JP2017200042A - Image forming device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image forming device capable of increasing a processing speed.SOLUTION: An image forming device 10 comprises: a skew block reading part 12 that sequentially reads out an image in a block unit smaller than a size of one page while changing a start-point coordinate in accordance with an inclination corrected by a memory 20 to which an input image is stored; a block inclination correction part 14 that corrects the inclination of each image of the block unit which is sequentially read by the skew block reading part 12, and outputs the image of which the inclination is corrected in the block unit; an extension/contraction filter processing part 16 that performs processing including at least extension/contraction processing of an arbitrary multiplication to the image of the block unit outputted by the block inclination correction part 14; a block writing-out part 18 that sequentially writes the image processed in the extension/contraction filter processing part 16 to the memory 20.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an image forming apparatus, and is suitable for application to an image forming apparatus having image processing functions such as skew correction (tilt correction) and filter processing.

  Conventionally, in an image forming apparatus such as a copying machine or a multifunction peripheral, an image forming apparatus that performs skew correction for the purpose of eliminating a skew of a document generated at the time of reading has been proposed (for example, see Patent Document 1).

  As a method of performing the skew correction, the image is read out from the memory in a block unit obtained by cutting out a part of the image into a rectangle, the block unit image is temporarily stored in the buffer, and the data stored in the buffer is changed according to the inclination. By reading out and correcting the inclination of the image in block units, writing the corrected image to the memory in block units, determining the size of the read block so that there is no overlap of blocks at the time of export, There is known a method of obtaining an image with corrected inclination by performing the entire image while changing the start coordinates of the read block in accordance with the inclination to be corrected. In the image forming apparatus, the memory for holding the entire image is configured with a large-capacity low-speed memory, and the buffer used for correcting the tilt of the image is configured with a small-capacity high-speed memory. Corrects tilt at high speed. This large-capacity low-speed memory is composed of, for example, a DRAM (Dynamic Random Access Memory) or the like, and has a low cost, high-speed sequential access but low-speed random access. The small-capacity high-speed memory is composed of, for example, an SRAM (Static Random Access Memory) or the like, and has a high cost, and high-speed sequential access and random access. Hereinafter, the large-capacity low-speed memory is denoted as a memory, and the small-capacity high-speed memory is denoted as a buffer.

  As a general process of the image forming apparatus, there is a filter processing group for the purpose of eliminating blur of a read original and suppressing the occurrence of moire due to screen interference between the read original and the output original. In addition to such processing, general processing of the image forming apparatus includes enlargement / reduction processing at an arbitrary magnification for the purpose of enlargement / reduction to a magnification specified by the user. Some image forming apparatuses that perform skew correction perform three processes on a read document image: skew correction, filter processing for the purpose of suppressing the occurrence of moire, and enlargement / reduction processing to an arbitrary magnification. .

JP 2005-352703 A

  In such an image forming apparatus that performs skew correction, processing of skew correction and enlargement / reduction processing to an arbitrary magnification may be performed on a read document image. In this case, the image forming apparatus performs an enlargement / reduction process to an arbitrary magnification on the image after skew correction written in the memory. Therefore, in the processing configuration for performing the filter processing on the image after skew correction written in the memory, memory access by reading and writing the image related to the skew correction is increased by the amount of the original image, and therefore only the enlargement / reduction processing without performing the skew correction. There was a problem that the processing was slower than the case of performing.

  The present invention has been made in consideration of the above points, and intends to propose an image forming apparatus capable of speeding up the processing.

  In order to solve such a problem, in the image forming apparatus of the present invention, the skew is sequentially read out in units of blocks smaller than the size of one page while changing the start point coordinates in accordance with the inclination to be corrected from the memory in which the input image is stored. The block reading unit, the block inclination correcting unit that corrects the inclination of each block-by-block image sequentially read by the skew block reading unit, and outputs the corrected image for each block, and the block inclination correcting unit output An enlargement / reduction filter processing unit that performs processing including at least an enlargement / reduction process at an arbitrary magnification and an image writing unit that sequentially writes images processed by the enlargement / reduction filter processing unit to a memory are provided. .

  According to the present invention, even when the scaling process is included in the filter process, the image tilt correction process and the filter process can be executed in a linked manner without using a memory. Therefore, it is possible to prevent the processing from becoming slower than when only the filter processing is performed without performing the inclination correction processing.

  According to the present invention, an image forming apparatus capable of speeding up the processing can be realized.

1 is a block diagram illustrating a configuration of an image forming apparatus. It is a figure which shows the input image and output image on memory. It is a figure which shows the operation | movement outline | summary of a skew block reading part. It is a figure which shows the operation | movement outline | summary of a block inclination correction part. It is a figure which shows the coordinate movement in a block inclination correction part. It is a figure which shows a X direction difference value and a Y direction difference value. It is a figure which shows the structure of the filter process part with expansion / contraction. It is a flowchart which shows the filter processing procedure with expansion / contraction. It is a figure which shows the operation | movement outline | summary of a block writing part. It is a figure which shows the coordinate of the input / output block of the filter process part with expansion / contraction. It is a figure which shows determination of the read start block when there is no filter process part. It is a figure which shows determination of the read start block in case there exists a filter process part. It is a figure which shows determination of the next block starting point coordinate. It is a flowchart which shows a block inclination correction process procedure (1). It is a flowchart which shows a block inclination correction process procedure (2). It is a flowchart which shows a block inclination correction process procedure (3). It is a flowchart which shows the block inclination correction process sequence by other embodiment. It is a flowchart which shows the stridew calculation process sequence by other embodiment. It is a flowchart which shows the strideh calculation process procedure by other embodiment.

  Hereinafter, modes for carrying out the invention (hereinafter referred to as embodiments) will be described with reference to the drawings.

[1. Embodiment]
[1-1. Configuration of image forming apparatus]
As shown in FIG. 1, the image forming apparatus 10 is a copier, a multifunction machine, or the like, and includes a control unit 11, a skew block reading unit 12, a block inclination correcting unit 14, an enlargement / reduction filter processing unit 16, a block writing unit 18, and a memory 20. Consists of. The control unit 11 is configured around a CPU (Central Processing Unit) (not shown), and controls and controls the image forming apparatus 10 by reading and executing a predetermined program from the memory 20. The skew block reading unit 12 reads the input image IMGi (FIG. 2) placed in the memory 20 in units of blocks while changing the start point coordinates in accordance with the inclination to be corrected, and inputs the block unit image to the block inclination correcting unit 14. .

  FIG. 2A shows an example of an input image IMGi placed on the memory 20, and a cross-hatched area is a document area AP indicating a document, and is a rectangular document image in a tilted state. The image forming apparatus 10 corrects the inclination of the inclined rectangular document image so that each side is along the X direction or the Y direction as shown in FIG. 2B, and outputs the image as an output image IMGo.

