JP6260372B2 - Image processing apparatus and image data processing method - Google Patents

Description

  The present invention relates to image data processing in which image data is compressed and stored in a nonvolatile memory.

  The copying machine reads an image from a document sheet by a scanner and prints it on a new sheet by a printer.

  After the copy command, the scanner can start reading immediately, but the printer has to raise the temperature of the fixing device, and it may take time to start printing. In general, the speed at which the scanner reads an image from one sheet of paper is faster than the speed at which the printer prints an image on one sheet of paper.

  Therefore, conventionally, a copying machine copies an image as follows. When an image is read by the scanner, the image data of the image is compressed and stored in a volatile memory, and the compressed image data is saved from the memory to a nonvolatile storage. Then, in accordance with the progress of processing in the printer, the compressed image data is read from the storage to the memory, decompressed to restore the image data, and the printer prints the image on paper.

  The reason why the image data is compressed and saved in the non-volatile storage is that the memory is expensive, so that a memory with a small capacity can be used to reduce the manufacturing cost.

Japanese Patent Laid-Open No. 2006-20029 JP 2003-198817 A

  The following methods have been proposed as a method for managing the memory of the copying machine. One page of image data is sequentially processed as band data of less than one page as follows. When the image output means is an output means that cannot be stopped after startup, the memory management means secures only a memory capacity of a size that can accommodate all the data for the page to be output and stores the bitmap memory in the page If only a memory size of less than a band unit is secured and the image output means is not an output means that cannot be stopped after startup, both the second code memory and the bitmap memory are secured in band units (Patent Document) 1).

  Alternatively, after performing data compression conversion on the input image by DMA control of the memory control unit, a plurality of data transfer (simultaneous transfer) is performed when performing a series of processes for transferring the converted data to the HDD through the buffer area of the image memory ) Depending on whether there is a request, each transfer operation of division (multiple times) / batch (once) is selected. A capacity corresponding to the divided transfer amount to the HDD is secured as a buffer area when one image is divided and transferred, and a capacity corresponding to the batch transfer amount is ensured when transferring one image (Patent Document 2).

  According to the above method, a memory for storing compressed image data can be used efficiently. However, there is a need for a more efficient method of using this memory.

  In view of such a problem, an object of the present invention is to more efficiently use a memory for storing compressed image data.

  An image processing apparatus according to an aspect of the present invention is an image processing apparatus that outputs any one of images input by each of one or a plurality of input units, and outputs the image for each of the images. A mode setting unit that sets a compression mode to one of the first mode and the second mode, and uncompressed image data that is uncompressed image data of the image are displayed in the mode of the image. When the first mode is set, compression is performed every first unit, and when the second mode is set, compression is performed every second unit smaller than the first unit. To store all the compressed data for each of the images for which the first mode is set in the compression means for generating compressed data, a volatile memory, and the memory. A first area is secured, a second area for storing the compressed data for each of the images for which the second mode is set is secured, and the current processing target by the output means Area securing means for securing a third area for storing all the compressed data of a certain image, a non-volatile storage, and all the compressed data of the image for which the first mode is set Are transferred to the storage and stored in association with the image, and the compressed data of the image for which the second mode is set is stored in the memory. A first transfer means for transferring the compressed data to the storage and storing it in the storage in association with the image; A second transfer means for transferring the compressed data of the image is the force of the object of processing to all the memory.

  According to the present invention, a memory for storing compressed image data can be used more efficiently.

FIG. 2 is a diagram illustrating an outline of a hardware configuration of an MFP according to an embodiment of the present invention. 2 is a diagram illustrating a configuration of an image processing unit of an MFP. FIG. It is a figure which shows the flow of the data in an image process part. It is a figure which shows the 1st memory usage method and 2nd memory usage method in connection with an image process part. It is a figure which shows the function structure of the controller of an image process part. It is a flowchart of the setting process of a memory usage method. It is a flowchart of the calculation routine of the required memory capacity of FIG. It is a flowchart of the calculation routine of the free memory capacity of FIG. FIG. 7 is a flowchart of the memory allocation routine of FIG. 10 is a flowchart of a band size calculation subroutine of FIG. 9. It is a figure which shows the 1st example of memory allocation. It is a figure which shows the 2nd example of memory allocation. It is a figure which shows the 3rd example of memory allocation. It is a figure which shows the 4th example of memory allocation. It is a flowchart which shows the other example of a setting process of a memory usage method. 15 is a flowchart of a required memory capacity calculation routine of FIG.

  An MFP (Multi-functional Peripheral) is cited as an image processing apparatus according to an embodiment of the present invention. An MFP is a composite information device that integrates a plurality of functions useful for office work, and operates as a copier, a printer, a network scanner, a facsimile machine, a document server, or the like according to a job to be input.

  As shown in FIG. 1, the MFP 1 includes a main controller 10, an image processing unit 13, a storage 14, a communication interface 15, an operation panel 16, two image scanners 17 </ b> A and 17 </ b> B, a printer engine 19, and a power supply circuit 20.

  The main controller 10 has a CPU (Central Processing Unit) that executes a program for controlling the MFP 1, a ROM (Read Only Memory) that stores the program, and a RAM (Random Access Memory) that is used as a work area for executing the program. doing.

  The image processing unit 13 performs a process of compressing image data (scan data) input from each of the image scanners 17A and 17B, and a process of expanding the compressed image data. In the image processing unit 13, as described later, the first memory usage method and the second memory usage method are set according to the state of input / output of image data, as will be described later. .

  The storage 14 is, for example, a hard disk drive having a storage capacity of several hundred gigabytes or more. The storage 14 is used for storing application programs, setting data, and temporarily storing image data compressed by the image processing unit 13. The storage 14 is provided with a group of boxes for storing documents. The box group includes a plurality of boxes such as a personal box assigned to each user and a shared box shared by a plurality of users.

  The communication interface 15 enables communication between the MFP 1 and an external device. The communication interface 15 includes a network interface card (NIC) that connects the MFP 1 to a LAN (Local Area Network) so as to be communicable, and a modem for facsimile communication via a public telephone line.

