JP2016096379A - Compression apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To effectively perform a processing for image data showing a reading image of an original.SOLUTION: A compression circuit compresses reading image data of an original provided by a reading device, and generates corresponded compressed image data. More specifically, the compressed image data that vertically reverses the reading image of original in a sub-scanning direction is generated (D). The compression circuit processes and compresses an image data group in a pixel block of the reading image data in each pixel block so as to be vertically reversed a position of the pixel in the pixel block. Thus, the compressed image data that has compressed block data showing a revered image in each pixel block is generated (C). Thereafter, the compression circuit replaces the compressed block data included in the compressed image data in the sub-scanning direction in a group unit of the pixel block along a main scanning direction, and generates the compressed image data that is vertically reversed the reading image in the sub-scanning direction (D).SELECTED DRAWING: Figure 3

Description

  The present invention relates to an apparatus having a function of compressing image data.

  2. Description of the Related Art Conventionally, an image reading system that reads a document and generates image data representing the read image is known. A compression device that compresses image data and generates compressed image data that is image data after compression is also known. As a compression method for image data, for example, the JPEG method is known. In the JPEG method, image data is compressed in units of blocks.

  As an expansion device that expands compressed image data, a device that expands compressed image data while executing a process of rotating an image represented by the compressed image data by 180 degrees is known (see, for example, Patent Document 1). In this decompression device, the compressed image data is read in order from the end for each block. Pixel data in each block is read sequentially from the end. By this reading, the image represented by the compressed image data is converted into an image rotated 180 degrees in the decompressed image data.

JP-A-9-2197882

  By the way, in the prior art in which image processing for reversing or rotating the read image is performed on the compressed image data representing the read image of the original after the entire original is read, the image processing is efficiently executed, It is difficult to generate or output compressed image data after image processing.

  Therefore, in one aspect of the present invention, it is desirable that image processing for image data representing a read image of a document can be efficiently executed.

  A compression device according to one aspect of the present invention includes a compression unit and a processing unit. The compression unit generates compressed image data corresponding to the image data by compressing the image data representing the read image of the document. The image data can be generated by executing a reading operation for reading the document for each line along the main scanning direction a plurality of times in the sub scanning direction.

  The compression unit compresses block data, which is data in the pixel block in the image data, for each pixel block by processing and compressing the position of each pixel in the pixel block in the sub-scanning direction. It can be configured to generate data. The pixel block can be defined as a region having a predetermined number of pixels in the main scanning direction and a predetermined number of lines in the sub-scanning direction. The compression unit may be configured to generate compressed image data having a plurality of compressed block data corresponding to the entire area of the read image as the compressed image data.

  The processing unit processes the compressed image data generated by the compression unit. The processing unit rearranges the compressed block data included in the compressed image data in units of bands, thereby converting the compressed image data so that the image represented by the compressed image data is converted into an image obtained by inverting the read image in the sub-scanning direction. It can be configured to work. The band unit mentioned here may be a set unit of compressed block data corresponding to a set of pixel blocks along the main scanning direction.

  According to one aspect of the present invention, compressed block data in which pixel positions are inverted in the sub-scanning direction is generated by processing and compressing the block data. Therefore, compressed image data obtained by inverting the read image can be generated more efficiently than when compressed block data is generated without inverting the pixel position. Therefore, according to one aspect of the present invention, it is possible to efficiently execute processing related to image inversion with respect to image data generated by a document reading operation.

  By the way, the compression unit may be configured to generate compressed image data in which a marker is attached to a set of compressed block data in band units as the compressed image data. In this case, the processing unit may be configured to identify a set of compressed block data in band units based on the markers in the compressed image data. The set of compressed block data can be efficiently or easily performed by using the marker.

  The compression unit may be configured to generate JPEG compressed image data as the compressed image data. In this case, the compression unit may be configured to generate JPEG compressed image data in which a restart marker is attached to a set of compressed block data in band units.

  The processing unit may be configured to specify a set of compressed block data in band units based on a restart marker in the compressed image data. The processing unit may be configured to process the compressed image data by rearranging the compressed block data in band units and re-adding the restart marker.

  The compression unit may be configured to start the compression before the entire image data is provided. Each of the line image data, which is data in line units from the first line to the last line constituting the image data, can be sequentially provided to the compression unit. In this case, the compression unit may be configured to start the compression before line image data up to the last line is provided.

  According to the compression apparatus having this configuration, the size of the data storage area (work area) necessary for generating the compressed image data can be suppressed. Therefore, it is possible to efficiently generate compressed image data using a storage device having a small storage capacity.

  When image data is compressed in units of pixel blocks, it may be necessary to add dummy data depending on the compression method. In other words, the addition of dummy data may necessitate that the number of lines of image data be an integer multiple of the number of lines of pixel blocks. The dummy data can be image data representing a blank image.

  In this case, if dummy data is added to the end line side of the image data, in the compressed image data processed by the processing unit, a blank image based on the dummy data appears on the top side of the entire image represented by the compressed image data. That is, the true read image excluding the blank image appears biased toward the end of the entire image. This bias can be frustrating to the user.

  Therefore, the compression apparatus may be provided with an additional unit that adds dummy data for a predetermined number of lines before the head line of the image data. In this case, the compression unit performs compression so that the position of each pixel in the pixel block is reversed in the sub-scanning direction for each pixel block in the image data with dummy, which is image data to which dummy data is added by the additional unit. It may be configured to generate block data and generate compressed image data.

  However, at the time when the dummy data is added, the number of lines of dummy data to be added before the first line of the image data due to the unknown number of lines from the first line to the last line of the image data. In some cases, it is not possible to specify an appropriate value. Therefore, the additional unit may be configured to generate a plurality of dummy image data having different numbers of lines of dummy data to be added before the first line as the dummy image data.

  In this case, the adding unit adds dummy data to at least one before the first line and after the last line of the image data, thereby adding dummy image data having a number of lines that is an integral multiple of the number of lines of the pixel block. Can be configured to generate.

  The compression unit generates, for each of the plurality of dummy image data, compressed block data processed so as to invert the position of each pixel in the pixel block for each pixel block in the dummy image data in the sub-scanning direction. In addition, by generating compressed image data having a plurality of compressed block data corresponding to the entire area of the dummy image data, a plurality of compressed image data can be generated.

  In this case, the processing unit selects one of the plurality of compressed image data as a processing target, and the compressed block data included in the selected compressed image data is divided into band units in the first line side and the last line side. It can be set as the structure rearranged so that it may invert. The processing unit can be configured to process the compressed image data so as to convert the image represented by the compressed image data into an image obtained by inverting the read image of the document in the sub-scanning direction by this rearrangement.