  FIG. 3 shows a state in which the skew block reading unit 12 performs reading in units of blocks, taking the input image IMGi of FIG. 2 as an example. 3 is a read block BLr corresponding to a block unit image read by the skew block reading unit 12, and a solid line frame block is a unit of output block BLo1 (FIG. 4) output by the block inclination correction unit 14. This is a block corresponding to an image area. The skew block reading unit 12 reads a rectangular range circumscribing each block input to the block inclination correcting unit 14 in units of the reading block BLr. Further, the skew block reading unit 12 starts reading from the reading block BLr corresponding to the upper left of the output of the document area AP as indicated by the arrow in the drawing as the reading order of each block, and when the reading of the reading block BLr is completed. The read block BLr corresponding to the output right direction of the read block BLr is read, and the read block BLr corresponding to the output right end is sequentially read. When the reading of the reading block BLr corresponding to the right end of the output is completed, the skew block reading unit 12 descends one step downward in the output, repeats these steps, and reads the entire image corresponding to the entire document area AP in units of the reading block BLr. . Further, the skew block reading unit 12 reads pixels in each reading block BLr in order of raster scanning from the upper left pixel. That is, the skew block reading unit 12 sequentially reads from the upper left to the right, goes down one step when reaching the end, repeats this, and reads all the pixels of the read block BLr. A method for determining the size of the read block BLr and the start point coordinates of each read block BLr will be described later.

  The block inclination correcting unit 14 corrects the inclination of each image in the read block BLr unit read and input by the skew block reading unit 12 and outputs each image in which the inclination is corrected in block units.

  FIG. 4 shows an outline of the inclination correction performed by the block inclination correction unit 14. The input block BLi1 in FIG. 4 corresponds to the read block BLr in FIG. As shown in FIG. 4, from the pixel group corresponding to the output area Ao surrounded by the solid line frame in the input block BLi1, as shown in the right diagram of FIG. Output as BLo1. Since the output order of the pixels of the output block BLo1 is a raster scan order, and the output of the output block BLo1 is output in order starting from the upper left pixel, the block inclination correction unit 14 is shown by the diagonal lines in FIG. It is necessary to have a buffer Bf having a size corresponding to the inclination to be corrected and the width of the input block BLi1 in the X direction (lateral direction).

  As described above, the block inclination correcting unit 14 requires a buffer Bf having a larger size as the inclination to be corrected is larger. Therefore, when the size of the buffer Bf is constant, the block inclination correcting unit 14 is inscribed in the input block BLi1 when the correction inclination is increased. When the size of the output area Ao, that is, the processable image is reduced, and the inclination to be corrected is reduced, the size of the output area Ao that is inscribed in the input block BLi1, that is, the processable image is increased. In addition, the size of the buffer Bf of the block inclination correcting unit 14 is determined so that an image of a block unit of the maximum width that can be processed by the enlargement / reduction filter processing unit 16 can be output when the correction inclination is smaller than 45 degrees. The

  The block inclination correcting unit 14 accumulates the image of the input block BLi1 input in the raster scan order in the buffer Bf, and the image of the input block BLi1 accumulated in the buffer Bf continuously connects pixels corresponding to one line of the output block BLo1. At the stage where only the data can be read out, the pixel corresponding to one line of the output block BLo1 is read in the tilt direction (arrow direction in the left diagram of FIG. 4), and one output line is generated and output. When the output of one line is completed, the block inclination correcting unit 14 assigns the buffer Bf that is no longer involved in the output to the accumulation of the unretained input pixels, and outputs the next line and repeats this. Thus, all the pixels corresponding to the output block BLo1 are sequentially generated and output.

FIG. 5 is a diagram for more specifically explaining the reading in the tilt direction. As shown in FIG. 5, the output corresponding coordinate Co that is the coordinate of each pixel corresponding to the output is inclined with respect to the input corresponding coordinate Ci that is the coordinate corresponding to each pixel of the input, and the block inclination correcting unit 14 When the pixel to be output is moved rightward, as shown in FIG. 5A, the coordinates of the next pixel Pn are added by adding Δx1 and Δy1 to the X and Y coordinates of the current pixel Pp. Can be requested. Δx1 and Δy1 are amounts of movement obtained by decomposing one-pixel movement vector Δxy1 corresponding to the horizontal direction of the output into the X direction and the Y direction. (Denoted as FIG. 2)) in the X and Y directions of the upper right and upper left vertex coordinates of the document area AP corresponding to the output image IMGo (FIG. 2B) in the input image IMGi shown in FIG. It can be obtained based on the difference values (hereinafter referred to as diwx and diwy, respectively) and can be expressed as the following equation.
Δx1 = diwx / outw
Δy1 = diwy / outw
When moving the output pixel downward, the block inclination correcting unit 14 sets Δx2 and Δy2 with respect to the X coordinate and the Y coordinate of the current line pixel Ppl as shown in FIG. 5B. The coordinates of the next line pixel Pnl can be obtained by addition. .DELTA.x2 and .DELTA.y2 are movement amounts obtained by decomposing one-pixel movement vector .DELTA.xy2 corresponding to the vertical direction of the output in the X direction and Y direction, and the number of pixels in the vertical direction of the entire document to be output (hereinafter referred to as outh (FIG. 2), and difference values in the X direction and Y direction of the vertex coordinates of the lower left and upper left of the document area AP corresponding to the output image IMGo (FIG. 2B) in the input image IMGi shown in FIG. (Hereinafter referred to as “dihx” and “dihy”, respectively), and can be expressed as the following equation.
Δx2 = dihx / outh
Δy2 = dihy / outh
Δx2 and Δy2 are Δx1 and Δy1, the resolution in the X direction and the resolution in the Y direction of the input image IMGi (hereinafter referred to as inxresol and inyresol), and the resolution in the X direction of the output image IMGo. And the resolution in the Y direction (hereinafter referred to as outxresol and outyresol) can also be expressed as the following equation.
Δx2 = −Δy1 × (inxresol / inyresol) / (outyresol / outxresol)
= (-Diwy * inxresol / inyresol) / (outw * outyresol / outxresol)
Δy2 = Δx1 × (inresol / inxresol) / (outyresol / outxresol)
= (Diwx x inyresol / inxresol) / (outw x outyresol / outxresol)

  If the input resolution and output resolution are in a multiplying relationship, the denominator and numerator of the above equation can be normalized using the minimum resolution. For example, if the X-direction resolution of the input image IMGi is 600, the Y-direction resolution is 600, the X-direction resolution of the output image IMGo is 600, and the Y-direction resolution is 600, the above equation can be normalized as the following equation (1) If the input image IMGi has an X-direction resolution of 300, a Y-direction resolution of 600, and the output image IMGo has an X-direction resolution of 600 and a Y-direction resolution of 600, the above equation can be normalized as the following equation (2): .