  The operation panel 16 has a touch panel display and a hard key panel. The touch panel display is an input / output device that displays an operation screen on which buttons as software keys are arranged and detects a touch operation on the screen. The touch panel display includes a liquid crystal panel and a touch panel having a translucent touch surface portion that is in close contact with the surface thereof. The hard key panel has a plurality of hard keys including a start key for instructing the start of processing.

  Each of the two image scanners 17A and 17B optically reads an image recorded on the original sheet placed on the platen glass by line sequential scanning. In this embodiment, these image scanners 17A and 17B are of the same type, can read color images and monochrome images, have a maximum reading size of A3 size, and both have an ADF (Auto Document Feeder). And The image scanners 17A and 17B have sensors that output a sheet detection signal for detecting the size of the original sheet by a known method.

  Hereinafter, the image scanners 17A and 17B may be collectively referred to as an image scanner 17.

  The printer engine 19 prints an image on paper by, for example, electrophotography in copying, printing, and facsimile reception. The printing method may be an inkjet method or other methods.

  The power supply circuit 20 converts the power supplied from the commercial AC power source into power of a voltage suitable for the drive system and the control system and supplies the converted power.

  Such MFP 1 executes a job designated by the user using the operation panel 16 and a job designated by access from an external device. The various jobs that can be executed by the MFP 1 include a job that involves reading an image from an original sheet (referred to as an “image reading job”). In one image reading job, one of the image scanners 17A and 17B is used.

  Image reading jobs include copy, scan to box, and data transmission. Copy is a job for printing a read original image on paper. The scan-to-box is a job for storing image data obtained by reading as saved data in a box of the storage 14. Data transmission is a job for generating image data for transmission based on image data obtained by reading, and sending the generated image data to a designated destination. The destination specified for data transmission is a user terminal represented by a personal computer, a file server, an e-mail server, a facsimile machine, or the like. When a server is provided in the MFP 1, the server can be the other party.

  The MFP 1 can read the image by the image scanner 17A and the image by the image scanner 17B in parallel. In the MFP 1, when one of the image scanners 17A and 17B is used by an image reading job (J) submitted by the user, the user or the other user uses the other of the image scanners 17A and 17B. The execution start of K) can be instructed.

  FIG. 2 shows the configuration of the image processing unit 13 related to the image reading job. As illustrated, the image processing unit 13 includes an ASIC (Application Specific Integrated Circuit) 30 and a code memory 35. The ASIC 30 includes a bitmap memory 31, a compression / decompression circuit 32, a DMA transfer circuit 33, and a controller 34.

  The bitmap memory 31 is used for temporary storage of raster format image data. The bitmap memory 31 has a memory capacity that makes it possible to print an image while reading a plurality of documents in parallel using both of the two image scanners 17A and 17B.

  The compression / decompression circuit 32 is responsible for compressing image data and decompressing the compressed image data. The compression / decompression circuit 32 is configured to be able to compress the image data obtained by the image scanners 17A and 17B in parallel. The configuration for compressing in parallel may include a plurality of compressors that can perform compression simultaneously, or may compress a plurality of image data in a time division manner.

  The DMA transfer circuit 33 is a circuit for performing data transfer inside the image processing unit 13 and data input / output between the image processing unit 13 and the outside by DMA (Direct Memory Access).

  The controller 34 manages the memory areas of the bitmap memory 31 and the code memory 35. The controller 34 designates a memory area to be accessed for storing or reading data in each of the bitmap memory 31 and the code memory 35 to the DMA transfer circuit 33. The controller 34 includes a processor 340 that executes a control program for realizing such processing. A control program executed by the processor 340 is stored in the ROM in the ASIC 30.

  FIG. 3 shows the flow of data in the image processing unit. In the image processing unit 13, data is transferred as in the following [1] to [8].

 [1] When the execution of the image reading job is started, image data D1 obtained by scanning the original sheet from one of the used image scanners 17 (for example, the image scanner 17A) is sequentially input to the image processing unit 13. The The input image data D1 is DMA-transferred and temporarily stored in the bitmap memory 31.

 [2] When image data D1 having a data amount for one page of the original or smaller is stored in the bitmap memory 31, the image data D1 is DMA-transferred from the bitmap memory 31 to the compression / decompression circuit 32. At this time, when the input of the image data D1 from the image scanner 17A is continued, the image data newly input from the image scanner 17A in parallel with the reading of the stored image data D1 from the bitmap memory 31. D1 is stored in the bitmap memory 31.

 [3] Image data D2 obtained by compressing the image data D1 is DMA-transferred from the compression / decompression circuit 32 to the code memory 35.

 [4] The image data D 2 is DMA-transferred from the code memory 35 to the storage 14.

  The above transfers [1] to [4] are performed in parallel. That is, while storing the image data D1 in the bitmap memory 31, the previously stored image data D1 is read and compressed, and the obtained image data D2 is temporarily stored in the code memory 35, and the image data D2 is appropriately stored. Is stored in the storage 14.

  Note that by compressing the image data D1 prior to transfer to the storage 14, the amount of data to be transferred can be reduced and the time required for transfer can be shortened.

  When the image reading job is a scan-to-box, the image data D2 corresponding to the image data D1 obtained from one or more original sheets is stored in the storage 14, and the job is completed.

  On the other hand, when the image reading job is copying or data transmission, the following transfers [5] to [8] are performed in parallel with the transfers [1] to [4]. FIG. 3 illustrates the case of copying.

 [5] One page of image data D2 is DMA-transferred from the storage 14 to the code memory 35.

 [6] One page of image data D2 is DMA-transferred from the code memory 35 to the compression / decompression circuit 32. The image data D2 is decompressed by the compression / decompression circuit 32.

 [7] Image data D3 obtained by expanding the image data D2 is DMA-transferred from the compression / expansion circuit 32 to the bitmap memory 31.

 [8] The image data D3 is DMA-transferred from the bitmap memory 31 to the printer engine 19. Thereafter, raster image data for printing is generated in the printer engine 19 based on the image data D3.

  When the transfers [1] to [4] and the transfers [5] to [8] are performed in parallel, the code memory 35 is used for input to store the image data D2 from the compression / decompression circuit 32. A memory area and an output memory area for storing image data D2 for one page read from the storage 14 are allocated.

