JP5843168B2 - Image processing apparatus, image processing method, and program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program.

  Conventionally, in an image processing apparatus having a photographing function such as a digital camera or a mobile phone, when generating an image, Bayer data (original image data) captured from an image sensor is converted into JPEG (Joint Photographic Experts Group), Processing such as conversion to YUV data, shading correction, image correction such as edge enhancement, noise reduction, distortion correction, and enlargement / reduction is performed.

  When processing each of the processes described above, if the entire image is processed, each processing circuit becomes very large. Therefore, the image is divided in the horizontal direction by a predetermined number of lines, and each divided image data In addition, various image processing including intermediate image processing is performed. At this time, in the image processing apparatus, a technique is known in which a predetermined area is secured on a memory and the area is used as a ring buffer in order to temporarily hold intermediate image data in the course of image processing. (For example, see Patent Documents 1 and 2).

  The ring buffer is a method for limiting the intermediate image data to a smaller area in order to use the memory effectively. The memory is used for programs, stacks, data for program processing, Bayer data fetched from the CCD, display data, and the like. For this reason, by reducing the amount of use of image data that requires a large area, it is possible to make a large area necessary elsewhere. For this reason, there is an advantage that the processing speed of image processing or the like can be increased.

  FIG. 9 is a conceptual diagram for explaining a memory use situation according to the conventional technique. For example, the JPEG conversion module cannot directly process Bayer data, and temporarily converts Bayer data 1 to YUV data 2 for size change, color adjustment, etc., and converts the YUV data to 2 JPEG data 3 Convert to By storing the temporarily required YUV data 2 in the ring buffer 4 in a cyclic manner, the amount of image data used in the memory is reduced as compared with the case where the ring buffer 4 is not used. The ring buffer 4 uses only a fixed area regardless of the amount of data to be used. When the amount of data exceeds a part of the area, it is called a ring buffer because it returns to the beginning of the area and uses the area cyclically.

  For this reason, when the data processing speeds of the two modules on the data input side to the ring buffer 4 and the data output side from the ring buffer 4 are different, data is overwritten or written to an area where reading has not been performed. There is a possibility that data is read from an area that has not been processed. Therefore, the data input side module controls the data output side, or the data output side module controls the data input side, so that the unwritten area is not read or the area where reading has not been completed. To prevent new writes from being made. The controlled module notifies the control module of the process end line, and controls RAM access based on the information, its own process end line, and the number of lines in the ring buffer.

  FIG. 10 is a block diagram showing an operation control method for the data output side module and the input side module in the prior art. The data output module 12 starts reading data from the ring buffer 11 when data is written to the ring buffer 11 up to the number of lines required by the data output module 12 itself. Thereafter, when the processing speed of the data output side module 12 is slower than that of the input side, data writing from the data output side module 12 to the data input side module 10 is stopped so that unread data is not overwritten. Control (data write control). On the contrary, when the processing speed of the data output module 12 is faster than that of the data input module 10, the data output module 12 reads from the data input module 10 so as not to read the data of the line that has not been written. (Data write end line notification).

JP 2010-86497 A JP 2007-88806 A

  However, in the above prior art, as shown in FIG. 11, in order to temporarily hold intermediate data in addition to the storage area 20 for the original image data and the storage area 21 for the processed image data, There was a problem that a storage area 22 for the ring buffer had to be secured separately. Further, there has been a problem that the memory capacity required for the ring buffer 22 increases as the image size increases and the processing path increases.

  SUMMARY An advantage of some aspects of the invention is that it provides an image processing apparatus, an image processing method, and a program that can effectively use a memory without requiring a storage area dedicated to a ring buffer.