Formula (1)
Δx1 = diwx / outw
Δy1 = diwy / outw
Δx2 = −diwy / outw
Δy2 = diwx / outw
Formula (2)
Δx1 = (diwx × 2) / (outw × 2)
Δy1 = (diwy × 2) / (outw × 2)
Δx2 = −diwy / (outw × 2)
Δy2 = (diwx × 4) / (outw × 2)

  By using the above equation, the values corresponding to the decimal point of the X-direction coordinate and the Y-direction coordinate of the coordinate on the input image IMGi (the value of the denominator of the equation), the X direction of each of the horizontal movement and the vertical movement of the output, and The denominator value of the movement amount in the Y direction can be shared by the X direction and the Y direction. In addition, in order to simplify the following explanation about the value of the numerator and denominator of the above formula, the values corresponding to the numerators of Δx1, Δy1, Δx2, and Δy2 are vx1, vy1, vx2, and vy2, respectively. The common denominator of Δx1 and Δx2 is denoted as xvbase, and the common denominator of Δy1 and Δy2 is denoted as yvbase.

  Using the above-described Δx1, Δy1, Δx2, and Δy2, the block inclination correcting unit 14 outputs one pixel from the coordinates corresponding to the output start point of each input block BLi1 in each input block BLi1. Are sequentially determined in the raster scan order, and pixels corresponding to the size of the predetermined output block BLo1 are generated and output. Note that the size of the output block BLo1, the output start point corresponding coordinate Ccos of the input block BLi1 corresponding to the output start point coordinate Cos of each output block BLo1, the method of determining the size of the buffer Bf provided, and more detailed description of pixel generation will be described later. To do.

[1-2. Configuration of filter processing unit with scaling]
The enlargement / reduction filter processing unit 16 includes a process that requires at least peripheral pixel reference for the input block BLi2 unit image input from the preceding block inclination correction unit 14, and a process including an enlargement / reduction process at an arbitrary magnification. And output as an output block BLo2 as shown in FIG. For example, in the case of a copying machine or a multi-function peripheral, the scale of an arbitrary magnification is reduced for the purpose of eliminating blurring of a scanned document, suppressing moiré due to screen interference between a scanned document and an output document, or converting an image to a user-specified magnification. Filter processing including processing is performed. FIG. 7A shows a configuration example of various filter processes including an enlargement / reduction process at an arbitrary magnification. The enlargement / reduction filter process unit 16 performs image area separation in which a determination is made for each character other than a character part and a character. The filter 30, the smoothing filter 32 that performs smoothing on pixels determined to be other than characters by the image area separation filter 30, the block enlargement / reduction unit 40 that performs enlargement / reduction processing at an arbitrary magnification, and the image area separation filter 30 The emphasis filter 34 is used to emphasize the pixels determined to be characters. The image area separation filter 30 and the smoothing filter 32 detect and smooth a specific pattern in the input image. Since the processing is based on an input image having a certain resolution, an enlargement / reduction process at an arbitrary magnification is performed. You need to do some processing before you do it. In order to process one pixel of interest for each filter, peripheral pixel reference is necessary. Taking the smoothing filter 32 as an example, processing of one pixel of interest includes 9 × 9 pixels in the vicinity including the pixel of interest. An average value is calculated, and the output of the average value as the output pixel value of the target pixel is performed for each input pixel. In the filter processing that requires peripheral pixel reference, the line buffer LBf for holding the preceding and succeeding pixels becomes large when processing is performed in units of lines of one page. By adopting a configuration having only the line buffer LBf for the input block BLi2 divided in the direction (X direction), the size of the line buffer LBf can be reduced, and the cost for configuring the filter can be suppressed. When such processing is performed, for the pixels in the vicinity of the boundary of the input block BLi2 (FIG. 7) corresponding to the output block BLo1 (FIG. 4), the pixel to be referenced is out of the boundary, resulting in an invalid pixel output. Therefore, the enlargement / reduction filter processing unit 16 needs to input in advance an input block BLi2 that is large in size corresponding to invalid pixels, and delete invalid pixels from the pixels after each filter process or all filter processes to obtain only valid pixels. There is. For example, in the example of the enlargement / reduction filter processing unit 16 in FIG. 7, invalid pixels having a width of 14 pixels (image region separation filter 5 pixels + smoothing filter 4 pixels + enhancement filter 5 pixels) are input to the boundary of the upper, lower, left, and right input blocks BLi2. Therefore, it is necessary to input the input block BLi2 having a size larger by 14 pixel widths to the enlargement / reduction filter processing unit 16 in advance. Further, since the overhead of reading increases when the proportion of invalid pixels with respect to the input block BLi2 to be input increases, the line buffer LBf included in the enlargement / reduction filter processing unit 16 becomes too small so that the overhead is reduced. It is necessary to decide not to. In the present embodiment, the case where the width of the line buffer LBf of the enlargement / reduction filter processing unit 16 is defined as 284 (the effective output width is 256) so that the overhead is about 10% when the vertical and horizontal output sizes are the same. The following explanation will be made on the assumption. The overhead is the number of pixels that need to be read out by an amount corresponding to invalid pixels, and the overhead in this case is obtained by 284 × 284−256 × 256.

  The block enlargement / reduction unit 40 performs enlargement / reduction processing on a specified size for an image input in units of blocks. The block enlargement / reduction unit 40 includes an X-direction enlargement / reduction unit 42 that performs enlargement / reduction in the X direction and a Y-direction enlargement / reduction unit 44 that performs enlargement / reduction in the Y direction. The direction is enlarged and reduced, and then the Y direction is enlarged and reduced.

  The X-direction enlargement / reduction unit 42 includes an X-direction enlargement / reduction counter and an X-direction output counter. The X-direction enlargement / reduction counter is set to a counter initial value determined in advance for each block at the head of the line (line head of the block image). The set block image is enlarged / reduced according to the X direction enlargement / reduction counter, the number of output pixels in the X direction is counted by the X direction output counter, and the X direction output counter is within a predetermined range of a predetermined value. By outputting only the pixels, the inputted block image is enlarged / reduced in the X direction, and the block image enlarged / reduced in the X direction is inputted to the Y direction enlargement / reduction unit 44.