  Regarding the use of the bit map memory 31 and the code memory 35 in the image processing unit 13 to which the transfer of [1] to [8] is performed, there are a first memory use method and a second memory use method. In the image processing unit 13, these two memory usage methods are dynamically used in accordance with the increase or decrease in the free memory capacity of the code memory 35 after the execution of the image reading job is started.

  FIG. 4A shows a first memory usage method, and FIG. 4B shows a second memory usage method. 4A and 4B, it is assumed that the image scanner 17A reads an image of a document 40 (which is referred to as a document) composed of a plurality of document sheets. However, the document 40 may be a single document sheet, or the image scanner 17B may read an image of the document 40. In the figure, diagonal lines attached to the memory areas 60 and 70 indicate that the memory capacity (memory size) is equivalent to one page. The memory capacity of the memory areas 50 and 80 not hatched is less than one page.

  In this specification, the term “page” means one side of an original sheet. “One page” means an amount corresponding to the entire image read from one side of the document sheet.

  In the first memory use method shown in FIG. 4A, image data D1 of an amount smaller than one page of the original 40 in the bitmap memory 31 (for example, half of one page) for each image reading job. Is secured, and the memory area 60 capable of storing one page of compressed image data D2 is secured in the code memory 35. This method is applied when the free memory capacity, which is the capacity of the released memory area in the code memory 35, is greater than or equal to the capacity necessary for securing the memory area 60.

  When the first method of using the memory is applied, for example, a stage in which image data D1 for a half page is stored in the memory area 50 (that is, more than a stage in which image data D1 for a page is stored). At an early stage), the compression of the image data D1 is started. The compressed image data D2 is stored in the memory area 60 for one page and transferred to the storage 14 in units of pages.

  In the second method of using the memory shown in FIG. 4B, for each image reading job, a memory area 70 capable of storing one page of image data D1 of the original 40 is secured in the bitmap memory 31. In this method, a memory area 80 capable of storing a smaller amount of image data D2 than one page is secured in the memory 35. This method is applied when the memory area 60 for one page cannot be secured in the code memory 35. The memory capacity of the memory area 80 is set to an amount that does not exceed the free memory capacity of the code memory 35.

  When the second method of using the memory is applied, after each page of image data D1 starts to be stored in the memory area 70, each time the image data D1 is stored in an amount that can be compressed and stored in the memory area 80. Then, the image data D1 is compressed. Then, the compressed image data D2 temporarily stored in the memory 80 is sequentially transferred to the storage 14. The storage 14 holds the image data D2 for one page.

  In the second memory usage method, when the free memory capacity of the code memory 35 is less than one page, another image reading job that is not an image reading job to be allocated for memory is terminated and free memory is used. There is an advantage that reading of the document 40 can be started without waiting for the capacity to increase to one page or more. However, compared to the first method of using the memory, the number of compression processes per page increases and the transfer efficiency decreases, and the performance (processing speed) of a series of operations for scanning the original 40 and storing data in the storage 14 ) May be lower than when the first memory usage method is applied.

  FIG. 5 shows a functional configuration of the controller of the image processing unit. The controller 34 includes an input mode determination unit 341, a necessary memory capacity calculation unit 342, a free memory capacity calculation unit 343, and a memory area allocation unit 344. These constituent elements are functional elements realized by the processor 340 executing the control program.

  The input mode determination unit 341 determines an input mode related to memory allocation in the image processing unit 13 based on information provided from the main controller 10 regarding the image reading job input to the MFP 1. The input modes are classified according to the type of input image reading job, the number of image scanners 17A and 17B to be used, and reading conditions (reading target page size, resolution, color mode, etc.).

  The required memory capacity calculation unit 342 calculates a required memory area (X) that is a memory capacity of a memory area to be allocated to the code memory 35 when storing compressed image data D2 for one page. The required memory area (X) depends on the reading conditions.

  The free memory capacity calculation unit 343 calculates the memory capacity (Y) of the memory area available for input in the code memory 35. The memory capacity (Y) depends on the type of image reading job (whether the job requires a memory area for output).

  The memory allocation unit 344 uses a memory usage method to be applied to page reading according to the required memory area (X) and the memory capacity (Y) obtained by the required memory capacity calculation unit 342 and the free memory capacity calculation unit 343. Decide. Then, the memory allocation unit 344 allocates a memory area having a memory capacity corresponding to the memory usage method to each of the bitmap memory 31 and the code memory 35 and notifies the DMA transfer circuit 33 of the address of the allocated memory area.

  FIG. 6 shows the flow of the setting process of the memory use method executed by the controller of the image processing unit. When notified from the main controller 10 that the image reading job has been input to the MFP 1, the controller 34 executes the processing of FIG. 6 (this is called a main routine).

  In this description, the execution of the main routine is started when the image reading job (J1) for reading the image of the original 40 is input, and the user selects the image scanner 17A when the image reading job (J1) is input. And It is possible that another image reading job (J2) is submitted before the image reading job (J1) is completed. In this case, the image scanner 17B is used for the image reading job (J2), and the image data D1 is input to the image processing unit 13 from each of the two image scanners 17A and 17B.

  In the process in which the image scanner 17A sequentially reads images one page at a time sequentially from the document 40, when the reading of each page is started (YES in S01), the controller 34 determines whether the mode of the document 40 is “mixed”. It is checked whether or not (S02). “Mixed loading” is an aspect in which the sizes of a plurality of original sheets constituting the original 40 are not the same, that is, an aspect in which the size of at least one original sheet is different from the sizes of other original sheets.

  In this embodiment, the check in step S02 is a check as to whether or not the mixed loading mode is selected by the user when the image reading job (J1) is submitted. The controller 34 checks whether or not the mixed loading mode is selected based on the information regarding the setting of the image reading job (J1) acquired from the main controller 10. The controller 34 determines that the mode of the original 40 is mixed when the mixed loading mode is selected, and determines that the mode of the original 40 is not mixed when the mixed loading mode is not selected.

  When the form of the document 40 is not mixed (NO in S02), the controller 34 acquires the current number of image inputs (S03) and determines whether the number of image inputs has changed by comparing with the previously acquired number of image inputs. Check (S04). The number of image inputs is the number of image scanners 17A and 17B that are in use. The used state is a state from when the document 40 is set to the ADF paper feed tray until the last page is scanned.