The present invention includes a storage means having at least a first area for storing original image data and a second area for storing image data after image processing for the original image, and image processing means for performing predetermined image processing. The original image data stored in the first area of the storage means is sequentially read out in a predetermined division unit, supplied to the image processing means, and the image data processed by the image processing means is stored in the storage In the process of sequentially storing in the second area of the means, the work area for the predetermined image processing by the image processing means is defined as the area in which the processed image data in the second area is not stored. to the first read already image data in the region has been storage area, and an image data transfer means for securing based on the use status of said storage means, said image data transfer means, said first A work area for performing the predetermined image processing until the size of the area in which the read image data is stored is equal to or larger than the size of the work area required for the predetermined image processing. The processed image data in the second area is secured in an unstored area, and the size of the area in which the read image data in the first area is stored is the predetermined image processing. At a predetermined timing after the size of the required work area is reached, the work area for performing the predetermined image processing is changed to an area in which the read image data in the first area is stored. It is characterized by switching.

The present invention includes a storage means having at least a first area for storing original image data and a second area for storing image data after image processing for the original image, and image processing means for performing predetermined image processing. The original image data stored in the first area of the storage means is sequentially read out in a predetermined division unit, supplied to the image processing means, and the image data processed by the image processing means is stored in the storage In the process of sequentially storing in the second area of the means, the work area for the predetermined image processing by the image processing means is defined as the area in which the processed image data in the second area is not stored. An image data transfer unit that secures the read image data in one area based on the use status of the storage unit, and the image processing unit includes the first area. The first image processing is performed in a predetermined division unit on the original image data stored in the image data, and the second image is processed in a predetermined division unit on the image data after the first image processing is performed. Processing is performed, and the work area is a work area for temporarily storing the result of the first image processing.

  According to the present invention, there is an advantage that a memory area for the ring buffer is not required and the memory can be used effectively.

It is a block diagram which shows the schematic structure of the digital camera by embodiment of this invention. It is a block diagram which shows the structure of ASIC31 by this embodiment. It is a conceptual diagram for demonstrating the memory utilization condition by this embodiment. 5 is a flowchart for explaining an operation of an image data transfer unit 45 according to the present embodiment. It is a conceptual diagram which shows the usage example of RAM32 by the belt unit by the image data transfer part 45 of this embodiment. 7 is a flowchart for explaining the operation of an image data transfer unit 47 according to the present embodiment. It is a conceptual diagram which shows the usage example of RAM32 by the belt unit by the image data transfer part 47 of this embodiment. It is a conceptual diagram which shows the usage example of RAM32 by the belt unit by the image data transfer part 47 of this embodiment. It is a conceptual diagram for demonstrating the memory utilization condition by a prior art. In the prior art, it is a block diagram which shows the operation control method of a data output side module and an input side module. It is a schematic diagram which shows the memory utilization state in a prior art.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

A. Configuration of Embodiment FIG. 1 is a block diagram showing a schematic configuration of a digital camera according to an embodiment of the present invention. In the figure, the CPU 30 controls the operation (photographing, image processing, etc.) of each part of the digital camera described later by executing a predetermined program. The ASIC 31 performs YUV data conversion, image correction such as edge enhancement and shading correction, and image enlargement and reduction on Bayer data input from the CCD 33 described later. Further, the ASIC 31 performs various image processing and the like on the image data input from the CCD 33 described later, and displays the image data on the LCD 35.

  The RAM 32 stores various parameters related to the operation of the CPU 30 and the ASIC 31, captured image data, image data after image processing is performed on the image data, and temporarily stores intermediate results in the image processing. save. The CCD 33 takes in an image formed through an optical system such as a lens as an electrical signal and supplies it to the ASIC 31 as captured image data (hereinafter referred to as original image data or Bayer data).

  The ROM 34 stores predetermined programs executed by the CPU 39 and the ASIC 31, operation parameters, and the like. The LCD 35 displays various menu screens, various setting items on the menu screen, operation parameters, a through image at the time of shooting, original image data that has been shot, and the like. The keyboard 36 includes buttons for setting and specifying various shooting parameters and operation modes, a shutter button, and the like. The power supply unit 37 includes various batteries (a primary battery, a secondary battery, and the like), and supplies power for operating the above-described units.

  Next, FIG. 2 is a block diagram showing a configuration of the ASIC 31 according to the present embodiment. In the figure, an ASIC 31 includes a DMAC (Dynamic Memory Access Controller) 40, a memory control unit 41, a CCD control unit 42, an image conversion unit 43, an image correction unit 44, an image data transfer unit 45, a pixel number conversion unit 46, an image data transfer A unit 47, a keyboard control unit 48, and an LCD control unit 49.