  Specifically, the processing unit may be configured to select compressed image data having a minimum number of lines of dummy data attached after the end line among a plurality of compressed image data as a processing target. By such selection, the compressed image data having the minimum margin represented by the dummy data on the top side of the inverted image represented by the compressed image data after processing can be selected as the processing target.

  The compression unit may be provided with image data for each side of the document. That is, the compression unit may be provided with image data representing a first one-side scanned image of both sides of the document and image data representing a second one-sided scanned image of both sides of the document. These image data can be generated by executing a reading operation for reading both sides of the document for each line a plurality of times.

  In this case, for the image data representing the read image on the first single side, the compression unit converts the compressed block data for each pixel block without performing the process of inverting the position of each pixel in the pixel block in the sub-scanning direction. It can be configured to generate and generate compressed image data having a plurality of compressed block data corresponding to the entire area of the read image.

  In other words, for the image data representing the read image on the second single side, the compression unit converts the compressed block data obtained by reversing the position of each pixel in the pixel block in the sub-scanning direction for each pixel block. It can be configured to generate and generate compressed image data having a plurality of compressed block data corresponding to the entire area of the read image.

  The processing unit corresponds to the second single side by rearranging the compressed block data included in the compressed image data corresponding to the second single side among the compressed image data corresponding to the first and second single side in band units. The compressed image data may be processed so as to convert the image represented by the compressed image data to be converted into an image obtained by inverting the read image on the second single side in the sub-scanning direction.

  When a double-sided printed original for long-side binding is read to generate compressed image data for generating a double-sided printed original for short-side binding, the read image on the back side of the original may need to be turned upside down. If the above-described compression device is used in such a document generation process, it is possible to efficiently generate compressed image data suitable for short-edge binding.

1 is a block diagram illustrating a configuration of an image reading apparatus. FIG. 6 is a diagram illustrating a difference between a long-side binding original and a short-side binding original. It is a figure explaining the basic production | generation procedure of the compressed image data showing the upside down image of the read image. It is the figure explaining the production | generation procedure of the compressed image data showing the upside down image of the read image in the form including the insertion procedure of dummy data. 3 is a block diagram illustrating a detailed configuration of a JPEG compression circuit that generates compressed image data corresponding to read image data on the back side of a document. FIG. It is a block diagram showing the detailed structure of a compression unit and compressed image data. FIG. 7A is a diagram showing the reading order of the pixel data group from the read image data by arrows, and FIG. 7B shows the input order to the DCT conversion unit of the pixel data group constituting the dummy image data by arrows. It is a figure. It is a flowchart showing the process which a compression unit performs. FIG. 9A is a flowchart showing processing executed by the processing unit, and FIG. 9B is a diagram showing the configuration of compressed image data to be processed in the upper stage and the configuration of inverted compressed image data in the lower stage. It is a time chart which shows operation | movement of a JPEG compression circuit. 11 is a time chart showing the operation of the JPEG compression circuit following FIG. 10.

  Embodiments of the present invention will be described below with reference to the drawings. However, the present invention is not construed as being limited in any way by the following examples. Embodiments of the present invention are all aspects that can be considered without departing from the essence of the invention specified only by the words of the claims.

  The image reading apparatus 1 of the present embodiment shown in FIG. 1 has a function of optically reading a document P, and a function of converting image data representing a read image of the document P into compressed image data and storing it. This image data is converted into JPEG compressed image data. The image reading apparatus 1 is configured as an auto document feeder type scanner apparatus, for example.

  The image reading apparatus 1 may be a copier or a facsimile machine. The image reading apparatus 1 can be configured to realize a copy function by printing a read image of a document P on paper using compressed image data. The image reading apparatus 1 can also be configured as a digital multi-function peripheral having two or more of a scanner function, a printer function, a copy function, and a facsimile function.

  The image reading apparatus 1 includes a CPU 11, a ROM 13, a RAM 15, a reading device 20, a reading control circuit 31, an A / D (analog / digital) converter 33, a reading correction circuit 35, and an image processing circuit 37. With.

  The CPU 11, the ROM 13, the RAM 15, the reading control circuit 31, the reading correction circuit 35, and the image processing circuit 37 are connected to each other via a bus. The image reading apparatus 1 may be configured with a user interface (not shown). In this case, the image reading apparatus 1 can accept an operation from the user, display various information for the user, and output a sound via the user interface.

  The CPU 11 performs overall control of the image reading apparatus 1 by executing processing according to a program recorded in the ROM 13 to realize various functions. The RAM 15 is used as a work area when the CPU 11 executes processing. The image reading apparatus 1 may include a memory (flash ROM, EEPROM, or the like) that can electrically rewrite data as the ROM 13 or separately from the ROM 13.

  Details of the reading device 20 are schematically shown in the rectangular area indicated by the dashed lines in FIG. As illustrated, the reading device 20 optically reads an image sensor 21A for optically reading the first surface (front surface) of the document P and the second surface (back surface) of the document P. Image sensor 21 </ b> B and a transport mechanism.

  The transport mechanism provided in the reading device 20 is configured to transport the document P in the sub-scanning direction as indicated by an arrow. The transport mechanism includes a first roller pair including a transport roller 25 and a driven roller 26 and a second roller pair including a paper discharge roller 27 and a driven roller 28 on the transport path 23 of the document P.

  The conveyance roller 25 and the paper discharge roller 27 are rotated by receiving power from a common electric motor (not shown) or individual electric motors. The document P supplied from the document tray between the transport roller 25 and the driven roller 26 is transported downstream in the sub-scanning direction by this rotation. The rotation of the transport roller 25 and the paper discharge roller 27 is controlled by the reading control circuit 31. Thereby, the document P is conveyed in the sub-scanning direction at a predetermined speed.

  However, the transport mechanism provided in the reading device 20 is not limited to the configuration shown in FIG. The transport mechanism may be any mechanism that can change the relative position between the image sensors 21A and 21B and the document P in the sub-scanning direction.

  Each of the image sensors 21A and 21B is configured as a line sensor that can read an image for one line along the main scanning direction orthogonal to the sub-scanning direction. These image sensors 21 </ b> A and 21 </ b> B are controlled by the reading control circuit 31 to perform a reading operation of reading the document P for each line along the main scanning direction.

  Each of the image sensors 21A and 21B generates image data representing a read image of a corresponding surface (a front surface or a back surface) of the document P by repeating the reading operation. Hereinafter, image data in one-sided units (in other words, in page units) representing the read images on each side constituting both sides of the document P will be expressed as read image data DI.

  More specifically, the image sensors 21A and 21B can be configured as line sensors having image sensors for each pixel arranged along the main scanning direction. In this case, each of the image sensors 21 </ b> A and 21 </ b> B has, as line image data representing a read image for one line of the original P, pixel data indicating a value corresponding to the charge accumulated in the image sensor due to the photoelectric effect, for one line. It can be configured to generate line image data for each pixel of minutes. As described above, the line image data has a group of pixel data.