  The number of image inputs acquired for the first time after execution of the main routine is started is “1”. When the image input number is acquired for the first time, the previous image input number is set to “0”. In the check in step S04, it is determined that the number of image inputs has changed from “0” to “1”, and the check result is YES.

  When the execution of the image reading job (J2) is started during the execution of the image reading job (J1), the number of image inputs acquired for the first time after the execution of the image reading job (J2) is started is “2”. . At this time, since the previous number of image inputs is “1”, it is determined in step S04 that the number of image inputs has changed from “1” to “2”, and the check result is YES.

  In addition, when the image reading job (J2) is started during the execution of the image reading job (J1) and the use of the image scanner 17 by one of the image reading jobs is finished thereafter, the use of the image scanner 17 is stopped. The number of image inputs acquired for the first time after the completion is “1”. At this time, since the previous number of image inputs is “2”, it is determined in step S04 that the number of image inputs has changed from “2” to “1”, and the check result is YES.

  On the other hand, the number of image inputs acquired in the second and subsequent executions of step S03 when only one of the image reading job (J1) and the image reading job (J2) is being executed is “1”. . At this time, since the number of previous image inputs is “1”, it is determined that the number of image inputs has not changed in the check in step S04, and the check result is NO. If the check result in step S04 is NO, the flow proceeds to step S08.

  The number of image inputs acquired in the second and subsequent executions of step S03 in a state where both of the two image scanners 17A and 17B are being used by the image reading job (J1) and the image reading job (J2) is “2”. It is. At this time, since the previous number of image inputs is “2”, it is determined in the check in step S04 that the number of image inputs has not changed, and the check result is NO.

  If the check result in step S04 is YES, the flow proceeds to step S05, and the controller 34 performs a necessary memory capacity calculation routine (S05), a free memory capacity calculation routine (S06), and a memory allocation routine (S07). Run. By executing these three routines, the memory allocation of the bitmap memory 31 and the code memory 35 is changed as necessary.

  After executing the routine of step S07, the controller 34 checks whether or not the image input from the image scanner 17A (input of the image data D1) has been completed (S08). When the image scanner 17B is also used, it is also checked whether the image input from the image scanner 17B is completed. That is, it is checked whether or not one or both of the image scanners 17A and 17B are changed from being in use to both of them.

  If the check result in step S08 is YES, the controller 34 ends the execution of the main routine. At that time, the controller 34 performs a memory management process for releasing the input memory area secured in each of the bitmap memory 31 and the code memory 35 to make an empty memory area.

  FIG. 7 is a flowchart of the required memory capacity calculation routine of FIG.

  The controller 34 acquires the resolution of reading by the image scanner 17, the color mode of reading, and the image size of the page on which scanning by the image scanner 17 is started or about to start from the main controller 10 (S51, S52, S53). The resolution is one selected by the user from the options of 200 dpi, 300 dpi, 400 dpi, and 600 dpi. The color mode is any one selected by the user from options such as full color, gray scale, and monochrome two gradations, and is acquired as information specifying the number of bits per pixel. The image size is the size of the original sheet placed on or about to be placed on the platen glass of the image scanner 17.

  The controller 34 calculates the necessary memory capacity (X) based on the acquired resolution, color mode, and image size (S54). The necessary memory capacity (X) is a memory capacity necessary for storing in the code memory 35 the image data D2 obtained by compressing the image data D1 for one page. For example, the reading condition is a condition that an image of an A3 size (297 mm × 420 mm) document sheet, which is the maximum reading size, is read in full color (24 bits per pixel) at 600 dpi (this is referred to as “maximum condition”). In some cases, the data amount of the image data D1 is approximately 200 megabytes. The product of the data amount and the compression rate is the required memory capacity (X). The compression rate depends on the content of the image, and the true compression rate is unknown at the stage where compression is not performed. Therefore, in step S54, the required memory capacity (X) is calculated using a compression rate (for example, 70%) set to be lower than the lowest possible compression rate. If the data amount of the image data D1 is 200 megabytes as in this example, the required memory capacity (X) is 140 megabytes.

  FIG. 8 is a flowchart of the free memory capacity calculation routine of FIG.

  The controller 34 acquires the current free memory capacity (Z) of the code memory 35 (S61). The free memory capacity (Z) depends on the memory capacity of the memory area already allocated for input or output. That is, the free memory capacity (Z) is obtained by subtracting the memory capacity of the memory area that is not released from the memory capacity of the code memory 35. Note that it is not always necessary to acquire the free memory capacity (Z) as the first processing of the routine, and it is sufficient to acquire the free memory capacity (Z) before executing step S70 described later.

  The controller 34 checks whether or not the document reading job of interest is a job that requires a memory area for output. The document reading job of interest is a document reading job to which memory allocation is to be performed among one or two document reading jobs being executed, that is, the start of reading one page that triggered the execution of this routine (FIG. 6). This is a document reading job corresponding to step S01).

  When the document reading job to be noticed is scan-to-box, the image data D2 is not read from the storage 14, so that an output memory area is unnecessary. On the other hand, when the document reading job to be noticed is copy or data transmission, an output memory area is required to decompress the compressed image data D2 and send it to the printer engine 19 or the communication interface 15. is there.

  When the controller 34 determines that the document reading job of interest does not require the output memory area by referring to the information indicating the necessity of the output memory area for each job type (S62) NO), the flow proceeds to step S63.

  In step S63, the controller 34 sets the number of output pages (Q) to “0”. The number of output pages (Q) is a variable indicating how many pages of output memory area are required. Since expansion is performed in units of pages, the number of output pages (Q) is determined in order to calculate the necessary memory capacity based on the amount of image data for one page. The number of output pages (Q) being “0” means that the required memory capacity is 0 pages, that is, no output memory area is required.

  On the other hand, if the controller 34 determines that the document reading job of interest is a job that requires an output memory area (YES in S62), the flow proceeds from step S62 to step S64, and the output mode is set as follows. Accordingly, the number of output pages (Q) is determined.

  First, in step S64, the controller 34 needs a memory area of at least one page, so the number of output pages (Q) is set to “1”.