  The DMAC 40 directly accesses the RAM 32 and the ROM 34 using the memory control unit 41 without going through the CPU 30, and controls the storage and reading of the original image data or the transfer to various image processing units described later. The memory control unit 41 directly accesses the RAM 32 and the ROM 34 under the control of the DMAC 40, and exchanges data between the DMAC 40 and the RAM 32 and the ROM 34. The CCD control unit 42 drives and controls the CCD 33 and supplies the original image data captured by the CCD 33 to the DMAC 40.

  The image conversion unit 43 converts the original image data supplied via the image data transfer unit 45 into YUV data and supplies the YUV data to the image correction unit 44. The image correction unit 44 performs image correction such as edge enhancement and shading correction on the supplied image data, and supplies the image data to the image transfer unit 45. The image data transfer unit 45 controls transfer processing with the DMAC 40 when the image conversion unit 43 and the image correction unit 44 perform input / output of image data with the RAM 32. At this time, the intermediate data after the image correction by the image correction unit 44 is temporarily stored in ring buffers A and B described later.

  The pixel number conversion unit 46 enlarges or reduces the supplied image data. The image data transfer unit 47 controls transfer processing with the DMAC 40 when the pixel number conversion unit 46 inputs and outputs image data with the RAM 3. At this time, the intermediate data after the image correction by the image correction unit 44 temporarily stored in ring buffers A and B described later is supplied to the pixel number conversion unit 46. The keyboard control unit 48 controls input (scanning) of the keyboard 36. The LCD control unit 49 controls the display on the LCD 35 of original image data supplied from the DMAC 40 and image data subjected to image processing.

  FIG. 3 is a conceptual diagram for explaining a memory usage state according to the present embodiment. In this embodiment, the unused area of the image data storage area 51 after processing is used as the storage area 52 of the ring buffer A, and the available area of the storage area 50 of the original image data before image processing is used as the storage area 52 of the ring buffer A. The storage area 52 is characterized by being used while being switched at a predetermined timing based on the progress of image processing. That is, as the image processing proceeds, the original image data becomes used, so that an available storage area is generated in the storage area 50 for the original image data. The processed image data storage area 51 stores the processed image data as the image processing proceeds, but there is an unused area at the initial stage of the image processing. Therefore, in the present embodiment, when image processing is started, first, a part of the unused area of the processed image data storage area 51 is used as the storage area 52 of the ring buffer A, and then a predetermined timing is used. Thus, a part of the storage area 50 of the original image data is switched to be used as the storage area 53 of the ring buffer B. In the present embodiment, switching from the ring buffer A to the ring buffer B is performed at a timing when image processing has progressed to such an extent that a storage area sufficient as a work area can be secured in the storage area 50 of the original image data.

  The ring buffers A and B in the present embodiment are buffers that secure work areas while moving, and are not cyclically used so-called ring buffers. Since it functions as a work area corresponding to a ring buffer used for temporarily storing data, it will be referred to as a ring buffer in the following description. However, as described above, the ring buffers A and B of the present embodiment are not exclusively reserved on the memory as in the conventional ring buffer, but read out image data in the storage area 50 of the original image data. Is different in that it is used by switching the area where the image is stored or the unused area of the storage area 51 of the processed image data.

B. Operation of Embodiment Next, the operation of the above-described embodiment will be described.
FIG. 4 is a flowchart for explaining the operation of the image data transfer unit 45 according to the present embodiment. FIG. 5 is a conceptual diagram showing a usage example of the RAM 32 in belt units by the image data transfer unit 45 of the present embodiment. FIG. 6 is a flowchart for explaining the operation of the image data transfer unit 47 according to this embodiment. FIGS. 7A to 7D and FIGS. 8A to 8D are conceptual diagrams showing an example of use of the RAM 32 in units of belts by the image data transfer unit 47 of the present embodiment.