  The reading control circuit 31 controls the reading device 20 according to a command from the CPU 11, thereby controlling the reading operation of the document P by the image sensors 21A and 21B and the transporting operation of the document P by the transport mechanism. With this control, the reading operation is executed every time the document P moves by a predetermined amount in the sub-scanning direction, and the read image data DI having line image data as a constituent element for each of the first surface and the second surface of the document P. Is generated.

  In the image sensors 21 </ b> A and 21 </ b> B, line image data generated for each line with respect to each of the first surface and the second surface of the document P is sequentially input to the A / D converter 33. The line image data input to the A / D converter 33 is converted from analog data to digital data and then input to the reading correction circuit 35.

  The reading correction circuit 35 performs various correction processes including shading correction on each of the line image data of the first surface and the second surface input from the A / D converter 33. The line image data of the first surface and the second surface corrected by the reading correction circuit 35 is stored in the RAM 15. In this manner, the RAM 15 stores the line image data group constituting the read image data DI, in other words, the pixel data group constituting the read image data DI, with respect to each of the first surface and the second surface. Are sequentially stored in accordance with the reading operation.

  The image processing circuit 37 is configured to execute compression processing on the pixel data group constituting the read image data DI, and execute various image processing as necessary. The image processing circuit 37 includes JPEG compression circuits 40A and 40B. The JPEG compression circuit 40A JPEG compresses the read image data DI corresponding to the first surface of the document P, and generates the compressed image data. The JPEG compression circuit 40B performs JPEG compression on the read image data DI corresponding to the second side of the document P, and generates the compressed image data. The generated compressed image data is stored in the RAM 15.

  Specifically, the JPEG compression circuits 40A and 40B sequentially read out the pixel data group constituting the read image data DI from the RAM 15, and compress them by the JPEG method in units of blocks determined by a predetermined rectangular area. As is well known, in the JPEG method, the read image data DI is compressed in units of MCU (Minimum Coded Unit). Each block is, for example, a rectangular area having a size of 16 pixels on each side. In other words, each block is a rectangular area having a size of 16 pixels in the main scanning direction and 16 lines in the sub-scanning direction.

  Hereinafter, each block is also expressed as a pixel block. In addition, the pixel data group included in the read image data DI is arranged on a two-dimensional coordinate system, and the pixel data group on the two-dimensional coordinate system is divided into pixel blocks, and the pixel data in each pixel block is determined. A set is also expressed as block data. The two-dimensional coordinate system is a coordinate system having a vertical axis along the sub-scanning direction and a horizontal axis along the main scanning direction. In this specification, according to the definition of the two-dimensional coordinate system, the sub-scanning direction of the read image data DI is regarded as the vertical direction, and the term “upper and lower” is used.

  That is, the JPEG compression circuits 40A and 40B compress the read image data DI for each block data, and generate the compressed image data including a group of compressed block data that is compressed block data. The compressed image data including the compressed block data corresponding to each pixel block is stored in the RAM 15.

  Here, the function of the JPEG compression circuit 40B that handles the read image data DI on the second side (back side) will be described in detail. The JPEG compression circuit 40B has a function of generating compressed image data by converting the read image of the document P represented by the read image data DI into an inverted image.

  This function is used, for example, to generate compressed image data for printing the original P1 for short side binding from the read image data DI of the original P0 for long side binding. This function can also be used when the document P is read from the lower end by the reading device 20 due to the opposite direction of the document P and the conveyance direction of the document P.

  The difference between the long side binding original P0 and the short side binding original P1 is as shown in FIG. The originals P0 and P1 shown in FIG. 2 are examples of the original P in which the letter “F” is written on the front side and the letter “B” is written on the back side. As shown in the left area of FIG. 2, in the long side-bound document P0, the characters “F” and “B” are placed either on the top or bottom and on the left or right side of the front side of the document P0 or the back side of the document P0. Also, it is arranged in a normal orientation that is not reversed. The hatched area in FIG. 2 represents a binding margin.

  On the other hand, in the original P1 for short edge binding, as shown in the right area of FIG. 2, the character “B” is rotated 180 degrees on the back side of the original P1 (in other words, the state is reversed up and down and left and right). Be placed. The right area of FIG. 2 shows the character arrangement of the front and back surfaces when the top and bottom surfaces of the original P1 are set to one top and bottom. This inversion is due to the fact that when the user turns the document P1 bound at the upper end of the front surface and looks at the back surface, the apparent lower end of the back surface corresponds to the upper end of the front surface.

  In this way, in order to print the original P1 for short side binding from the read image data DI of the long side binding original P0, at least an image that vertically inverts the read image with respect to the read image data DI on the back side. It is necessary to perform processing. For this reason, the JPEG compression circuit 40B is provided with a function for generating a compressed image data by inverting the read image of the document P upside down. The process of reversing the read image horizontally is realized, for example, by vertically reversing the read image and then expanding the compressed image data so that the left and right sides of the reversed image are switched.

  When the JPEG compression circuit 40B generates the compressed image data so as to convert the read image into an upside down image based on the read image data DI of the document P shown as an example in the area (A) of FIG. Compress the read image data DI. In the area (A) of FIG. 3, a simple geometric image is illustrated as an image represented by the document P using a white area and a hatched area.

  When the reading operation for the document P is started, the JPEG compression circuit 40B, when the pixel data group constituting the read image data DI is accumulated in the RAM 15 in an amount corresponding to one band, the pixel block for one band. Are converted into compressed block data. At this time, the JPEG compression circuit 40B processes the block data so as to invert the position of each pixel in the pixel block in the sub-scanning direction, and compresses the image represented by the pixel block upside down. Generate compressed block data to represent.

  One band corresponds to a set of pixel blocks for one row along the main scanning direction. When the vertical size of the pixel block is 16 pixels (lines), the pixel data group for one band corresponds to the pixel data group for continuous 16 lines in the pixel data group constituting the read image data DI. In the area (B) of FIG. 3, the boundary between bands is indicated by a broken line in the read image data DI of the original P having line image data of 16 lines × 4 = 64 lines.

  In the area (B) of FIG. 3, in the case where the reading operation of the document P is executed line by line by the reading device 20 from the base point of the arrow along the sub-scanning direction in the area (A) of FIG. The region of the pixel data group for the band is represented with reference numerals Z1, Z2, Z3, and Z4 in the order of compression processing. The numbers included in the symbols Z1, Z2, Z3, and Z4 represent the order of compression processing.