  Similar to the X direction enlargement / reduction unit 42, the Y direction enlargement / reduction unit 44 includes a Y direction enlargement / reduction counter and a Y direction output counter, and blocks the Y direction enlargement / reduction counter at the first line (first line of the block image). A counter initial value determined in advance is set every time, and the input block image is enlarged / reduced according to the Y-direction enlargement / reduction counter, and the number of output lines in the Y-direction is counted by the Y-direction output counter. Output only the lines within a predetermined value range, so that the input block image is enlarged / reduced in the Y direction and output. The Y-direction enlargement / reduction unit 44 needs to have a plurality of line memories when performing an enlargement / reduction method other than simple reduction (for example, interpolation enlargement / reduction method).

[1-3. Scaled filter processing procedure]
The details of the enlargement / reduction processing in the X direction and the Y direction in the enlargement / reduction filter processing unit 16 will be described with reference to the flowchart shown in FIG. explain.

  In step SP301, the X-direction enlargement / reduction unit 42 sets 0 in the output counter (rxocnt) at the head of the line, and sets the initial value (rxcnt0) of the enlargement / reduction counter for each block image in the enlargement / reduction counter (rxcnt). Proceed to

  In step SP302, the X-direction enlargement / reduction unit 42 determines whether the value of the enlargement / reduction counter (rxcnt) is smaller than the numerator value (rxratio) of the enlargement / reduction magnification. If it is smaller, the process proceeds to step SP303 and outputs the pixel. Otherwise, the process proceeds to step SP305 without outputting the pixel. In step SP303, the X-direction enlargement / reduction unit 42 outputs the pixel, adds 1 to the output counter (rxocnt), and further adds the denominator value (rxbase) of the enlargement / reduction magnification to the enlargement / reduction counter (rxcnt). Proceed to step SP304. Here, the output pixel value is determined by the following equation, for example, in the case of the interpolation enlargement / reduction method.

  (Output pixel value) = ((Current pixel value) × ((Magnification numerator value) − (Enlargement / reduction counter value)) + (Next pixel value) × (Enlargement / reduction counter value)) / (Molecular value of magnification)

  In step SP304, the X-direction enlargement / reduction unit 42 determines whether the value of the enlargement / reduction counter (rxcnt) updated in step SP303 is equal to or larger than the numerator value (rxratio) of the enlargement / reduction magnification. The process proceeds to step SP305, where the input pixel position is updated. Otherwise, the process proceeds to step SP306.

  In step SP305, the X-direction enlargement / reduction unit 42 subtracts the numerator value (rxratio) of the enlargement / reduction magnification from the value of the enlargement / reduction counter (rxcnt), and updates the input pixel position (the pixel immediately to the right of the current pixel). To the next current pixel), the process proceeds to step SP306. In step SP306, the X-direction enlargement / reduction unit 42 determines whether or not the value of the output counter (rxocnt) has reached the number of output pixels (rxoutw) of one line. The pixel line is processed, and if not reached, the process returns to step SP302. The X-direction enlargement / reduction unit 42 performs enlargement / reduction in the X direction by performing the above processing on all the pixels of the block image.

  The block writing unit 18 writes the image processed by the pre-enlargement / reduction filter processing unit 16 to the memory 20 in units of writing blocks BLw shown in FIG. As described above, the input order of each writing block BLw is input from the upper left writing block BLw in the raster scan order, and the pixel inputs in the writing block BLw are also input in the raster scan order. When pixels outside the output document area (outside the output image IMGo) are included in the input block, the block writing unit 18 writes only an image other than the pixels outside the document area to the memory 20 (that is, cross-hatching in FIG. 9). All pixels are written in the memory 20 while only the pixels in the original area are written into the memory 20, thereby obtaining an image of the entire original.

[1-4. Parameter calculation method]
Hereinafter, a method for calculating parameters necessary for the operation of the image forming apparatus 10 will be described. FIG. 10 is a diagram illustrating the block image input to the enlargement / reduction filter processing unit 16 and the start point coordinates of each block image output from the enlargement / reduction filter processing unit 16. The hatched rectangular frame in the figure corresponds to the block image of the output block BLo2 output from the enlargement / reduction filter processing unit 16, and the width and height of the block image of the output block BLo2 are defined as foutbw and foutbh, respectively. When the X coordinate and Y coordinate of the output block start point coordinate CBLos that is the start point coordinate of the output block BLo2 that is output first are each 0, the next output block that is the start point coordinate of the output block BLo2 (not shown) adjacent to the right The start point coordinates CBLons are coordinates shifted by the width (foutbw) of the block image output by the enlargement / reduction filter processing unit 16, and the next stage output block start point coordinates which are the start point coordinates of the output block BLo2 (not shown) adjacent thereto below CBLoncs are coordinates shifted by the height (foutbh) of the block image output from the enlargement / reduction filter processing unit 16, and the width and height (foutbw and foutbh) of the block image output from the enlargement / reduction filter processing unit 16 are constant. Since it is a value, the enlargement / reduction filter processing unit 16 starts the subsequent output block in each direction. Coordinates can also be obtained by going to simply adding the width and height of the output block image.

  The input block BLi2 input to the enlargement / reduction filter processing unit 16 needs to input a block size that is larger by an invalid pixel generated by each filter processing in the enlargement / reduction filter processing unit 16, and the start point coordinates of the input block BLi2 accordingly. The input block start point coordinates CBLis (fxtop and fytop) are shifted from the output block start point coordinates CBLos. In order to obtain the input block start point coordinate CBLis, the enlargement / reduction filter processing unit 16 obtains the enlargement / reduction unit input start point coordinate Cis (rxtop and rytop) needs to be determined first. The enlargement / reduction unit input start point coordinates Cis are the denominator and numerator values (rxbase, rybase, rxratio, and ryratio) of the enlargement / reduction factor of arbitrary magnification in each direction, and the filter processing (enhancement filter 34) subsequent to the block enlargement / reduction unit 40 Can be calculated using the widths of invalid pixels (obnw2 and obsnh2) generated in the following equation.

Formula (10)
rxtop = floor [− (obnw2 × rxbase) / rxratio]
rytop = floor [− (obnh2 × rybase) / ryratio]

  Using the above-mentioned rxtop and rytop and the width of invalid pixels (obnw1 and obnh1) generated in the previous stage filter processing (image area separation filter 30 and smoothing filter 32) of the block enlargement / reduction unit 40, processing is performed at the beginning of the page. The starting point coordinate (input block starting point coordinate CBLis) of the input block BLi2 to the enlargement / reduction filter processing unit 16 is obtained by the following equation.

Formula (11)
fxtop = rxtop-obnw1
fytop = rytop-obnh1

  Next, for the input block BLi1 of the enlargement / reduction filter processing unit 16, the next input block start point coordinate CBLins which is the start point coordinate of the next block (rightward block) and the next coordinate which is the start point coordinate of the next block (downward block). A method for calculating the stage input block start point coordinates CBLincs will be described.