  In the following description, Bayer data is 16 bits / pix, intermediate data, and final data after image enlargement / reduction is 16 bits / pix of YYUV data. The input image size is 320 × 240, and the output image size is 640 × 480, which is four times the input image. Further, it is assumed that the intermediate image secures a ring buffer A for 320 × 60 pixels at 16 bits / pix of YYUV data of 320 × 240.

  The image data transfer unit 45 calculates the start address of the ring buffer A, outputs the ring buffer change signal as disabled (invalid), and initially uses the ring buffer A (step S10). As shown in FIG. 6A, the head address of the input image is set to 0x00000000, and the head address of the output image is set to 0x00025800. The size of the ring buffer A is (320 × 60 × 16) / 8 = 38400 (0x00009600 in hexadecimal) bytes. The start address of the ring buffer A is output image start address + output image data size−ring buffer size = 0x00025800 + 0x000BB800-0x00009600 = 0x000D7A00.

  Next, the image data transfer unit 45 calculates the number of belt lines according to the processing content (step S12). The number of belt lines that can be processed by each processing unit is set to MAX 32 lines. Assuming that each processing unit is used in MAX, the number of belt lines processed by the image data transfer unit 45 is 30 on both the input side and the output side, and the number of belt lines processed by the image data transfer unit 47 is The side is expanded to 15 lines, and the output side is expanded to 30 lines.

  Next, the image data transfer unit 45 calculates a ring buffer change address ADDRa (= start address + ring buffer size) that may change the ring buffer during reading of the input image (step S14). In this case, the ring buffer change address ADDRa = input start address + ring buffer size = 0x00009600. Next, the image data transfer unit 45 calculates the belt unit input start address ADDRb (step S16). In this case, since it is the first belt, the belt unit input start address ADDRb = 0x00000000.

  Next, the image data transfer unit 45 determines whether or not ADDRa> ADDRb (step S18). Here, since ADDRa> ADDRb (YES in step S18), the image data transfer unit 45 calculates a belt unit output start address, and performs data input / output in belt units according to the belt unit output start address. The number of output lines is calculated (step S24). Specifically, in the case of the first belt, the belt unit output start address is the start address “0x000D7A00” of the ring buffer A. The image data transfer unit 45 supplies the original image data for 30 lines (the first belt of 0x0000000 to 0x00004B00 in FIG. 5) to the image conversion unit 43, and outputs the output data (for 30 lines) from the image correction unit 44. As shown in FIG. 6A, the data is output to the storage area 52 (0x000D7A00) of the ring buffer A set in the storage area 51 of the processed image data. The number of output lines is 30 lines.

  Next, the image data transfer unit 45 determines whether or not all the processes have been completed (step S26). If not completed (NO in step S26), the process returns to step S16 and performs the above-described processes. The process for the second belt is repeated. In the case of the second belt, in step S16, 30 lines are added to the belt unit input start address ADDRb, and the belt unit input start address ADDRb = 0x00004B00 (see FIG. 6B, second belt). In this case, since ADDRa> ADDRb is determined as in the case of the first belt (YES in step S18), the image data transfer unit 45 calculates a belt unit output start address in step S24 (0x000D7A00 + 30 line = 0x000C500). In accordance with the belt unit output start address, 30 lines of the original image data (second belt of 0x00004B00) are supplied to the image conversion unit 43, and the output data (30 lines) from the image correction unit 44 is shown in FIG. As shown in (b), the data is output to the storage area 52 (0x000C500-) of the ring buffer A set in the storage area 51 of the processed image data. The number of output lines is 60 lines.

  Next, the image data transfer unit 45 returns to step S16 again, repeats the above-described processing, and performs processing for the third belt. In the process of the third belt, as shown in FIG. 6C, since the belt unit input start address ADDRb = 0x00009600 and ADDRa = ADDRb (NO in step S18), the image data transfer unit 45 It is determined whether or not the buffer change signal is enabled (step S20). In this case, since the ring buffer change signal is disabled (invalid) in the previous step S10 (NO in step S20), the image data transfer unit 45 outputs the output line number 60 as the changed line number LNc. Then, the ring buffer change signal is enabled and output to the image data transfer unit 47 and changed to the ring buffer B (step S23).