  The pixel data group for one band is converted into a compressed block data group for one band by compressing the pixel data in units of pixel blocks while being inverted upside down. In the area (C) of FIG. 3, an image represented by each band in a state where the read image is vertically inverted for each pixel block is shown.

  For the compressed image data having the compressed block data group processed in this way, the JPEG compression circuit 40B compresses the compressed block data in units of bands so as to reverse the band arrangement order as shown in the area (D) of FIG. Rearrange the groups. By this rearrangement, the JPEG compression circuit 40B generates compressed image data representing an image obtained by inverting the read image in the sub-scanning direction as compressed image data corresponding to the read image data DI.

  In other words, the JPEG compression circuit 40B converts the incomplete read image data DI into compressed block data in which the pixel position is vertically inverted in units of pixel blocks in a state where the reading of the entire document is not completed. After all the reading is completed, the compressed block data is rearranged in band units, thereby efficiently generating compressed image data obtained by vertically inverting the read image of the original P.

  The procedure for generating compressed image data in the example in which the number of lines G of the read image data DI is an integer multiple of the number of lines H of the pixel block has been described above with reference to FIG. However, the number of lines G of the read image data DI is not uniformly determined due to the influence of the paper size of the original P, the size of the margin area, and the like. For this reason, there may occur a case where the number of lines G of the read image data DI is not an integral multiple of the number of lines H of the pixel block.

  At this time, the JPEG compression circuit 40B adds dummy data representing a blank image to the read image data DI, and generates dummy image data with the number of lines adjusted. The dummy image data referred to here is data obtained by adding dummy data to the read image data DI. To the read image data DI, dummy data having a minimum number of lines in which the number Q of lines of image data with dummy is an integral multiple of the number of lines H of the pixel block (for example, 16 lines) is added.

  As shown in FIG. 4, when the JPEG compression circuit 40B generates inverted compressed image data DO, which is compressed image data obtained by inverting the read image, the number of dummy data added before the first line of the read image data DI is A plurality of different dummy image data D11, D12, D13, and D14 are generated. The lattice-shaped hatching area shown in FIG. 4 corresponds to the dummy data area. The dummy data is added as necessary after the last line of the read image data DI. If the sum of the number of lines L of dummy data attached before the first line and the total number of lines G from the first line to the last line is not an integral multiple of the number of lines H of the pixel block, the dummy data is It is added after the last line of the read image data DI so that the number Q of lines of the dummy image data is an integral multiple of the number of lines H of the pixel block.

  The dummy image data D11 is dummy image data in which the number of lines L of dummy data attached before the first line is zero. As an example of the dummy image data D11, the number G of lines of the read image data DI is an integral multiple of the number of lines H of the pixel block, so that no dummy data is added after the end line of the read image data DI. Is included. In the present specification, since the number of lines G of the read image data DI is an integer multiple of the number of lines H of the pixel block, the read image data to which dummy data is not added as a result is also formalized as dummy image data. Express.

  The JPEG compression circuit 40B generates compressed image data D21, D22, D23, D24 corresponding to the dummy image data D11, D12, D13, D14, respectively. Thereafter, the JPEG compression circuit 40B selects one of the compressed image data D21, D22, D23, and D24 corresponding to the dummy image data D11, D12, D13, and D14 as a processing target. Then, in the compressed image data to be processed, the compressed block data group is rearranged in units of bands so as to reverse the arrangement order, thereby converting the image represented by the read image data DI into an image obtained by inverting the read image in the sub-scanning direction. The converted inverted compressed image data DO is generated.

  Specifically, the JPEG compression circuit 40B selects the compressed image data D23 having the smallest number (including zero) of dummy data attached after the last line of the read image data DI as a processing target. Thereby, the JPEG compression circuit 40B generates the inverted compressed image data DO so that the inverted image of the read image does not appear biased downward in the image frame represented by the compressed image data.

  Next, the detailed configuration of the JPEG compression circuit 40B will be described. As shown in FIG. 5, the JPEG compression circuit 40B includes a plurality of compression units 41A, 41B, 41C, and 41D, a selection unit 45, and a processing unit 47. In FIG. 5, the RAM 15 is drawn in the block of the JPEG compression circuit 40B. This is because the compressed image data D21, D22, D23, and D24 output from the compression units 41A, 41B, 41C, and 41D are temporarily stored in the RAM 15. It means that it is memorized.

  Each of the compression units 41A, 41B, 41C, and 41D sequentially reads the pixel data group constituting the read image data DI stored in the RAM 15, and the number L of dummy data lines that precede the first line of the read image data DI is determined. Different dummy image data D11, D12, D13, and D14 are generated, and compressed image data D21, D22, D23, and D24 corresponding to the dummy image data D11, D12, D13, and D14 are output to the RAM 15.

  When the number of lines H of the pixel block is 16, and when it is necessary to generate inverted compressed image data DO obtained by inverting the read image, the compression units 41A, 41B, 41C, and 41D follow the operation parameters set by the CPU 11. It works as follows.

  The compression unit 41A adds dummy data as necessary after the end line of the read image data DI without attaching dummy data before the start line of the read image data DI, and is an integer of the number H of pixel block lines. Dummy image data D11 having twice the number of lines Q is generated.

  The compression unit 41A sequentially compresses the dummy image data D11 in units of pixel blocks while generating it, and outputs the compressed image data D21 to the RAM 15. Here, the compressed image data D21 output from the compression unit 41A is compressed image data in which the position of each pixel in the pixel block is inverted in the sub-scanning direction as illustrated in FIG.

  The compression unit 41B adds four lines of dummy data before the first line of the read image data DI, and adds dummy data as necessary after the last line of the read image data DI. Dummy image data D12 having a line number Q that is an integral multiple of H is generated. The compression unit 41B sequentially compresses the dummy image data D12 in units of pixel blocks while generating it, and outputs the compressed image data D22 to the RAM 15. Here, the compressed image data D22 output from the compression unit 41B is compressed image data in which the position of each pixel in the pixel block is inverted in the sub-scanning direction, as in the compression unit 41A.

  The compression unit 41C operates in the same manner as the compression unit 41B, except that the number L of dummy data lines attached before the first line of the read image data DI is eight. That is, the compression unit 41C adds dummy data for 8 lines before the first line of the read image data DI, and generates dummy image data D13 having a line number Q that is an integral multiple of the number L of pixel block lines. The corresponding compressed image data D23 is output to the RAM 15.

  The compression unit 41D operates in the same manner as the compression units 41B and 41C except that the number L of dummy data lines attached before the first line of the read image data DI is 12. That is, the compression unit 41D adds dummy data for 12 lines before the first line of the read image data DI, and generates dummy image data D14 having a line number Q that is an integral multiple of the number H of pixel block lines. The corresponding compressed image data D24 is output to the RAM 15.