  As described above, the start point coordinates of the output block BLo1 can be obtained by simply increasing foutbw in the X direction and foutbh in the Y direction, and the start point coordinates of each input block BLi1 on the input side are moved accordingly. It can be obtained by doing. However, since the coordinate value on the input side needs to be converted into the coordinate value before scaling at an arbitrary magnification in the block scaling unit 40, in order to process seamlessly between blocks, the coordinate value after the decimal point is also considered. Need to ask. The coordinate values below this decimal point correspond to the scaling counter initial values rxcnt0 and rycnt0 in the above-described scaling processing with the filtering process (FIG. 8). Can be obtained.

Formula (12)
rxcnt0 = (rxratio− (obnw2 × rxbase)% rxratio)% rxratio
rycnt0 = (ryratio− (obnh2 × rybase)% ryratio)% ryratio

  The initial value of the enlargement / reduction counter in the X direction of the next block (right block) and the initial value of the enlargement / reduction counter in the Y direction of the next block (downward block) are the input block BLi1 positioned on the left and upper, respectively. Using the counter initial value used, it can be obtained by the following formula.

Formula (13)
rxcnt0 ′ (initial value of enlargement / reduction counter in the X direction of the next block)
= (Foutbw × rxbase + rxcnt0)% rxratio
rycnt0 ′ (initial value of the enlargement / reduction counter in the Y direction of the next block)
= (Foutbh × rybase + rycnt0)% ryatio

  In addition, the enlargement / reduction unit next block input start point coordinate Cins which is the X-direction coordinate position of the next block and the enlargement / reduction unit next-stage block input which is the Y-direction coordinate position of the next block of the enlargement / reduction unit input block BLi3 of the block enlargement / reduction unit 40 The starting point coordinates Cincs is expressed by the following equation using the initial counter values used in the input block BLi1 positioned on the left and on the upper side, respectively.

Formula (14)
rxtop '(X-direction coordinate position of the next block) = rxtop
+ Floor [(foutbw × rxbase + rxcnt0) / rxratio]
rytop '(Y-direction coordinate position of the next block) = rytop
+ Floor [(foutbh × rybase + rycnt0) / ryratio]

  The right term on the right side of the above expression is the amount of movement of the coordinates of the start point of the block on the input side, and the X-direction coordinate position of the next block of the input block BLi1 of the enlargement / reduction filter processing unit 16 and the Y-direction coordinate position of the next-stage block As for the above, it can be obtained by the following equation similarly to the above equation.

Formula (15)
fxtop ′ (X coordinate position of the next block) = fxtop
+ Floor [(foutbw × rxbase + rxcnt0) / rxratio]
fytop '(the Y-direction coordinate position of the next block) = fytop
+ Floor [(foutbh × rybase + rycnt0) / ryratio]

[1-5. Operation of skew block reading unit and block inclination correcting unit]
Next, operations of the skew block reading unit 12 and the block inclination correcting unit 14 will be described. Since the skew block reading unit 12 and the block inclination correcting unit 14 perform an enlargement / reduction process at an arbitrary magnification at a later stage, it is necessary to consider when obtaining each parameter value necessary for the operation.

  FIG. 11 and FIG. 12 explain details of a read start block which is a block that is first read by the skew block reading unit 12. FIG. 11 illustrates the read block BLr when the enlargement / reduction filter processing unit 16 is not provided, and FIG. 12 illustrates the read block BLr when the enlargement / reduction filter processing unit 16 is provided. In the present embodiment, the configuration with the enlargement / reduction filter processing unit 16 is used as the embodiment. However, for the sake of clarity, the configuration without the enlargement / reduction filter processing unit 16 is shown in FIG. 11 (B).

  A frame indicated by a solid line in FIG. 11 is an output area Ao that is an area corresponding to the output of the block inclination correction unit 14 at the subsequent stage, and a rectangular frame indicated by a broken line in FIG. 11 circumscribing the output area Ao as described above. Is set as the read block BLr, and the skew block read unit 12 performs read. The block start point coordinates CBLs (xtop and ytop), which are the start point coordinates of the read block BLr, are obtained when vy1 is negative (slope 0 to −45 degrees) (FIG. 11A) and when vy1 is positive (slope 0 to 0). 45 degrees) (FIG. 11B), the sides of the read block BLr that are in contact with the output start point corresponding coordinates Ccos (FIG. 4) of the block inclination correcting unit 14 are different. Using the CAPs (xorg and yorg), the output block size (outbw and outbh) of the subsequent block inclination correction unit 14, and values (vx1, vy1, vx2, vy2, xvbase and yvbase) related to the movement amount, the following equation (3) ) To determine.

Formula (3)
If vy1 is negative:
xtop = xorg
ytop = yorg + vy1 × (outbw−1) / yvbase
If vy1 is positive:
xtop = xorg + vx2 × (outbh−1) / xvbase
ytop = yorg

  Next, the configuration of the present embodiment in which the enlargement / reduction filter processing unit 16 is provided will be described with reference to FIGS. Similar to the configuration without the previous enlargement / reduction filter processing unit 16, a frame indicated by a solid line in FIG. 12 is an output region Ao in an area corresponding to the output of the block inclination correction unit 14 at the subsequent stage, and as described above, this output area Ao A circumscribed rectangular frame indicated by a broken line in FIG. 12 is set as a reading block BLr, and the skew block reading unit 12 performs reading. Here, the output area Ao of the block inclination correcting unit 14 is large in size corresponding to the invalid pixels generated by the processing of the enlargement / reduction filter processing unit 16, and the output effective area Aov which is an area that finally becomes an effective output is In FIG. 12, the frame area is indicated by a one-dot chain line, and the document start point coordinates CAPs are present at the upper left of the output effective area Aov. Therefore, the skew block reading unit 12 needs to read the starting point coordinates of the reading block BLr from a position shifted from the original starting point coordinates CAPs by invalid pixels, the original starting point coordinates CAPs (xorg and yorg), and the subsequent block inclination correcting unit. In addition to 14 output block sizes (outbw and outbh) and values related to movement amounts (vx1, vy1, vx2, vy2, xvbase and yvbase), they are generated at the left and top edges of the block in the processing of the expansion / contraction filter processing unit 16 It is determined by calculating with the following equation (4) using the width (obnw and obnh) of the invalid pixel.