  Next, in step S24, the image data transfer unit 45 calculates a belt unit output start address. From now on, since the ring buffer B is used, the belt unit output start address = 0x00000000, and 30 lines of the original image data (the third belt from 0x00009600) are supplied to the image conversion unit 43 according to the belt unit output start address Then, as shown in FIG. 6C, the output data (for 30 lines) from the image correction unit 44 is stored in the storage area 53 (0x0000000) of the ring buffer B set in the empty area of the storage area 50 of the original image data. To ~). The number of output lines is 90 lines.

  Next, the image data transfer unit 45 returns to step S16 again, repeats the above-described processing, and performs processing for the fourth belt. In the process of the fourth belt, as shown in FIG. 6D, the belt unit input start address ADDRb = 0x00012C00 and ADDRa <ADDRb (NO in step S18). In this case, in the previous step S22, the ring Since the buffer change signal is enabled (YES in step S20), the image data transfer unit 45 calculates the belt unit output start address in step S24. In this case, since the ring buffer B is used, the belt unit output start address = 0x00004B00, and 30 lines of the original image data (the fourth belt of 0x00012C00) are supplied to the image conversion unit 43 according to the belt unit output start address. The output data (for 30 lines) from the image correction unit 44 is stored in the storage area 53 (0x0004B00 to 0x0004B00) of the ring buffer B set as a free area in the storage area 50 of the original image data, as shown in FIG. ). The number of output lines is 120 lines.

  Thereafter, the image data transfer unit 45 reads out the original image data and supplies it to the image conversion unit 43, and the processed intermediate data from the image correction unit 44 is set as a free area in the storage area 50 of the original image data. The process of storing in the ring buffer B is repeated until the final belt. When the process is completed for all of the original image data (YES in step S26), the process is terminated.

  On the other hand, the image data transfer unit 47 first calculates the number of belt lines according to the processing content (step S30). In this embodiment, as described above, the number of belt lines processed by the image data transfer unit 47 is 15 lines on the input side and 30 lines on the output side. Next, the image data transfer unit 47 determines whether there is input data in the ring buffer A or B (step S32). If there is no input data (NO in step S32), the process returns to step S32 and waits until there is input data.

  On the other hand, if there is input data in the ring buffer A or B (YES in step S32), the image data transfer unit 47 determines whether or not the ring buffer change signal is enabled (step S34). For example, in the processing of the image data transfer unit 45 described above, the ring buffer change signal is disabled from the first belt to the second belt. As described above, when the ring buffer change signal is disabled (NO in step S34), the image data transfer unit 47 selects the ring buffer A as a buffer to be used (step S40). Next, the image data transfer unit 47 receives the intermediate data from the ring buffer A, supplies the intermediate data to the pixel number conversion unit 46, and stores the image data from the pixel number conversion unit 46 in the storage area 51 of the processed image data. Then, the input line number LNd is calculated (step S42).

  Specifically, as shown in FIG. 7A, the image data transfer unit 47 starts from the storage area 52 (0x000D7A00) of the ring buffer A set in the storage area 51 for the processed image data. Intermediate data (15 lines) is supplied to the pixel number conversion unit 46, and the enlarged data (30 lines: the first half of the first belt) from the pixel number conversion unit 46 is stored in the image data storage area 51 after processing. (0x00025800 ~). At this time, since the input belt has 15 lines, the number of input lines LNd = 15.

  Next, the image data transfer unit 47 determines whether or not all the processes have been completed (step S44). If the processes have not been completed (NO in step S44), the process returns to step S32 to perform the above-described processes. repeat. Also in this case, since the ring buffer change signal is disabled, as shown in FIG. 7B, the image data transfer unit 47 sets the ring buffer A in the storage area 51 for the processed image data. The next intermediate data (15 lines) is supplied from the storage area 52 to the pixel number conversion unit 46, and the enlarged image data (30 lines: the second half of the first belt) from the pixel number conversion unit 46 is processed. The image data is stored in a storage area 51 (0x0002A300-). At this time, since the input belt has 15 lines, the number of input lines LNd = 30.