  Next, whether or not the output mode is a mode in which the output mode is “rotation” for changing the orientation of the image by rotating it by a specified angle of 90 degrees or “enlarging / reducing” for changing the size of the image. Is checked (S65). Each of “rotation” and “enlargement / reduction” requires a memory area for two pages, one for each page, for storing data before processing and for storing data after processing. Therefore, when it is determined that the output mode is a mode for performing “rotation” or “enlargement / reduction” (YES in S65), the controller 34 sets the number of output pages (Q) to “2” (S66). If the check result in step S65 is NO, the flow jumps to step S66 and proceeds to step S67.

  In the output mode in which both “rotation” and “enlargement / reduction” processes are performed, one process and the other process are sequentially performed using the memory area for a total of two pages. That is, also in this case, the number of output pages (Q) is set to “2”.

  In step S67, the controller 34 checks whether or not the output mode is a mode for performing “aggregation” (also referred to as “N in 1”) in which images of a plurality of pages are arranged on one page. N is the number of pages (2, 4, 6, 9, etc.) collected on one side of the recording sheet. In the “aggregation”, the image data D2 for N pages is stored in the code memory 35, so a memory area for N pages is required. When “rotation” or “enlargement / reduction” and “aggregation” are performed, a memory area of 2N pages is required.

  If the check result in step S67 is YES, the controller 34 replaces the number of output pages (Q) with a value “N * Q” that is N times that number (S68). That is, in step S68, the number of output pages (Q) becomes “N” or “2N”.

  The flow proceeds from step S68 to step S69. If the check result in step S67 is NO, the flow jumps to step S67 and proceeds to step S69.

  In step S <b> 69, the controller 34 calculates the memory capacity (V) of the output memory area to be allocated to the code memory 35. The calculation performed here is the required memory capacity (X) calculated in step S54 of FIG. 7 corresponding to the data amount of the image data D2 for one page, and the number of output pages obtained by the processing from step S62 to step S68 ( Q). That is, the product of the required memory capacity (X) and the number of output pages (Q) is calculated as the memory capacity (V). When the number of output pages (Q) is “0”, the memory capacity (V) is “0”.

  Subsequently, the controller 34 calculates the memory capacity (Y) of the memory area available for input in the free memory area (Z) of the code memory 35 (S70). The memory capacity (Y) is a value obtained by subtracting the output memory capacity (V) calculated in step S69 from the free memory capacity (Z) acquired in step S61. The controller 34 having calculated the memory capacity (Y) returns to the main routine and executes the memory allocation routine.

  FIG. 9 is a flowchart of the memory allocation routine of FIG.

  The controller 34 determines whether or not one page of the image data D2 compressed by the compression / decompression circuit 32 can be stored in the code memory 35 (S71). In this determination, the controller 34 compares the already-calculated required memory capacity (X) with the memory capacity (Y). When the memory capacity (Y) is equal to or larger than the necessary memory capacity (X), the controller 34 determines that one page can be stored and sets the determination result to “possible”. If the memory capacity (Y) is less than the required memory capacity (X), the controller 34 determines that one page cannot be stored and sets the determination result to “impossible”.

  When the determination result is “possible”, the controller 34 performs memory allocation to the code memory 35 and the bitmap memory 31 to which the first memory usage method is applied. That is, as shown in FIG. 4A, a memory area 60 having a memory capacity capable of storing one page of image data D2 is secured in the code memory 35. At this time, the memory capacity of the memory area 60 is set to a necessary memory capacity (X). Further, a memory area 50 having a memory capacity capable of storing less than one page of image data D1 is secured in the bitmap memory 31.

  On the other hand, if the determination result is “impossible”, the controller 34 executes a band size setting subroutine for setting the band size (R) (S74), and thereafter performs memory allocation using the second memory usage method. Perform (S75). The band size (R) is a data amount of the image data D1 that is read from the bitmap memory 31 and transferred to the code memory 35 in each compression process in the process of compressing the image data D1 for one page in a plurality of times. is there. Note that the subroutine of step S74 may be executed after the process of step S75 is executed.

  FIG. 10 is a flowchart of the band size setting subroutine of FIG. In this subroutine, the largest possible band size (R) is set according to the memory capacity (Y) of the free memory area available for input. This is because by increasing the band size (R), the number of transfers from the bitmap memory 31 to the compression / decompression circuit 32 is reduced and the performance is improved.

  The controller 34 obtains the necessary memory capacity (X) and the memory capacity (Y) described above (S701, S702), and calculates the ratio (Y / X) of the memory capacity (Y) to the necessary memory capacity (X) ( S703). Then, the controller 34 determines the band size as follows according to the magnitude relationship between the predetermined threshold values (50%, 34%, 25%, 20%, 17%) and the ratio (Y / X). (R) is set. Note that the threshold values and the number of threshold values are not limited to examples, and can be selected as appropriate.

  When the ratio (Y / X) is not less than 50% and less than 100% (YES in S704), the controller 34 sets 0.5 times the necessary memory capacity (X) as the band size (R) (S705). In this case, one page of image data D1 is compressed in two steps.

  When the ratio (Y / X) is not less than 34% and less than 50% (YES in S706), the controller 34 sets 0.34 times the necessary memory capacity (X) as the band size (R) (S707). In this case, one page of image data D1 is compressed in three steps.

  When the ratio (Y / X) is 25% or more and less than 34% (YES in S708), the controller 34 sets 0.25 times the necessary memory capacity (X) as the band size (R) (S709). In this case, the image data D1 for one page is compressed in four steps.

  When the ratio (Y / X) is 20% or more and less than 25% (YES in S710), the controller 34 sets the band size (R) to 0.2 times the necessary memory capacity (X) (S711). In this case, one page of image data D1 is compressed in five steps.

  When the ratio (Y / X) is not less than 17% and less than 20% (YES in S712), the controller 34 sets 0.17 times the necessary memory capacity (X) as the band size (R) (S713). In this case, one page of image data D1 is compressed in six steps.

  When the ratio (Y / X) is less than 17% (NO in S712), the controller 34 sets the predetermined minimum band size (Rmin) to the band size (R) (S714). Instead of the minimum band size (Rmin), the memory capacity (Y) may be set to the band size (R). In any case, in this case, the image data D1 for one page is compressed in seven or more times.

  As described above, in the MFP 1, the memory allocation for the image processing unit 13 is optimized every time reading of one page from the document 40 is started by either the image scanner 17A or the image scanner 17B. Next, several examples of memory allocation are given.