  Here, the widths of invalid pixels (obnw and obsnh) generated at the left end and the upper end of the block of Expression (4) are the filter processing (image region separation filter 30 and smoothing filter 32) in the previous stage of the block enlargement / reduction unit 40. The width of invalid pixels (obnw1 and obnh1) generated, the width of invalid pixels (obnw2 and obnh2) generated in the subsequent filter processing (enhancement filter 34) of the block enlargement / reduction unit 40, and the scaling factor of arbitrary magnification The denominator and numerator values (rxbase, rybase, rxratio, and ryratio) must be used in the following equation.

Formula (16)
obnw = obnw1
+ Floor [(obnw2 × rxbase + rxratio−1) / rxratio]
obnh = obnh1
+ Floor [(obnh2 × rybase + ryratio-1) / ryratio]

Formula (4)
If vy1 is negative:
xtop = xorg + (− obnw × vx1−obnh × vx2) / xvbase
ytop = yorg + (− obnw × vy1−obnh × vy2 + by1 × (outbw−1)) / yvbase
If vy1 is positive:
xtop = xorg + (− obnw × vx1−obnh × vx2 + vx2 × (outbh−1)) / xvbase
ytop = yorg + (− obnw × vy1-obnh × vy2) / yvbase

  The skew block reading unit 12 sets the coordinates corresponding to the integer values of the X coordinate and the Y coordinate calculated by the above formula as the start point coordinates of the first read block BLr (for example, −2, 2.5 if −1.4). If there is, use 2).

  The size (inbw and inbh) of the read block BLr can be determined by calculating with the following equation (5).

Formula (5)
If vy1 is negative:
inbw = (vx1 × outbw + vx2 × outbh) / xvbase + α
inbh = (− vy1 × outbw + vy2 × outbh) / yvbase + α
If vy1 is positive:
inbw = (vx1 × outbw−vx2 × outbh) / xvbase + α
inbh = (vy1 × outbw + vy2 × outbh) / yvbase + α

  Α in the above equation corresponds to the pixel input required due to the influence of the interpolation operation (generated by interpolating the output pixel with 4 input pixels) in the process performed by the block inclination correcting unit 14 in the subsequent stage and the quantization of the start coordinate For the margin for this, a value greater than the required number of pixels (approximately 2 or more) may be used.

  Further, as shown in FIG. 13, the skew block reading unit 12 outputs the start point coordinates of the read block BLr adjacent to the left as to the next block start point coordinates CBLns that are the start point coordinates of the next block. The position shifted by the width (stridew) of the area Aov is determined by obtaining the position shifted downward by the height (strideh) of the output effective area Aov with respect to the downward movement. The left block start point coordinates CBLls (left_xtop and left_ytop) are the start point coordinates of the read block BLr adjacent to the left, and the start block coordinates CBLus (up_xtop and up_ytop) are the start point coordinates of the read block BLr adjacent to the upper side (6 ) To calculate.

  Here, stridew and strideh described as values corresponding to the width and height of the output effective area Aov in Expression (6) are the width and height (foutbw and foutbh) of the block image output by the filter processing unit 16 with scaling. ), The initial value (rxcnt0 and rycnt0) of the enlargement / reduction filter processing unit 16 in the corresponding block, and the denominator and numerator values (rxbase, rybase, rxratio, and ryratio) of the enlargement / reduction factor of arbitrary magnification It is necessary to find and use the following formula using.

Formula (17)
stridew = floor [(foutbw × rxbase + rxcnt0) / rxratio]
strideh = floor [(foutbh × rybase + lycnt0) / ryratio]

Formula (6)
(Block movement to the right)
xtop = left_xtop + stridew × vx1 / xvbase
ytop = left_ytop + stridew × vy1 / yvbase
(Block movement down)
xtop = up_xtop + strideh * vx2 / xvbase
ytop = up_ytop + strideh x vy2 / yvbase

  The size of the output block BLo1 of the block inclination correction unit 14, the output start point corresponding coordinates Ccos that are the coordinates corresponding to the output start point of each output block BLo1, and the size required for the buffer Bf provided are the size of the output block BLo1. The output start point corresponding coordinates Ccos of each input block BLi1 are the coordinates at the upper left of the output area Ao corresponding to the output of the block inclination correcting unit 14 indicated by the solid line frame in FIGS. 11 and 12 (xstart and ystart), and the required size of the buffer Bf provided is obtained from the difference value of the Y coordinate value of the upper side of the output area Ao corresponding to the output and the width (inbw) of the input block BLi1, and is calculated by the following equation (7). .

Formula (7)
Buffer size = (abs [vy1 × outbw / yvbase] + β) × inbw
If vy1 is negative:
xstart = 0
ystart = −floor [(ycnt00 + vy1 × (outbw−1)) / yvbase]
If vy1 is positive:
xstart = −floor [(xcnt00 + vx2 × (outbh−1)) / xvbase]
ystart = 0
Here, abs [] represents the absolute value in [].

Β in the above equation can be a value of approximately 2 or more as a margin to deal with pixel input required due to interpolation calculation (generated by interpolating output pixels with 4 input pixels) and quantization of start coordinates. It ’s fine. Further, floor [] is a floor function (for example, floor [-1.4] =-2, floor [2.5] = 2) for making an integer value, and xcnt00 and ycnt00 are output start points of each input block BLi1. This is a value corresponding to the decimal value of the coordinate in the X direction and the Y direction of the corresponding coordinate Ccos. Xcnt00 and ycnt00 of the first read block BLr can be calculated by the following equations.
xcnt00 = (− obnw × vx1-obnh × vx2)% xvbase
ycnt00 = (− obnw × vy1-obnh × vy2)% yvbase
("%" Represents the remainder operation)
Further, xcnt00 and ycnt00 of the next block (block movement to the right) and the next block (block movement to the bottom) are xcnt00 and ycnt00 of the read block BLr adjacent to the left as left_xcnt00 and left_ycnt00, and the read block BLr adjacent above. Xcnt00 and ycnt00 can be obtained by the following equation as up_xcnt00 and up_ycnt00.
(Block movement to the right)
xcnt00 =
(Left_xcnt00 + (stridew × vx1)% xvbase)% xvbase
ycnt00 =
(Left_ycnt00 + (stride × by1)% yvbase)% yvbase
(Block movement down)
xcnt00 =
(Up_xcnt00 + (strideh × vx2)% xvbase)% xvbase
ycnt00 =
(Up_ycnt00 + (strideh × vy2)% yvbase)% yvbase

  The image forming apparatus 10 obtains parameters necessary for the processing of each input block based on the values obtained above, and performs the following block inclination correction processing, thereby performing a filter including enlargement / reduction at an arbitrary magnification after skew correction. It is possible to cope with processing.