  Thereafter, similarly, as shown in FIG. 7C, the image data transfer unit 47 sets the next intermediate data (from the storage area 52 of the ring buffer A set in the storage area 51 of the processed image data to the next intermediate data ( 15 lines: the first half of the second belt) is supplied to the pixel number conversion unit 46, and the enlarged image data (30 lines: the first half of the second belt) from the pixel number conversion unit 46 is converted into the processed image data. Store in the storage area 51 (0x0002EE00-). At this time, since the input belt has 15 lines, the number of input lines LNd = 45.

  Further, as shown in FIG. 7D, the image data transfer unit 47 sets the next intermediate data (15th line: 15th line) from the storage area 52 of the ring buffer A set in the storage area 51 of the processed image data. 2nd half of the belt) is supplied to the pixel number conversion unit 46, and the enlarged image data (30 lines: second half of the second belt) from the pixel number conversion unit 46 is stored in the image data storage area 51 ( 0x0002EE00-). At this time, since the input belt has 15 lines, the input line number LNd = 60.

  Since the image data transfer units 45 and 47 operate asynchronously, the ring buffer change signal is enabled during the above-described processing, but the change line number LNc of the image data transfer unit 45 is 60 lines. Therefore, LNd <LNc, and the ring buffer is not changed.

  Then, at the stage shown in FIG. 7D, that is, when the processing for the second half of the second belt is completed, the input line number LNd = 60 and LNd <LNc is not satisfied (NO in step S36). Switching to use the buffer B (step S38). The start address of the ring buffer B is 0x00000000 because it is the start address of the input original image. Next, the image data transfer unit 47 receives the intermediate data from the ring buffer B, supplies the intermediate data to the pixel number conversion unit 46, and stores the image data from the pixel number conversion unit 46 in the image data storage area 51 after processing. Then, the input line number LNd is calculated (step S42).

  Specifically, as shown in FIG. 8A, the image data transfer unit 47 sets the next intermediate data (0x00000000) from the storage area 53 (0x00000000) of the ring buffer B set in the storage area 50 of the original image data. 15 lines: the first half of the third belt) is supplied to the pixel number conversion unit 46, and the enlarged image data (30 lines: the first half of the third belt) from the pixel number conversion unit 46 is converted into the processed image data. Store in the storage area 51. At this time, since the input belt has 15 lines, the number of input lines LNd = 75.

  Next, the image data transfer unit 47 also has the ring buffer change signal enabled and LNd> LNc in this case, so that the image data transfer unit 47 returns the original image data as shown in FIG. The next intermediate data (15th line: second half of the third belt) is supplied from the storage area 53 of the ring buffer B set in the storage area 50 to the pixel number conversion unit 46, and the enlargement process from the pixel number conversion unit 46 The processed image data (30 lines: the second half of the third belt) is stored in the storage area 51 of the processed image data. At this time, since the input belt has 15 lines, the number of input lines LNd = 90.

  Thereafter, in the same manner, the image data transfer unit 47 sets the next intermediate data (15 lines) from the storage area 53 of the ring buffer B set in the storage area 50 of the original image data, as shown in FIG. : The first half of the third belt) is supplied to the pixel number conversion unit 46, and the enlarged image data from the pixel number conversion unit 46 (30 lines: the first half of the fourth belt) is stored in the storage area for the processed image data. 51. At this time, since the input belt has 15 lines, the number of input lines LNd = 105.

  Further, as shown in FIG. 8D, the image data transfer unit 47 starts the next intermediate data (15th line: fourth belt) from the storage area 53 of the ring buffer B set in the storage area 50 of the original image data. ) Is supplied to the pixel number conversion unit 46, and the enlarged image data (30 lines: the second half of the fourth belt) from the pixel number conversion unit 46 is stored in the storage area 51 of the processed image data. . At this time, since the input belt has 15 lines, the number of input lines LNd = 120.

  Thereafter, the image data transfer unit 47 reads the intermediate data from the ring buffer B set in the storage area 50 of the original image data, supplies the intermediate data to the pixel number conversion unit 46, and receives the image data from the pixel number conversion unit 46. The process of storing the processed image data in the storage area 51 is repeated until the final belt. Then, when the process is completed for all (YES in step S44), the process is terminated.