  FIG. 6 shows a detailed configuration of the compression unit 41. The compression unit 41 corresponds to any one of the compression units 41A, 41B, 41C, and 41D. The compression units 41A, 41B, 41C, and 41D are configured to add dummy data for the number L of lines set by the CPU 11 before the first line of the read image data DI according to the operation parameters set by the CPU 11. And have a configuration common to each other.

  As shown in FIG. 6, the compression unit 41 includes a data input unit 51, a DCT conversion unit 53, a quantization unit 55, a zigzag sequence unit 56, an encoding unit 57, and a data output unit 59.

  The data input unit 51 sequentially reads out the pixel data group constituting the read image data DI from the RAM 15 and executes a process of adding dummy data and a process of inverting the position of each pixel in the pixel block in the sub-scanning direction. The pixel data group after the processing is sequentially input to the DCT conversion unit 53.

  That is, the data input unit 51 sequentially converts the pixel data group constituting the read image data DI into the pixel data group constituting the dummy image data D1, and inverts the position of each pixel in the pixel block in the sub-scanning direction. The converted pixel data group is input to the DCT conversion unit 53. The dummy image data D1 referred to here corresponds to one of the above-described dummy image data D11, D12, D13, and D14.

  A detailed configuration of the data input unit 51 is shown in a rectangular area indicated by a broken line in FIG. As illustrated, the data input unit 51 includes a main input unit 511, a dummy addition unit 513, a selector 515, and an input control unit 517.

  The main input unit 511 reads pixel data corresponding to the address designated by the input control unit 517 from the RAM 15 and inputs it to the selector 515 in accordance with a control signal from the input control unit 517. The dummy adding unit 513 inputs dummy data to the selector 515.

  The selector 515 selectively selects one of the pixel data input from the main input unit 511 and the dummy data input from the dummy addition unit 513 according to the selection signal from the input control unit 517 to the DCT conversion unit 53. input.

  The input control unit 517 inputs the selection signal to the selector 515 and inputs the control signal to the main input unit 511, whereby dummy data corresponding to the number L of lines set by the CPU 11 becomes the first line of the read image data DI. The selection operation of the selector 515 and the reading operation by the main input unit 511 are controlled so that the pixel position in the pixel block is added before and the pixel position in the pixel block is inverted in the sub-scanning direction.

  With this control, the selector 515 causes the pixel data group constituting the dummy image data D1 in which dummy data is added to the read image data DI, in a state where the position of each pixel in the pixel block is inverted in the sub-scanning direction. The data are sequentially input to the DCT conversion unit 53.

  The process of inverting the position of each pixel in the pixel block in the sub-scanning direction is controlled by the input control unit 517, and the main input unit 511 converts the pixel data group in the read image data DI according to the arrow shown in FIG. 7A. This is realized by reading. In FIG. 7A, the area in the read image data DI constituting one pixel block in the dummy image data D1 is represented by a broken line. An area where dummy data is inserted is represented by a grid-like hatching area. That is, in the process of reading out the pixel data group constituting one pixel block from the RAM 15, the pixel data group in the pixel block is read out from the reverse direction in the sub-scanning direction for each line. This is realized by inverting the group arrangement in the sub-scanning direction.

  Through such processing, the DCT conversion unit 53 has the pixel data group constituting the image data D1 with the dummy inverted in the sub-scanning direction in the sub-scanning direction along the arrow shown in FIG. 7B. In this state, the signals are sequentially input to the DCT conversion unit 53. The data in the lattice-shaped hatching area shown in FIG. 7B is dummy data.

  More specifically, the DCT conversion unit 53 includes corresponding pixel data groups in the order of the pixel block (1, 1), the pixel block (1, 2), and the pixel block (1, 3) in the main scanning direction. Is input to the DCT converter 53. Here, the pixel block (m, n) indicates a pixel block in the m-th row and the n-th column. The variables m and n are natural numbers of 1 or more. Then, when the pixel data group of the pixel block (1, N) in the last column in the same row is input to the DCT conversion unit 53, the pixel block (2, 1) in the first row in the next row is sequentially followed by the same row. The pixel data groups of the pixel block (2, n) are sequentially input to the DCT conversion unit 53.

  The DCT transform unit 53 performs the discrete cosine transform (DCT) on the pixel data group input from the data input unit 51 in this way in units of pixel blocks, and then inputs them to the quantization unit 55. That is, the DCT conversion unit 53 sequentially performs discrete cosine transform on block data for each pixel block constituting the image data with dummy, and the position of each pixel in the pixel block is inverted in the sub-scanning direction. And input to the quantization unit 55. The DCT transform unit 53 controls the input timing of the pixel data group from the data input unit 51 by inputting an update signal to the input control unit 517 every time the discrete cosine transform of each block data is completed.

  The quantization unit 55 quantizes each block data subjected to discrete cosine transform using a quantization table, and inputs the quantized block data, which is the quantized block data, to the zigzag sequence unit 56.

  For each quantized block data input from the quantizing unit 55, the zigzag sequence unit 56 converts this quantized block data into a one-dimensional data string and inputs the converted data to the encoding unit 57. The encoding unit 57 entropy encodes the input data from the zigzag sequence unit 56. The encoded data is input to the data output unit 59. The encoded data for each pixel block corresponds to the compressed block data described above.

  The data output unit 59 attaches attached data such as a header, an end marker (EOI), and restart markers (FFD0 to FFD7) to the encoded data input from the encoding unit 57, thereby compressing the JPEG method. The image data D2 is output to the RAM 15. The compressed image data D2 referred to here corresponds to one of the compressed image data D21, D22, D23, and D24 described above.

  As shown in FIG. 6, the compressed image data D2 has a header at its head as ancillary data and an end marker at its end. The compressed image data D2 is further configured to have a restart marker for a set of compressed block data in band units. The band data shown in FIG. 6 corresponds to a set of compressed block data for one band. The i-th band data corresponds to a set of compressed block data in the i-th row from the top of the dummy image data D1. The variable i is a natural number of 1 or more.

  As is well known, the restart marker is assigned an identification number in the range from 0 to 7. A restart marker having an identification number (i-1) is attached to the rear end of the i-th band data. When i> 8, a restart marker whose identification number is (i−1)% 8 is attached to the rear end of the i-th band data. The symbol% used here is a remainder operator, and (i-1)% 8 indicates the remainder when the value (i-1) is divided by 8.

  Here, the flow of processing realized by the compression unit 41 will be conceptually described with reference to FIG. When generating the compressed image data D2 corresponding to the read image data DI, the compression unit 41 outputs a header via the data output unit 59 (S110). Further, the pixel data group constituting the pixel block to be processed is read from the RAM 15 via the data input unit 51, dummy data is added as necessary, and the block data of the pixel block to be processed in the dummy image data D1 is used. Then, block data is generated by inverting the pixel position in the pixel block in the sub-scanning direction (S120). The block data is compressed (encoded) through the DCT conversion unit 53, the quantization unit 55, the zigzag sequence unit 56, and the encoding unit 57, and converted into compressed block data. This compressed block data is output as data following the header (S130).