[1-6. Block inclination correction processing procedure]
Further, a more detailed description of pixel generation performed by the block inclination correction unit 14 will be described according to the flowcharts shown in FIGS. 14, 15 and 16 based on the values determined by the above formula.

  First, the block inclination correcting unit 14 obtains integer values (xstart and ystart) of the starting point coordinates corresponding to the output of each input block BLi1 from xcnt00 and ycnt00, and xcnt_lttop, ycnt_lttop, xpos_lttop and ypos_lttop of the line head coordinate value management parameters. And the Y-direction output line number counter (yocnt) is set to 0 (SP1).

  Next, the block inclination correcting unit 14 sets the current coordinate value management parameters xcnt, ycnt, xpos, and ypos, and sets the output pixel number counter (xocnt) in the X direction to 0 (SP2).

Next, the block inclination correcting unit 14 reads out pixels corresponding to the current coordinate value management parameters xpos and ypos from the buffer Bf, generates an output pixel, and outputs a corresponding one pixel (SP3). The block inclination correcting unit 14 calculates, for example, using the following expression when generating the output pixel by interpolating with the input four pixels.
Output pixel value =
(
(P [xpos, ypos] × (xvbase-xcnt) +
p [xpos + 1, ypos] × xcnt) / xvbase × (yvbase-ycnt)
+ (P [xpos, ypos + 1] × (xvbase-xcnt) +
p [xpos + 1, ypos + 1] × xcnt) / xvbase × ycnt) / yvbase
In the above equation, p [] is a pixel value corresponding to xpos, ypos.

  Next, the block inclination correcting unit 14 adds 1 to the output pixel number counter (xocnt) in the X direction, and adds vx1 and vy1 to xcnt and ycnt, respectively. When xcnt after addition of vx1 is equal to or greater than xvbase, xvbase is subtracted from xcnt and 1 is added to xpos. On the other hand, when xcnt after addition of vx1 is less than xvbase and xcnt is less than 0, the block inclination correction unit 14 adds xvbase to xcnt and subtracts 1 from xpos. The block inclination correcting unit 14 performs the same for ycnt. As a result, the coordinates of the next pixel (the pixel immediately adjacent to the current pixel) are calculated (SP4).

  Next, the block inclination correcting unit 14 determines whether xocnt matches with outbw, and if not, performs processing from SP3 (SP5).

  If xocnt matches outbw in SP5, the output of one line corresponding to the output is completed, so the block inclination correcting unit 14 adds 1 to the output line number counter (yocnt) in the Y direction. Then, the top coordinate of the next line is calculated (SP6).

  Next, the block inclination correcting unit 14 determines whether or not yocnt matches with outbh, and if not, performs processing from SP2, and repeats the processing until this determination matches (SP7).

  The block inclination correction unit 14 performs the above-described processing for each block of the input document by performing the above processing on each input block BLi1 input to the block inclination correction unit 14.

[1-7. Effect]
Here, the enlargement / reduction process at an arbitrary magnification can be realized by changing the output resolution of the block inclination correction unit 14, but in the filter process for the purpose of suppressing the occurrence of moire, an input image having a predetermined resolution is processed. Therefore, the image forming apparatus 10 performs a skew correction process and an enlargement / reduction process at an arbitrary magnification, respectively. There is a need.

  Such skew correction and scaling processing at an arbitrary magnification requires accurate coordinate management in each process, so an image obtained by either processing is output to a memory, and the other image is processed with respect to the other image. It is considered that a method of connecting and processing through a memory so as to perform processing is generally performed, and an increase in memory access and an increase in necessary memory occur. However, in order to connect the filter processing including the skew correction and the enlargement / reduction process at any magnification without using the memory, the accurate coordinate management necessary for each process of the skew correction and the enlargement / reduction process at the arbitrary magnification is simultaneously broken. It was necessary to do so and it was difficult to realize.

  On the other hand, the image forming apparatus 10 does not read the image from the memory 20 after writing the image after skew correction in the memory 20 and performs the filter process, but performs the input image IMGi on the memory 20 according to the inclination to be corrected. The image is read from each read block BLr, the inclination of the image is corrected for each read block BLr, and the image whose inclination is corrected is subjected to filter processing including enlargement / reduction processing to an arbitrary magnification with respect to the block unit image. Later, the filtered image is written into the memory 20. As a result, the image forming apparatus 10 can prevent memory access between the skew correction and the filter process including the enlargement / reduction process to the arbitrary magnification. Therefore, the filter process including the enlargement / reduction process to the arbitrary magnification without performing the skew correction. It can be suppressed that the processing is slower than the case where only the process is performed.

  Further, in the case of a conventional processing configuration that performs a filtering process including an enlargement / reduction process to an arbitrary magnification on an image after skew correction written in a memory, a filter process including a skew correction and an enlargement / reduction process to an arbitrary magnification Since an intermediate memory is required between them, the necessary memory is increased as compared with the case where only the filtering process including the scaling process to an arbitrary magnification is performed without performing the skew correction. On the other hand, since the image forming apparatus 10 does not require an intermediate memory between the skew correction and the filter process including the enlargement / reduction process to an arbitrary magnification, both the skew correction and the filter process including the enlargement / reduction process to the arbitrary magnification are performed. Even when this processing is performed, the memory required for the processing can be reduced.

  Further, the block inclination correcting unit 14 has a buffer Bf of a predetermined size, and performs correction of inclination using the buffer Bf. At this time, when the inclination to be corrected increases, the size of the processable image decreases, and correction is performed. The skew block reading unit 12 reads the image in units of the read block BLr with a size that can be processed by the block inclination correcting unit 14. I made it. That is, the image forming apparatus 10 sets the size of the input block BLi1 in accordance with the size of the buffer Bf used in the block inclination correcting unit 14 (that is, not to exceed the size).

  Further, the enlargement / reduction filter processing unit 16 includes a line buffer LBf corresponding to the processing of an image having a width smaller than the width of one page, and the skew block reading unit 12 can be processed by the block inclination correction unit 14. An image in a read block BLr unit is read in a size that can be processed by the enlargement / reduction filter processing unit 16. That is, the image forming apparatus 10 uses the block inclination correction unit 14 in accordance with the size of the line buffer LBf used in the enlargement / reduction filter processing unit 16 having a processing load larger than that of the block inclination correction unit 14 (that is, not to exceed the size). The size of the buffer Bf to be used is set.