  According to the above-described embodiment, the usable area that has been read in the storage area 50 of the original image data before use and the unused area in the storage area 51 of the processed image data according to the usage status of the RAM 32. Since the storage area used for the ring buffer is secured, a storage area dedicated to the ring buffer, which is necessary separately from the storage area for the input / output image data, becomes unnecessary, and the memory can be used effectively.

  In addition, according to the present embodiment, the storage area of the original image data from the ring buffer A secured in the storage area that is not stored in the storage area 51 of the processed image data at a predetermined timing based on the usage status of the RAM 32. Since the switching to the ring buffer B secured in the storage area in which 50 read image data has been stored is made, there is no need for a dedicated storage area for the ring buffer without hindering memory use by image processing, Memory can be used effectively.

  Further, according to the present embodiment, until the size of the empty area in the storage area 50 of the original image data is equal to or larger than the size of the work area required for the image processing, it is secured in the storage area of the processed image data 51. The storage area for the original image data is used at a predetermined timing after the size of the empty area in the storage area 50 for the original image data becomes equal to or larger than the size of the work area required for image processing. Since the switching to the ring buffer B secured to 50 is performed, a memory area dedicated to the ring buffer is not required and the memory can be used effectively without causing any trouble in using the memory by image processing.

  Further, according to the present embodiment, after the size of the free area in the storage area 50 of the original image data becomes equal to or larger than the size of the work area required for the image processing, it is secured in the storage area of the processed image data 51. The work area for performing image processing is changed to a free area in the storage area 50 of the original image data at a predetermined timing until the size of the ring buffer A becomes smaller than the size of the work area required for image processing. Since switching is performed, a memory area dedicated to the ring buffer is not required and the memory can be effectively used without causing any trouble in using the memory by image processing.

  In addition, according to the present embodiment, the work area for performing image processing at the timing when the size of the free area in the storage area 50 of the original image data is equal to or larger than the size of the work area required for image processing, Since switching to a free area in the storage area 50 of the original image data is performed, a memory area dedicated to the ring buffer is not required and the memory can be used effectively without causing any trouble in using the memory by image processing.

  Further, according to the present embodiment, the ring buffer A secured in the storage area 51 of the processed image data is used as a work area for temporarily storing intermediate data after the image processing by the image conversion unit 43 and the image correction unit 44. And the ring buffer B secured in the storage area 50 of the original image data are used, so that a memory area dedicated to the ring buffer is not required and the memory is effectively used without hindering the memory use by the image processing. Can do.

  In this embodiment, if there is a difference in processing speed due to the difference in processing contents of the image data transfer units 45 and 47, the processing of the image data transfer unit 45 does not wait, and intermediate data for the image data transfer unit 47 is lost. There is a possibility that the preparation of will be delayed. In the present embodiment, it is possible to absorb the difference in processing speed between the image data transfer units 45 and 47 by adjusting the capacities of the ring buffers A and B. However, if the difference in processing speed between the two cannot be absorbed for some reason, it is necessary to wait for processing by the image data transfer unit 47 (reading of data from the ring buffers A and B). In such a case, for example, every time the processing in units of belts by the image data transfer unit 45 ends (after being stored in the ring buffers A and B), the image data transfer unit 47 is notified of the end. It may be. When the image data transfer unit 47 receives notification from the image data transfer unit 45 that processing for a certain belt has been completed, the image data transfer unit 47 may perform processing for the belt for which processing has been completed. Alternatively, a management unit that manages the processing progress of the image data transfer units 45 and 47 may be provided separately, and the management unit may adjust the progress of processing by the image data transfer unit 45 and the image transfer unit 47. .

As mentioned above, although several embodiment of this invention was described, this invention is not limited to these, The invention described in the claim, and its equal range are included.
Below, the invention described in the claims of the present application is appended.

(Appendix 1)
The invention according to appendix 1 includes at least storage means having a first area for storing original image data and a second area for storing image data after image processing for the original image, and predetermined image processing. The image processing means to be applied and the original image data stored in the first area of the storage means are sequentially read in predetermined division units and supplied to the image processing means, and the image processed by the image processing means In the process of sequentially storing data in the second area of the storage means, the work area for the predetermined image processing by the image processing means is not stored in the processed image data in the second area. And image data transfer means for securing the area and the area where the read image data in the first area is stored based on the usage status of the storage means. It is a device.