  The compression unit 41 executes the processing according to S120 and S130 for each pixel block for one band. The compression unit 41 executes these processes for each of the pixel blocks for one band (Yes in S140), and if the execution for all the pixel blocks is incomplete (No in S150), 1 A restart marker is output following the compressed block data group for the band (S160).

  The compression unit 41 repeatedly executes the above-described processing (S120, S130, S140, S150, S160) until it is executed for all pixel blocks. When the execution for all the pixel blocks is completed (Yes in S150), an end marker is output following the compressed block data group of the last band (S170), and the compressed image data D2 corresponding to the read image data DI is stored in the RAM 15. To generate.

  By the operation of the compression units 41A, 41B, 41C, and 41D described above, the RAM 15 stores a plurality of compressed image data D21, D22, D23, and D24 with different numbers of lines L of dummy data.

  The selection unit 45 (see FIG. 5) selects one of the compressed image data D21, D22, D23, and D24 as a processing target, and inputs the selected compressed image data to be processed to the processing unit 47. Specifically, the selection unit 45 selects a processing target based on the total number of lines G of the read image data DI specified by the completion of the reading operation from the first line to the last line of the document P.

  More specifically, the selection unit 45 selects a processing target based on the remainder R = G% H when the total number of lines G is divided by the number of lines H of the pixel block = 16. Specifically, the selection unit 45 includes the line number L and the remainder R of the dummy data added before the first line of the read image data DI among the compressed image data D21, D22, D23, and D24 stored in the RAM 15. The compressed image data D2 having the maximum addition value L + R equal to or less than the number of lines H of the pixel block is selected as a processing target. However, when the remainder R = 0, the compressed image data D21 having the number of lines L = 0 is selected as a processing target. In this way, the selection unit 45 selects the compressed image data D2 having the smallest number of dummy data lines attached after the last line of the read image data DI as a processing target.

  Further, the processing unit 47 executes the processing shown in FIG. 9A based on the compressed image data D2 selected as the processing target, thereby obtaining inverted compressed image data DO representing an image obtained by inverting the read image in the sub-scanning direction. Generate.

  The processing unit 47 writes the copy data of the header of the compressed image data D2 to be processed on the RAM 15 as the header of the inverted compressed image data DO (S210). Thereafter, as shown in FIG. 9B, the compressed image data D2 to be processed is referred to in order from the end in the reverse direction (S220).

  If a restart marker is detected during this reference operation (Yes in S230), the processing unit 47 interrupts the reference operation. Then, the data located between the detected restart marker and the marker (end marker) that appears first in the forward direction (opposite to the reference direction) is a set of compressed block data for one band, that is, Assuming that the data is band data, the copy data of the band data is written in succession to the header of the inverted compressed image data DO (S250). Further, a restart marker (FFD0) is added (S260), and the process proceeds to S220.

  In S220, the reference operation is resumed from the interruption point. When the restart marker is detected (Yes in S230), the processing unit 47 converts the data located between the restart marker and the first marker (restart marker) appearing in the positive direction as band data. Accordingly, the copy data of the band data is added to the inverted compressed image data DO being generated (S250). Further, a restart marker (FFD1) is written (S260), and the process proceeds to S220.

  In this way, the processing unit 47 refers to the inside of the compressed image data D2 to be processed in the reverse direction, and each time the restart marker is detected, the copy data of the band data specified based on the restart marker is obtained. Then, it is added to the reverse compression image data DO being generated (S250). Furthermore, the renumbered restart marker is written (S260).

  After that, when the processing unit 47 detects the header instead of the restart marker (Yes in S240), the copy data of the band data located between the header and the restart marker that appears first in the forward direction is being generated. Is added to the inverted compressed image data DO (S270). Further, an end marker is added (S280), and the inverted compressed image data DO is completed on the RAM 15.

  By executing the processing shown in FIG. 9A in this way, the processing unit 47 rearranges the compressed block data group included in the compressed image data D2 to be processed in units of bands, and further attaches a restart marker. Thus, the inverted compressed image data DO configured as shown in the lower part of FIG. 9B is generated.

  As described above, according to the image reading apparatus 1 of the present embodiment, the reading device 20 repeatedly performs a reading operation of reading both surfaces of the document P for each line along the main scanning direction in the sub scanning direction. The read image data DI is generated for each side of the document P. Specifically, the reading device 20 generates line image data including a pixel data group for one line for each reading operation, and sequentially writes the line image data in the RAM 15.

  The JPEG compression circuits 40A and 40B each store a pixel data group for one band each time a pixel data group for one band is accumulated in the RAM 15 before the line image data from the reading device 20 to the last line is provided to the RAM 15. The compression process for each is started.

  Further characteristically, when generating the compressed block data corresponding to the pixel block, the compression unit 41 reads the pixel data group in the pixel block stored in the RAM 15 in the reverse direction in the sub-scanning direction. Thereby, the compression unit 41 processes and compresses the pixel data group in the pixel block so as to invert the position of each pixel in the pixel block in the sub-scanning direction, and the read image in the pixel block is inverted upside down. Generate compressed block data. In this way, the compression unit 41 generates compressed image data D2 having compressed block data for each pixel block in which the read image in the pixel block is inverted upside down.

  The processing unit 47 converts the image represented by the compressed image data D2 into a vertically inverted image of the read image by rearranging the compressed block data included in the compressed image data D2 generated by the compression unit 41 in units of bands. The compressed image data D2 is processed as described above.

  As described above, according to this embodiment, each time a pixel data group for one band is accumulated in the RAM 15 before the reading operation up to the final line (end line) is completed, the pixel data group is stored in units of pixel blocks. Process and compress. For this reason, it is possible to generate the inverted compressed image data DO efficiently.

  For example, when the inverted compressed image data DO is generated after the read image data DI is completed, the read image data DI having a large data size that is not compressed needs to be stored in the RAM 15. On the other hand, according to the present embodiment, the RAM 15 can be used efficiently, and the size of the data storage area necessary for generating the inverted compressed image data DO can be suppressed.

  Further, according to the present embodiment, there is an advantage that the inverted compressed image data DO can be completed more quickly than the case where the compression is started after the read image data DI is completed. Therefore, according to the present embodiment, the inverted compressed image data DO can be quickly provided to the user, and the printed matter based on the inverted compressed image data DO can be quickly provided to the user. That is, according to the present embodiment, it is possible to efficiently generate and output the inverted compressed image data DO suitable for short-edge binding.