  FIG. 11 shows a first example of memory allocation.

  In FIG. 11A, one image reading job (Ja) is being executed, and the document 43 is read using the image scanner 17A. The image scanner 17B is not used. The document 43 has a plurality of document sheets of A3 size (maximum reading size), and reading of pages of A3 size is in progress. The image reading job (Ja) is a scan-to-box. In scan-to-box, there is no need to allocate a memory area for output.

  In FIG. 11A, the first memory usage method is applied to reading a page. That is, a memory area 51 capable of storing an amount of image data D1a less than one page is allocated to the bitmap memory 31, and the compressed image data D2a for one page can be stored in the code memory 35. A memory area 61 is allocated.

  The memory capacity of the code memory 35 is selected to be a minimum amount capable of storing the image data D2 for two pages when reading under the above-mentioned maximum conditions when designing the circuit of the image processing unit 13. Accordingly, in the state of FIG. 11A, the code memory 35 has a free memory capacity for at least one page under the maximum condition.

  FIG. 11B shows a state shifted from the state of FIG. In FIG. 11B, an image reading job (Jb) for scanning the document 44 using the image scanner 17B is being executed together with the above-described image reading job (Ja) using the image scanner 17A. The image reading job (Jb) is a scan-to-box. The original 44 has a plurality of A3 size original sheets. The image scanner 17B reads the A3 size page (SB) and the image scanner 17A that has already been in progress at the start of the reading (SB). A3 page read (SA) is in progress.

  Assume that at the start of reading (SB), reading (SA) has already been performed by applying the first method of using the memory as in FIG. At the start of reading (SB), execution of the main routine of FIG. 6 determines the memory usage to be applied to reading (SB).

  When determining the memory usage method to be applied to reading (SB), the code memory 35 has an empty memory area for at least one page under the maximum condition as described above. That is, the code memory 35 has a free memory capacity (Z) that is equal to or larger than the required memory capacity (X) for reading (SB), and the determination result in step 72 in FIG.

  Therefore, in FIG. 11B, the first memory usage method is applied to reading (SB). That is, a memory area 52 capable of storing an amount of image data D1b smaller than one page is allocated to the bitmap memory 31, and the compressed memory 35 can store one page of compressed image data D2b. A memory area 62 is allocated. The bitmap memory 31 has two memory areas 51 and 52 for input, and the code memory 35 has two memory areas 61 and 62 for input.

  FIG. 12 shows a second example of memory allocation.

  In FIG. 12A, one image reading job (Jc) is being executed, and the document 45 is read using the image scanner 17A. The image scanner 17B is not used. The document 45 has a plurality of A3 size document sheets, and reading of an A3 size page is in progress. The image reading job (Jc) is a scan-to-box.

  In FIG. 12A, the first memory usage method is applied to reading a page. That is, a memory area 53 capable of storing an amount of image data D1c less than one page is allocated to the bitmap memory 31, and the compressed image data D2c for one page can be stored in the code memory 35. A memory area 63 is allocated. In the state of FIG. 12A, the code memory 35 has a free memory capacity (Z) for at least one page under the maximum condition.

  FIG. 12B shows a state shifted from the state of FIG. In FIG. 12B, an image reading job (Jd) for reading the document 46 using the image scanner 17B is being executed together with the above-described image reading job (Jc) using the image scanner 17A. The image reading job (Jd) is a copy. The document 46 has a plurality of A3 size document sheets. The image scanner 17B reads an A3 size page (SD) and the image scanner 17A that has already been in progress at the start of the reading (SD). A3 page read (SC) is in progress.

  Assume that at the start of reading (SD), reading (SC) has already been performed by applying the first method of using the memory as in FIG. At the start of reading (SD), execution of the main routine of FIG. 6 determines the memory usage applied to reading (SD).

  When determining the memory usage method applied to reading (SD), the code memory 35 has a free memory area (Z) for at least one page under the maximum condition as described above. However, since the image reading job (Jd) accompanied by reading (SD) is a copy, it is necessary to allocate a memory area 91 for output to the code memory 35. Therefore, the memory capacity obtained by subtracting the memory capacity (V) of the output memory area 91 from the free memory capacity (Z) of the code memory 35, that is, the memory capacity (Y) of the memory area 81 available for input is compressed. Less than one page of the subsequent image data D2d. The determination result in step 72 in FIG. 9 is “impossible”.

  Therefore, in FIG. 12B, the second memory use method is applied to reading (SD). That is, a memory area 71 capable of storing one page of image data D1d is allocated to the bitmap memory 31, and the code memory 35 can store compressed image data D2d in an amount less than one page. A memory area 81 is allocated. The bitmap memory 31 has two memory areas 53 and 71 for input and one memory area 92 for output, and the sign memory 35 has two memory areas 63 and 81 for input and one memory area for output. 91.

  When allocating the memory area 81, the band size setting subroutine shown in FIG. 10 is executed, and the above-described band size (R) related to reading from the memory area 71 of the bitmap memory 31 is set. The image data D1d for one page stored in the memory area 71 is read for each band size and transferred to the compression / decompression circuit 32, and the compressed image data D2d for the band size is temporarily stored in the memory area 71. After that, it is transferred to the storage 14. In the storage 14, image data D2d for one page inputted for each band size is stored as a set of data for output.

  The memory area 92 of the bitmap memory 31 stores the image data D3d read from the storage 14 to the memory area 91 of the code memory 35 and then decompressed by the compression / decompression circuit 32. Thereafter, the image data D3d is transferred from the memory area 92 to the printer engine 19.

  FIG. 13 shows a third example of memory allocation. The state shown in FIG. 13A is the same as the state shown in FIG. In the state shown in FIG. 13B, the execution of the image reading job (Jc) being executed in FIG. 13A is completed, and the execution of the image reading job (Jd) using the image scanner 17B is continued. State.

  In FIG. 13B, reading (SD) of an A3 size page of the document 46 is in progress. However, the current page to be read is a page different from the page to be read in FIG. It is assumed that the image reading job (Jc) using the image scanner 17A has already ended at the start of such reading (SD).