  According to the above configuration, the image forming apparatus 10 buffers the image in units of the read block BLr smaller than the size of one page while changing the start point coordinates according to the inclination to be corrected from the memory 20 in which the input image IMGi is stored. The skew block reading unit 12 that sequentially reads the image and the inclination of each of the read block BLr units sequentially read by the skew block reading unit 12 are corrected, and the image whose inclination is corrected is output to the buffer Bf in block units as the output block BLo1. A block inclination correction unit 14 that outputs as a block, an enlargement / reduction filter processing unit 16 that performs a process including an enlargement / reduction process of at least an arbitrary magnification on an image of the output block BLo1 output from the block inclination correction unit 14, and an expansion / reduction filter process This block sequentially writes the images processed by the unit 16 into the memory 20. Tsu was to be provided and click writing unit 18.

  Thus, the image forming apparatus 10 can execute the skew correction process and the filter process by connecting them without using the memory even when the filter process includes an enlargement / reduction process at an arbitrary magnification. Therefore, it is possible to prevent the processing from being delayed compared to the case where only the filter processing is performed without performing the skew correction processing.

[2. Other Embodiments]
In the above-described embodiment, by adding SP10 shown in FIG. 17 before SP1 (FIG. 14), rxcnt00 and rycnt00 set the initial value of the enlargement / reduction counter initially input by the block enlargement / reduction unit 40. Thus, the initial value of the enlargement / reduction counter corresponding to each input block may be automatically set, and the stridew and strideh of each input block may be set based on the initial value of the enlargement / reduction counter. The strideh calculation subroutine in SP10 is the flowchart shown in FIG. 18, and the stridew calculation subroutine is the flowchart shown in FIG. In this case, it is not necessary to calculate stridew and strideh for each read block BLr, and the processing can be simplified.

  In the above-described embodiment, the case where the enlargement / reduction filter processing unit 16 includes the image area separation filter 30, the smoothing filter 32, the block enlargement / reduction unit 40, and the enhancement filter 34 has been described. The present invention is not limited to this, and the enlargement / reduction filter processing unit 16 performs color conversion processing for performing color conversion processing from RGB data to CMYK data, N-value conversion processing, background detection processing, blank page detection processing, document color determination processing, and the like. For example, a configuration including other processing necessary for a copying machine, a multifunction machine, or the like may be used. In particular, the color conversion process to CMYK data and the N-value conversion process are indispensable processes for forming a dot pattern with a toner or the like on a recording medium such as paper in a copying machine or a multi-function machine. When the amount of pixel data before conversion is reduced by the conversion of (for example, converting 1-bit 8-bit data into 1-bit data), the amount of data to be finally written in the memory can be reduced. It is desirable to have a configuration that includes a digitization process.

  Further, in the above-described embodiment, the case where the present invention is applied to an image forming apparatus such as a copying machine or a multifunction machine has been described. The present invention is not limited to this, and the present invention may be applied to various devices such as a printer and a FAX machine.

  Further, in the above-described embodiment, the skew block reading unit 12 as the skew block reading unit, the block inclination correcting unit 14 as the block inclination correcting unit, and the enlargement / reduction filter processing unit 16 as the enlargement / reduction filter processing unit, The case where the image forming apparatus 10 as the image forming apparatus is configured by the block writing unit 18 as the block writing unit has been described. The present invention is not limited to this, and an image forming apparatus may be configured by a skew block reading unit, a block inclination correction unit, an enlargement / reduction filter processing unit, and a block writing unit having various other configurations.

  The present invention can also be used in various electronic devices that perform various processes relating to images, such as image scanners and facsimile machines.

  DESCRIPTION OF SYMBOLS 10 ... Image forming apparatus, 11 ... Control part, 12 ... Skew block reading part, 14 ... Block inclination correction part, 16 ... Filter processing part with expansion / contraction, 18 ... Block writing part, 20 ... Memory, 30... Image area separation filter, 32... Smoothing filter, 34... Emphasis filter, 40... Block enlargement / reduction section, 42 ... X-direction enlargement / reduction section, 44 ... Y-direction enlargement / reduction section, IMGi. Input image, IMGo ... Output image, BLr ... Read block, BLw ... Write block, BLi1 ... Input block, BLo1 ... Output block, BLi2 ... Input block, BLo2 ... Output block, BLi3 ... Enlargement / reduction unit Input block, AP: Document area, Ao: Output area, Aov: Output valid area, Bf: Buffer, LBf: Line buffer CAPs: Document start point coordinates, CBLs: Block start point coordinates, Cos: Output start point coordinates, Ccos: Output start point corresponding coordinates, Ci: Input corresponding coordinates, Co: Output corresponding coordinates, Pp: Current pixel, Pn ... Next pixel, Ppl ... Current line pixel, Pnl ... Next line pixel, CBLis ... Input block start point coordinates, CBLins ... Next input block start point coordinates, CBLincs ... Next stage input block start point coordinates, CBLos ... Output Block start point coordinates, CBLons ... Next output block start point coordinates, CBLoncs ... Next stage output block start point coordinates, Cis ... Enlargement / reduction part input start point coordinates, Cins ... Enlargement / reduction part next block input start point coordinates, Cincs ... Enlargement / reduction part next stage Block input start point coordinates.

Claims (5)

A skew block reading unit that sequentially reads out images in block units smaller than the size of one page while changing the start point coordinates in accordance with the inclination to be corrected from the memory in which the input image is stored;
A block inclination correction unit that corrects the inclination of each block-by-block image sequentially read out by the skew block reading unit, and outputs the image in which the inclination is corrected in block units;
An enlargement / reduction filter processing unit that performs processing including at least enlargement / reduction processing on an image in block units output by the block inclination correction unit;
An image forming apparatus comprising: a block writing unit that sequentially writes an image processed by the enlargement / reduction filter processing unit to a memory. The block inclination correcting unit includes a buffer having a predetermined size, and performs inclination correction using the buffer. At this time, when the correction inclination increases, the size of the processable image decreases, and the correction inclination decreases. Is configured to increase the size of the image that can be processed,
The image forming apparatus according to claim 1, wherein a size of an image of a block unit read by the skew block reading unit is a size that can be processed by the block inclination correcting unit. The enlargement / reduction filter processing unit includes a line buffer for processing an image having a width smaller than the width of one page,
3. The image forming apparatus according to claim 1, wherein a size of an image in a block unit read by the skew block reading unit is a size that can be processed by the block inclination correction unit and can be processed by the enlargement / reduction filter processing unit. . The skew block reading unit and the block inclination correcting unit have a counter for managing the coordinates of inclination correction and enlargement / reduction at an arbitrary magnification, and the starting point coordinates of each process are changed based on the counter. The image forming apparatus according to claim 1, wherein the processing is performed. The block inclination correction unit outputs the image whose inclination is corrected without writing to the memory in units of blocks,
The image forming apparatus according to claim 1, wherein the enlargement / reduction filter processing unit performs processing including enlargement / reduction processing at the arbitrary magnification on the output image in block units.

Images