  In addition, according to the present embodiment, in order to rearrange the set of compressed block data in band units, restart markers (FFD0 to FFD7) are attached to each set of compressed block data (band data) in band units. Then, the processing unit 47 identifies a set of compressed block data in band units based on the restart marker, rearranges the compressed block data in band units, and re-adds the restart marker. The compressed image data D2 is processed. Therefore, according to the present embodiment, conversion from the compressed image data D2 to the inverted compressed image data DO can be performed appropriately and efficiently.

  In the JPEG system, the number of lines of the read image data DI needs to be an integral multiple of the pixel block in order to compress the read image data DI in units of pixel blocks. Conventionally, such adjustment of the number of lines has been performed by adding dummy data representing a blank image after the last line of the read image data DI.

  However, if dummy data is added only after the end line of the read image data DI, when image printing is performed on a sheet based on the inverted compressed image data DO, a blank image based on the dummy data is displayed on the leading side of the sheet. Appear, and the true read image appears biased toward the lower end of the paper. This bias can be frustrating to the user.

  Therefore, in this embodiment, dummy data is added before the first line of the read image data DI. Therefore, according to the present embodiment, when the reverse image of the read image is printed on the paper based on the reverse compression image data DO, it is possible to suppress the reverse image from being printed biased toward the trailing edge of the paper. it can.

  In particular, in this embodiment, the JPEG compression circuit 40B is provided with a plurality of compression units 41A, 41B, 41C, and 41D. Then, in each of the compression units 41A, 41B, 41C, and 41D, dummy data is added to at least one of the read image data DI before the first line and after the last line, so that the number of lines H of the pixel block is increased. Dummy image data D11, D12, D13, and D14 having a line number Q that is an integral multiple is generated. Between the compression units 41A, 41B, 41C, and 41D, the number L of dummy data added before the first line is set. I tried to change each other.

  Then, among the compressed image data D21, D22, D23, and D24 corresponding to the dummy image data D11, D12, D13, and D14, compressed image data having the smallest number of dummy data lines on the end line side is processed. The inverted compressed image data DO is generated by selection. Therefore, according to the present embodiment, even when the number G of lines of the read image data DI cannot be specified in advance, the reverse compressed image data DO that can suppress the reverse image from being printed biased toward the rear end of the paper. Can be generated.

  Here, a supplementary description will be given of the image reading apparatus 1 described above. The JPEG compression circuit 40B may be configured to convert the read image data DI of the second side (back side) of the document P into compressed image data without executing the process of inverting the read image upside down. The case where the inverted compressed image data DO needs to be generated is limited to some cases such as the case of printing the original P1 for short side binding. Accordingly, the JPEG compression circuit 40B is in either an operation mode for generating the inverted compressed image data DO or an operation mode for converting the read image data DI into the compressed image data without inverting the read image in accordance with an instruction from the CPU 11. It may be configured to operate depending on the situation.

  The JPEG compression circuit 40A may be configured similarly to a known JPEG compression circuit. In other words, the JPEG compression circuit 40A represents the read image of the document P by compressing the read image data DI in units of pixel blocks without performing the process of inverting the position of each pixel in the pixel block in the sub-scanning direction. What is necessary is just to be comprised so that the compressed image data provided with the compressed block data for every pixel block may be produced | generated on RAM15 as compressed image data.

  If the number of lines G of the read image data DI is not an integral multiple of the number of lines H of the pixel block, dummy data is attached to the end line of the read image data DI, and the dummy image data with the adjusted number of lines is compressed. As described above, the JPEG compression circuit 40A may be configured. That is, the JPEG compression circuit 40A is a compression unit having the same hardware configuration as the compression unit 41, and can be configured as a compression unit that reads a pixel data group from the RAM 15 in the forward direction.

  As described above, since the function as the JPEG compression circuit 40A can be realized by using a partial configuration of the JPEG compression circuit 40B, the JPEG compression circuits 40A and 40B may be combined into one. In addition, the compression units 41A, 41B, 41C, and 41D may be configured as one compression unit in which overlapping circuits are shared. Further, the functions as the selection unit 45 and the processing unit 47 may be realized by software. The functions as the selection unit 45 and the processing unit 47 can be realized by the CPU 11 executing the program instead of the JPEG compression circuit 40B.

  In addition, the image reading apparatus 1 may be configured to operate according to the time charts shown in FIGS. 10 and 11. According to FIGS. 10 and 11, the compression unit 41D as the fourth compression unit adds 12 lines of dummy data before the first line of the read image data DI, so that the pixels constituting the first band data As a data group, when a pixel data group for four lines from the first line constituting the read image data DI is accumulated in the RAM 15, compression processing for one band is executed. In this compression processing, compressed block data for each pixel block constituting the first band data is generated and written to the RAM 15.

  Thereafter, each time the pixel data group from the (16 × J + 5) th line to the (16 × J + 20) th line constituting the read image data DI is accumulated in the RAM 15 (J is an integer of 0 or more). The compressed image data D24 is generated by generating compressed block data for each pixel block constituting the (J + 2) th band.

  On the other hand, since the compression unit 41C as the third compression unit adds dummy data for eight lines before the first line of the read image data DI, the read image data is used as a pixel data group constituting the first band data. When the pixel data group for 8 lines from the first line constituting the DI is accumulated in the RAM 15, compression processing for 1 band is executed. Thereafter, the compression unit 41C stores the (J + 2) -th band data every time the pixel data group from the (16 × J + 9) -th line to the (16 × J + 24-th) line constituting the read image data DI is accumulated in the RAM 15. The compressed image data D23 is generated by generating compressed block data for each pixel block constituting the block.

  Since the compression unit 41B as the second compression unit adds dummy data for four lines before the first line of the read image data DI, the read image data DI is used as a pixel data group constituting the first band data. When the pixel data group for 12 lines from the first line constituting the data is accumulated in the RAM 15, compression processing for one band is executed. Thereafter, each time the pixel data group from the (16 × J + 13) th line to the (16 × J + 28) th line constituting the read image data DI is accumulated in the RAM 15, the compression unit 41B stores the (J + 2) th band data. The compressed image data D22 is generated by generating compressed block data for each pixel block constituting the block.

  Since the compression unit 41A as the first compression unit does not add dummy data before the first line of the read image data DI, the first line constituting the read image data DI as a pixel data group constituting the first band data. When the pixel data group for 16 lines is accumulated in the RAM 15, the compression processing for one band is executed. Thereafter, the compression unit 41A stores the (J + 2) -th band data every time the pixel data group from the (16 × J + 17) th line to the (16 × J + 32) -th line constituting the read image data DI is accumulated in the RAM 15. The compressed image data D21 is generated by generating compressed block data for each pixel block constituting the block.

