sizes of the detected original paper (see FIG. 33), and cpx and cpy are virtual paper areas 30.
42, and rot_x and rot_y are coordinates representing the rotation center with respect to the origin of the reading area (see FIG. 37). First, when the angle θ> 0 and the side a,
b are both positive (θ> 0 in step S2561)
And if YES in step S2562) or if the angle θ <0 and both sides b and c are positive (θ <0 in step S2561 and step S
If YES in 2563), two sides sandwiching the rotation center have been extracted. Therefore, -rot_x is changed to pmd
, And -rot_y to pmday (step S2564). Also, when the angle θ> 0 and the side a,
b are both negative (NO in step S2562,
If YES in step S2565) or if angle θ <0 and sides b and c are both negative (NO in step S2563, YES in step S2566)
), -Rot_x is substituted for pmdex,
(-Rot_y + cpy-spy) is substituted for pmday (step S2567).

If the angle θ> 0 and both sides c and d are positive (steps S2562 and S2
NO at 565 and YES at step S2568) or when angle θ <0 and both sides d and a are positive (steps S2563 and S256)
No. 6 and YES in step S2569),
(−rot_x + cpx−spx) is substituted for pmdax, and −rot_y is substituted for pmday (step S
256A). If the angle θ> 0 and both sides d and a are positive (steps S2562 and S2
565 and NO in step S2568, step S256
B is YES) or the angle θ <0 and both sides a and b are positive (step S256)
3, NO in steps S2566 and S2569, and YES in step S256C), (-rot_x + c
(px-spx) into pmdax, and (-rot_y
+ Cpy-spy) is substituted for pmday (step S256D). If none of the above conditions are met,
{-Rot_x + (cpx-spx) / 2} is converted to pmda
x, and ｛−rot_y + (cpx−spx) /
2｝ is substituted for pmday (step S256E).

FIG. 39 is a flow chart of the destination setting for 90 ° rotation (FIG. 27, step S257). If 90 ° rotation is combined, it is necessary to change the destination criterion. If θ ₉₀ = 90 (YES in step S2571), pmday is set to p
mdx, and cpx-pmdax is substituted for pmday (step S2572). Also, θ ₉₀ = 180
(YES in step S2573), cpx-
Substituting pmdx into pmdx, cpx-pmday
Is substituted for pmday (step S2574). If θ ₉₀ = 270 (Y in step S2575)
ES), substituting cpy-pmday for pmdax, p
mmax is substituted for pmday (step S257)
6). In addition, the contents of the above destination setting are
Although the case of the normal mode has been described, the binding margin shift,
When combining with a saving copy that copies a plurality of originals into one, it is also possible to make settings in consideration of this.

FIG. 40 is a main flowchart of the third CPU 103 for controlling the memory unit 30. First, initial settings for initializing a RAM, a register, and the like are performed (step S31). Next, a command receiving process (step S32) and a status transmitting process (step S3)
3), image memory writing processing (step S34), compression control processing (step S35), decompression control processing (step S36), editing control processing (step S37, see FIG. 41), smoothing control processing (step S38), image memory The reading process (step S39) and other processes (step S3A) are performed. Next, step S
32, and repeats the above processing. FIG. 41 is a flowchart of the editing control (FIG. 40, step S37). Once the data for printing is ready (step S
371, YES), size parameters (cpx, cp
y) and (spx, spy) are set (step S37).
2) A 90 ° unit rotation parameter (90θ) is set (step S373), an arbitrary angle rotation parameter θ is set (step S374), a rotation coordinate parameter is set (step S375), and destination parameters (pmdax, pmday) are set. Is set (step S3).
76). Then, other parameters are set (step S377), and the process returns.

FIG. 42 shows a fourth CP for controlling the printer unit.
It is a main flowchart of U104. After the initialization for initializing the RAM and the registers is performed (step S41), an internal timer that defines the length of one routine is set (step S42). Next, control of the development / transfer system (step S43) and control of the transport system (step S43)
44), control of the fixing system (step S45), and control of the print processing unit (step S46). Then, other processing is performed (step S47). Next, after waiting for the end of the internal timer (YES in step S48), step S4
2 and repeat the above processing. FIG. 43 is a main flowchart of the fifth CPU 105 which controls the control of the entire copying machine. After the initialization for initializing the RAM and the registers is performed (step S51), an internal timer that defines the length of one routine is set (step S51).
52). Next, an input data analysis process for checking input data from another CPU (step S53), a mode setting process for determining an operation mode according to the operation content (step S54), an interrupt switching process (step S55), and a mode Command setting processing (step S56), output data setting for making the command wait in the communication port (step S57), and other processing (step S58).
do. Next, waiting for the end of the internal timer (YES in step S59), the process returns to step S52, and the above-described processing is repeated.

In the above, the arbitrary angle rotation processing of the image data input from the document reading apparatus has been described. Similarly, it is also possible to process image data (with angle parameters) transmitted from an external device on the network. . For example, image data input from an external device can be rotated by an arbitrary angle and printed by the print processing unit. Further, the image data read by the image reading device can be subjected to an arbitrary angle rotation process and output to an external device. In addition, image data input from an external device can be subjected to arbitrary angle rotation processing and output to the external device again.

[0051]

According to the present invention, when an original is placed on the original glass at an angle, even if the original is out of the reading area, the four corners of the original are estimated from the read image data, and the arbitrary angle rotation for correcting the inclination is performed. Various parameters required for can be calculated. As a result, even if the document is out of the reading area, the inclination angle is automatically corrected, and the originally required image is reproduced. Further, since various parameters can be set to the most reliable values based on the image data of the document, an optimal image can be obtained. If a reliable parameter cannot be obtained, the rotation at an arbitrary angle is prohibited, so that useless image reproduction can be avoided.