  The memory capacity (Y) available for input in the code memory 35 at the start of reading (SD) is the free memory capacity (Z) of the code memory 35, which is an amount of one page or more under the maximum conditions. This is because the code memory 35 has a memory area of two pages or more under the maximum condition, and the memory area 81 previously allocated is considered to be released, and the memory of the memory area 91 allocated for output This is because the capacity is an amount of one page or less under the maximum condition.

  Therefore, in FIG. 13B, the first memory usage method is applied to reading (SD). That is, a memory area 56 capable of storing less than one page of image data D1d is allocated to the bitmap memory 31, and a memory area capable of storing one page of compressed image data D2d is allocated to the code memory 35. 66 is assigned. The bitmap memory 31 has one memory area 56 for input and one memory area 92 for output, and the code memory 35 has one memory area 66 for input and one memory area 91 for output. ing.

  As described above, in the example of FIG. 13, the number of image inputs is changed from “2” to “1” during the execution of the image reading job (Jd) using the image scanner 17B. The memory usage method is changed from the second memory usage method to the first memory usage method.

  FIG. 14 shows a fourth example of memory allocation.

  In FIG. 14A, one image reading job (Je) is being executed, and the document 47 is read using the image scanner 17A. The image scanner 17B is not used. The image reading job (Je) is a copy. The document 47 has a plurality of A4 size document sheets, and reading of an A4 size page is in progress. In parallel with reading, printing by the printer engine 19 is performed. Output memory areas 93 and 94 are allocated to the code memory 35 and the bit map memory 31 as shown in FIG.

  In FIG. 14A, the first memory usage method is applied to reading a page. That is, a memory area 54 capable of storing an amount of image data D1e smaller than one page is allocated to the bitmap memory 31, and the compressed image data D2e for one page can be stored in the code memory 35. A memory area 64 is allocated. The free memory capacity (Z) of the code memory 35 is an amount obtained by subtracting the memory capacity of the input memory area 64 and the output memory area 93 from the memory capacity of the code memory 35.

  FIG. 14B shows a state shifted from the state of FIG. In FIG. 14B, an image reading job (Jf) for reading the image of the document 48 using the image scanner 17B is being executed together with the above-described image reading job (Je) using the image scanner 17A. The image reading job (Jf) is a scan-to-box. The original 48 has a plurality of A4 size original sheets. The image scanner 17B reads an A4 size page (SF) and the image scanner 17A that has already been in progress at the start of the reading (SF). A4 page read (SE) is in progress.

  Assume that at the start of reading (SF), reading (SE) has already been performed by applying the first method of using the memory as in FIG. At the start of reading (SF), the execution of the main routine of FIG. 6 determines the memory usage to be applied to reading (SF).

  When determining the memory usage to be applied to reading (SF), the free memory capacity (Z) of the code memory 35 was greater than the required memory capacity (X) for reading (SF).

  Therefore, in FIG. 14B, the first memory usage method is applied to reading (SF). That is, a memory area 55 capable of storing image data D1f of an amount less than one page is allocated to the bitmap memory 31, and compressed image data D2f for one page can be stored in the code memory 35. A memory area 65 is allocated. The bitmap memory 31 has two memory areas 54 and 55 for input and one memory area 94 for output, and the sign memory 35 has two memory areas 64 and 65 for input and one memory area for output. 93.

  FIG. 15 is a flowchart showing another example of the setting process of the memory usage method. The main routine of FIG. 6 determines a memory usage method to be applied to reading in accordance with the size of the page to be read when the document mode is mixed. On the other hand, in the example of FIG. 15, when the document mode is mixed loading, the size of all pages of the document is regarded as the maximum size in the document, and the memory use method applied to reading is used. It is a decision. The contents of the process in FIG. 15 are substantially the same as the contents of the process in FIG. 6 except that the size of the page to be read is uniformly regarded as the maximum size in the mixed loading mode.

  The controller 34 acquires the current number of input images when reading of each page is started in the process in which the image scanner 17 sequentially reads images one page at a time from the document 40 (YES in S21) (S22). Then, it is checked whether or not the image input number has changed by comparing with the previously acquired image input number (S23). If the check result in step S23 is NO, the flow proceeds to step S27.

  If the check result in step S23 is YES, the flow proceeds to step S24, and the controller 34 executes a required memory capacity calculation routine (S24). Subsequently, the controller 34 calculates the memory capacity (Y) of the memory area available for input by the procedure of FIG. 8 (S25), and executes the memory allocation of the procedure of FIG. 9 (S26). Thereafter, the controller 34 checks whether or not one or both of the image scanners 17A and 17B is changed from being used to being used (S27).

  If the check result in step S27 is NO, the flow returns to step S21. If the check result in step S27 is YES, the controller 34 finishes executing the process of FIG. At that time, the controller 34 releases the input memory area secured in each of the bitmap memory 31 and the code memory 35 to make an empty memory area.

  FIG. 16 is a flowchart of the required memory capacity calculation routine of FIG.

  The controller 34 acquires the resolution of reading by the image scanner 17 and the color mode of reading from the main controller 10 (S41, S42). Further, when the document mode is not mixed (NO in S43), the image size of the page on which scanning by the image scanner 17 is started or is about to start is acquired from the main controller 10 (S54). When the document mode is mixed (YES in S43), the controller 34 determines the largest document sheet size (maximum size) among the plurality of document sheets constituting the document as the image size (S46). Then, the controller 34 calculates the necessary memory capacity (X) based on the resolution, the color mode, and the image size (S45).

  In the above embodiment, the image scanners 17A and 18 need not be the same type. One or both of the image scanners 17 and 17B may not be able to read color, or may not be provided with an ADF.

  The present invention can also be applied when the number of image scanners 17A and 17B is three or more. Depending on the number of image scanners, the memory capacity of the code memory 35 can be increased, for example, by adding memory modules.

  The configuration of the image processing unit 13 is not limited to the example shown in FIG. For example, the code memory 35 may be formed in the ASIC 30. Further, the ASIC 30 may be provided in an external memory device without forming the bitmap memory 31 in the ASIC 30.

  A plurality of image reading jobs using an output memory area can be performed in parallel, such as copying and facsimile transmission. In that case, the memory area for input by applying the first memory usage method or the second memory usage method according to the remaining memory capacity (Y) in which a plurality of memory areas for output are allocated to the code memory 35. Can be assigned.