  Then, when the compression unit 41A completes the compression block data conversion for the pixel data group from the first line to the sixteenth line, the JPEG compression circuit 40B completes the unnecessary first line (first line) to the fourth line. Processing for releasing the area in the RAM 15 in which the pixel data group is written into an overwritable area is executed. Thereafter, the JPEG compression circuit 40B sequentially releases the area in the RAM 15 in which the pixel data group that has been converted into the compressed block data in all of the compression units 41A, 41B, 41C, and 41D is written into an overwritable area. Execute the process for

  Specifically, the JPEG compression circuit 40B releases the above by inputting a notification signal for notifying the overwritable area to the reading correction circuit 35 that writes the pixel data group provided from the reading device 20 to the RAM 15. Can be executed.

  According to the image reading apparatus 1 configured as described above, the read image data DI can be efficiently converted into the inverted compressed image data DO using the memory (RAM 15) having a small storage capacity, which is very significant.

  Finally, the correspondence between terms will be described. The compression units 41A, 41B, 41C, and 41D included in the JPEG compression circuit 40A and the JPEG compression circuit 40B correspond to an example of the compression unit included in the compression apparatus of the present invention, and the selection unit 45 and the processing unit 47 correspond to the compression unit of the present invention. This corresponds to an example of a processing unit provided in the apparatus. The function realized by the dummy adding unit 513, the selector 515, and the input control unit 517 corresponds to an example of the function realized by the additional unit.

DESCRIPTION OF SYMBOLS 1 ... Image reading apparatus, 11 ... CPU, 13 ... ROM, 15 ... RAM, 20 ... Reading device, 21A, 21B ... Image sensor, 31 ... Reading control circuit, 33 ... A / D converter, 35 ... Reading correction circuit, 37 ... Image processing circuit, 40A, 40B ... JPEG compression circuit, 41, 41A, 41B, 41C, 41D ... Compression unit, 45 ... Selection unit, 47 ... Processing unit, 51 ... Data input unit, 53 ... DCT conversion unit, 55 Quantizer, 56 ... Zigzag sequence unit, 57 ... Coding unit, 59 ... Data output unit, 511 ... Main input unit, 513 ... Dummy addition unit, 515 ... Selector, 517 ... Input control unit, D1, D11, D12 , D13, D14 ... Image data with dummy, D2, D21, D22, D23, D24 ... Compressed image data, DI ... Read image data, DO ... Inverted compressed image Data, P, P0, P1 ... manuscript.

Claims (8)

Compression corresponding to the image data by compressing image data representing a read image of the document generated by executing a reading operation for reading the document line by line in the main scanning direction a plurality of times in the sub-scanning direction. A compression unit for generating image data;
A processing unit for processing the compressed image data generated by the compression unit;
With
For each pixel block defined as an area having a predetermined number of pixels in the main scanning direction and a predetermined number of lines in the sub-scanning direction, the compression unit converts block data that is data in the pixel block in the image data into the pixels A compressed block data is generated by processing and compressing the position of each pixel in the block so as to be inverted in the sub-scanning direction, and a plurality of compressed blocks corresponding to the entire area of the read image is generated as the compressed image data. Generating compressed image data having data;
The processing unit rearranges the compressed block data included in the compressed image data in a band unit that is a set of the compressed block data corresponding to the set of the pixel blocks along the main scanning direction. A compression apparatus that processes the compressed image data so as to convert an image represented by data into an image obtained by inverting the read image in the sub-scanning direction. The compression unit generates, as the compressed image data, compressed image data with a marker attached to the set of compressed block data in units of bands,
The compression device according to claim 1, wherein the processing unit specifies the set of the compressed block data in the band unit based on the marker in the compressed image data. The compression unit generates, as the compressed image data, JPEG compressed image data with a restart marker attached to the set of compressed block data in the band unit,
The processing unit identifies the set of compressed block data in band units based on the restart marker in the compressed image data,
The compression apparatus according to claim 1, wherein the processing unit processes the compressed image data by rearranging the compressed block data in units of bands and further adding the restart marker. Each of the line image data that is data in line units from the first line to the last line constituting the image data is sequentially provided to the compression unit,
The compression apparatus according to any one of claims 1 to 3, wherein the compression unit starts the compression before the line image data up to the end line is provided. An additional unit for adding dummy data for a predetermined number of lines before the head line of the image data,
The compression unit reverses the position of each pixel in the pixel block in the sub-scanning direction for each pixel block in the image data with dummy, which is image data to which the dummy data is added by the additional unit. The compression apparatus according to claim 4, wherein the compressed image data is generated by generating the processed compressed block data. Addition of generating dummy image data having a number of lines that is an integral multiple of the number of lines of the pixel block by adding dummy data to at least one of the image data before the first line and after the last line With a unit
The additional unit generates, as the dummy image data, a plurality of dummy image data having different numbers of lines of the dummy data attached before the first line,
The compression unit is configured to process each of the plurality of dummy image data so that the position of each pixel in the pixel block is inverted in the sub-scanning direction for each pixel block in the dummy image data. Generating block data, generating a plurality of compressed image data by generating a compressed image data having a plurality of compressed block data corresponding to the entire area of the dummy image data,
The processing unit selects one of the plurality of compressed image data as a processing target, and the compressed block data included in the selected compressed image data includes a first line side and a last line side in the band unit. The compressed image data is processed so that an image represented by the compressed image data is converted into an image obtained by inverting the read image in the sub-scanning direction by rearranging the images so as to be reversed. Item 5. The compression device according to Item 4. The processing unit selects, as the processing target, compressed image data having a minimum number of lines of the dummy data attached after the end line among the plurality of compressed image data. The compression device described. The compression unit is provided with image data representing the read image on one side of each side of the document generated by executing a reading operation for reading both sides of the document for each line a plurality of times.
The compression unit is
For the image data representing the read image of the first single side of the two sides, the compression for each pixel block is performed without performing a process of inverting the position of each pixel in the pixel block in the sub-scanning direction. Generate block data, generate compressed image data having a plurality of compressed block data corresponding to the entire area of the read image,
For the image data representing the read image of the second single side of the two sides, the compressed block data obtained by processing the pixel block in the pixel block to be inverted in the sub-scanning direction for each pixel block. To generate compressed image data having a plurality of compressed block data corresponding to the entire area of the read image,
The processing unit is
By rearranging the compressed block data included in the compressed image data corresponding to the second single side among the compressed image data corresponding to the first and second single sides, the second unit The compressed image data is processed so that an image represented by the compressed image data corresponding to one side of the image is converted into an image obtained by inverting the read image on the second single side in the sub-scanning direction. The compression apparatus as described in any one of Claims 1-7.

Images