[Brief description of the drawings]

FIG. 1 is a schematic sectional view of a copying machine according to the present invention.

FIG. 2 is a plan view of an operation panel of the copying machine.

FIG. 3 is a diagram illustrating an example of a system configuration including a copying machine;

FIG. 4 is a block diagram of a part of a control unit of the copying machine.

FIG. 5 is a block diagram of a part of a control unit of the copying machine.

FIG. 6 is a block diagram of a controller.

FIG. 7 is a block diagram of an image signal processing unit.

FIG. 8 is a block diagram of a memory unit.

FIG. 9 is a diagram of a memory write operation.

FIG. 10 is a diagram of a memory read operation.

FIG. 11 is a diagram for explaining input of a document image by scanning.

FIG. 12 is a diagram illustrating an example of an image of a read image.

FIG. 13 is a diagram illustrating an example of an image of a read image.

FIG. 14 is a diagram illustrating an example of an image of a read image.

FIG. 15 is a diagram illustrating an example of an image of a read image.

FIG. 16 is a diagram illustrating an example of an image of a read image.

FIG. 17 is a diagram illustrating an example of an image of a read image.

FIG. 18 is a diagram illustrating an example of an image of a read image.

FIG. 19 is a block diagram of an edit processing unit in the memory unit.

FIG. 20 is a diagram for describing 90 ° rotation processing of image data.

FIG. 21 is a diagram for explaining bit swap processing of image data.

FIG. 22 is a diagram illustrating an arbitrary angle rotation process (shift shift process) of image data.

FIG. 23 is a diagram illustrating a process of simultaneously performing an arbitrary angle rotation process (shift shift process) and a scaling process on image data.

FIG. 24 is a main flowchart of the first CPU.

FIG. 25 is a main flowchart of the second CPU.

FIG. 26 is a flowchart of document detection.

FIG. 27 is a flowchart of edge extraction.

FIG. 28 is a flowchart of angle detection.

29 is a flowchart of an angle theta _a detection.

Is a flow chart of FIG. 30 the angle theta _b detected.

FIG. 31 is a flowchart of angle θ _c detection.

FIG. 32 is a flowchart of an angle theta _d detection.

FIG. 33 is a flowchart of size detection.

FIG. 34 is a flowchart of skew document size detection.

FIG. 35 is a flowchart of normal document size detection.

FIG. 36 is a flowchart of size setting.

FIG. 37 is a flowchart of setting rotation coordinates.

FIG. 38 is a flowchart of destination setting.

FIG. 39 is a flowchart of setting a 90 ° rotation destination.

FIG. 40 is a main flowchart of the third CPU.

FIG. 41 is a flowchart of editing control.

FIG. 42 is a main flowchart of the fourth CPU.

FIG. 43 is a main flowchart of the fifth CPU.

[Explanation of symbols]

30 memory unit section, 103 third CPU, 2
00 image reading unit, 304 image memory, 307
Editing section.

Claims (10)

[Claims] An image input means for inputting two-dimensional image data of an original placed on a reading area of an original glass, and image data of the original input by the image input means
Detecting means for detecting an area where the original placed on the original glass is present; and an arbitrary angle rotation for rotating the two-dimensional input image data by an arbitrary angle when the area of the original detected by the detecting means is inclined. Means for calculating the inclination angle of the document and the deviation of the position of the document from the coordinates of the area of the document detected by the detecting means when the document is out of the reading area, and based on these, the arbitrary angle rotation means An image forming apparatus comprising: a control unit that operates the image forming apparatus. 2. The image forming apparatus according to claim 1, wherein the control unit detects a length of a side of the document in the reading area from coordinates of the document area, and determines the inclination angle and the reading area. Set the length of the side outside the reading area from the length of the side of the original inside,
An image forming apparatus which calculates a paper size based on these, and operates the arbitrary angle rotation unit based on the inclination angle, the paper size and the deviation. 3. The image forming apparatus according to claim 2, wherein the control unit extracts a tilt angle by regarding a straight line obtained from coordinates of the document area as each side of the document area. Characteristic image forming apparatus. 4. The image forming apparatus according to claim 2, wherein the control unit calculates the inclination angle from a longest one of the sides of the extracted document. Image forming device. 5. The image forming apparatus according to claim 2, wherein the control unit does not operate the arbitrary-angle rotating unit when a side indicating the document area cannot be extracted. Image forming apparatus. 6. The image forming apparatus according to claim 2, wherein the control unit is significantly different from other calculation results among the respective inclination angles calculated from each side of the extracted document area. An image forming apparatus characterized in that when there is an angle, the arbitrary angle rotation unit is not operated. 7. The image forming apparatus according to claim 2, wherein said control means includes a document area detected by said document detection means, and a rectangular area having an extracted side as one side. An image forming apparatus for calculating a size. 8. The image forming apparatus according to claim 7, wherein the control unit does not operate the arbitrary-angle rotating unit when the calculated paper size is larger than a maximum paper size. Image forming apparatus. 9. The image forming apparatus according to claim 8, further comprising: a 90 ° rotation unit configured to rotate the image by 90 °, wherein the control unit determines that the calculated sheet size is larger than a maximum sheet size. An image forming apparatus characterized by operating a 90 ° rotation unit together with an arbitrary angle rotation unit when the length and width of the paper are switched to be smaller than the maximum paper size. 10. The image forming apparatus according to claim 1, wherein, when there are two or more adjacent sides among the extracted sides, the control unit determines an image based on the calculated size and the calculated paper size. An image forming apparatus for calculating a parallel movement amount.