  In reading using the first memory usage method, when transferring the image data D2 from the code memory 35 to the storage 14, a new memory area is secured in addition to the memory area storing the image data D2. You may do it. Thereby, reading to the storage 14 and writing to the code memory 35 can be performed in parallel to improve performance.

1 MFP (image processing device)
17A, 17B Image scanner (input means)
19 Printer engine (output means)
15 Communication interface (output means)
D1 Image data (uncompressed image data)
D2 Image data (compressed data)
32 Compression / decompression circuit (compression means, decompression means)
35 Code memory (volatile memory)
60 memory area (first area)
80 memory area (second area)
91, 93 Memory area (third area)
344 Memory allocation unit (mode setting means, area securing means)
14 Storage 33 DMA transfer circuit (first transfer means, second transfer means)

Claims (11)

An image processing apparatus for outputting any of images input by each of one or a plurality of input means by an output means,
Mode setting means for setting, for each image, a mode for compressing the image to one of the first mode and the second mode;
Uncompressed image data that is uncompressed image data of the image is compressed for each first unit when the mode of the image is set to the first mode, and the second mode Compression means for generating compressed data by compressing every second unit smaller than the first unit,
Volatile memory,
A first area for storing all the compressed data for each of the images for which the first mode is set is secured in the memory, and for each of the images for which the second mode is set. An area for securing a second area for storing one piece of the compressed data and a third area for storing all the compressed data of the image that is the target of the current processing by the output means Securing means;
Non-volatile storage,
When all the compressed data of the image for which the first mode is set is stored in the memory, all the compressed data is transferred to the storage and stored in association with the image, and the second mode is A first transfer means for transferring the compressed data of the set image to the storage each time the compressed data of the image is stored in the memory, and storing the compressed data in the storage in association with the image;
Second transfer means for transferring all the compressed data of the image that is the target of the current output processing by the output means from the storage to the memory;
An image processing apparatus comprising: The mode setting means sets the mode of the image to the first mode when the memory has a space for the first area of the image, and to the second mode otherwise. Set,
The image processing apparatus according to claim 1. The compression means has a size capable of storing the compressed data of the image in which the mode is set to the second mode in a free area in which the memory is not used or secured. Generated by compressing the uncompressed data of the image as the second unit;
The image processing apparatus according to claim 1. Each of the input means inputs each of the images by scanning a printing surface of paper,
The first unit is the size of one side of the paper,
The compression means stores the compressed data of the image in which the mode is set to the second mode, and the (1 / M) plane portion (where M ≧ 2) of the image is stored in the empty area. If the (1 / (M-1)) plane portion cannot be stored, the uncompressed data of the image is compressed using the (1 / M) plane size as the second unit. Generated by
The image processing apparatus according to claim 3. The area securing means secures the first area, the second area, and the third area based on the characteristics of the image in which the uncompressed data is stored.
The image processing apparatus according to claim 1. When a plurality of the images are continuously input from the same input means among the input means,
The mode setting means sets the mode in common for the plurality of images,
The area securing means secures one of the first area and the second area according to the set mode for the first image of the plurality of images, Then continue to reserve the reserved area for the remaining images,
The image processing apparatus according to claim 1. When a plurality of the images are continuously input from the same input means of the input means and the paper sizes of the plurality of images are different,
The mode setting means commonly sets the mode according to the largest size of the plurality of images for the plurality of images,
The area securing means secures one of the first area and the second area according to the set mode for the first image of the plurality of images, Then continue to reserve the reserved area for the remaining images,
The image processing apparatus according to claim 1. When a plurality of the images are continuously input from the same input means of the input means and the sizes of the plurality of images are different,
The mode setting means sets the mode for each of the plurality of images according to the size of each image,
When each of the plurality of images is input, the area securing unit is configured to set the first area and the second area for the input image according to a mode set for the input image. Secure any one of the areas, and release the area when all the compressed data of the input image is transferred to the storage;
The image processing apparatus according to claim 1. The area securing means performs processing for enlarging, reducing, rotating, or consolidating with one other image together with other images as image processing for the current output target of the output means In this case, a fourth area for storing processed image data of the image subjected to the image processing is further secured in the memory.
The image processing apparatus according to claim 1. An image processing apparatus having one or more input means, a volatile memory, a nonvolatile storage, and an output means for outputting any of the images input by each of the one or more input means An image data processing method in
For each image, the mode for compressing the image is set to one of the first mode and the second mode,
The uncompressed image data of the image is compressed for each first unit when the mode of the image is set to the first mode, and when the second mode is set Generating compressed data by compressing every second unit smaller than the first unit;
A first area for storing all the compressed data for each of the images for which the first mode is set is secured in the memory, and for each of the images for which the second mode is set. Securing a second area for storing one of the compressed data, securing a third area for storing all the compressed data of the image that is the target of the current processing by the output means;
Each time all the compressed data of the image for which the first mode is set is stored in the memory, all the compressed data is transferred to the storage and stored in association with each other, and the second mode is set Each time the compressed data of the image being stored is stored in the memory, the compressed data is transferred to the storage and stored in the storage in association with other compressed data of the image,
From the storage, to transfer all the compressed data of the image that is the target of the current output processing by the output means to the memory,
An image data processing method characterized by the above. A computer program used in a computer for performing processing for outputting any of images input by each of one or more input means,
On the computer,
For each image, a process for setting the mode for compressing the image to one of the first mode and the second mode is executed,
The uncompressed image data of the image is compressed for each first unit when the mode of the image is set to the first mode, and when the second mode is set By performing compression every second unit smaller than the first unit, a process of generating compressed data is executed,
A first area for storing all the compressed data for each of the images for which the first mode is set is secured in a volatile memory, and the image for which the second mode is set A process for securing a second area for storing one piece of the compressed data every time and a third area for storing all the compressed data of the image that is the target of the current output process And execute
Each time all the compressed data of the image for which the first mode is set is stored in the memory, all the compressed data is transferred to a non-volatile storage and stored in association with each other, and the second mode Each time the compressed data of the image for which is set is stored in the memory, the compressed data is transferred to the storage and associated with other compressed data of the image and stored in the storage.
Causing the storage to execute a process of transferring all the compressed data of the image to be processed for current output to the memory;
A computer program characterized by that.

Images