(Appendix 2)
According to the second aspect of the present invention, the image data transfer means has processed a work area for performing the predetermined image processing in the second area at a predetermined timing based on a usage status of the storage means. The image processing apparatus according to appendix 1, wherein the image processing apparatus switches from an area in which the image data is not stored to an area in which the read image data in the first area is stored.

(Appendix 3)
According to the third aspect of the present invention, in the image data transfer means, the size of the area where the read image data in the first area is stored is larger than the size of the work area required for the predetermined image processing. In the meantime, a work area for performing the predetermined image processing is secured in an area in which the processed image data in the second area is not stored, and the work area has been read out in the first area. A work area for performing the predetermined image processing at a predetermined timing after the size of the area in which the image data is stored is equal to or larger than the size of the work area required for the predetermined image processing. The image processing apparatus according to appendix 2, wherein the image processing apparatus switches to an area in which read image data in one area is stored.

(Appendix 4)
In the invention according to attachment 4, the image data transfer means is configured such that the size of the area in which the read image data in the first area is stored is larger than the size of the work area required for the predetermined image processing. After that, at a predetermined timing until the size of the area in which the processed image data in the second area is not stored becomes smaller than the size of the work area required for the predetermined image processing, The image processing apparatus according to appendix 3, wherein a work area for performing the predetermined image processing is switched to an area in which the read image data in the first area is stored.

(Appendix 5)
In the invention according to attachment 5, the image data transfer means is configured such that the size of the area in which the read image data in the first area is stored is equal to or larger than the size of the work area required for the predetermined image processing. The supplementary note 4 is characterized in that the work area for performing the predetermined image processing is switched to the area in the first area in which the read image data has been stored at the timing when An image processing apparatus.

(Appendix 6)
The invention according to appendix 6, wherein the image processing means performs first image processing on the original image data stored in the first area in a predetermined division unit; A second image processing unit that performs second image processing on a predetermined division unit on the image data that has been subjected to the first image processing by the first image processing unit; The image processing apparatus according to any one of appendices 1 to 5, wherein the image processing apparatus is a work area for temporarily storing the first image processing result by the first image processing means.

(Appendix 7)
The invention according to appendix 7 includes a step of sequentially reading the original image data stored in the first area in the memory in a predetermined division unit, and the original image data sequentially read in the predetermined division unit. Performing a predetermined image processing; a work area for the predetermined image processing; an area in which the processed image data in the second area is not stored; and a read area in the first area. Securing the image data in the area where the image data is stored based on the use status of the memory, and the image data on which the predetermined image processing has been performed, the second area of the memory via the work area The image processing method is characterized in that it includes a step of sequentially storing data in the image processing method.

(Appendix 8)
The invention according to appendix 8 includes a function of sequentially reading original image data stored in a first area in a memory in a predetermined division unit into a computer, and to original image data sequentially read in the predetermined division unit. A function for performing predetermined image processing on the area, a work area for the predetermined image processing, an area in which the processed image data in the second area is not stored, and a read in the first area A function for securing the image data on the basis of the usage status of the memory and the image data subjected to the predetermined image processing are stored in the second area of the memory via the work area. It is a program characterized by causing a function to be stored sequentially.

30 CPU
31 ASIC
32 RAM
33 CCD
34 ROM
35 LCD
36 Keyboard 37 Power supply 40 DMAC
41 Memory control unit 42 CCD control unit 43 Image conversion unit 44 Image correction unit 45 Image data transfer unit 46 Pixel number conversion unit 47 Image data transfer unit 48 Keyboard control unit 49 LCD control unit 50 Original image data storage area 51 After processing Image data storage area 52 Ring buffer A storage area 53 Ring buffer B storage area

Claims (8)

Storage means having at least a first area for storing original image data and a second area for storing image data after image processing on the original image;
Image processing means for performing predetermined image processing;
The original image data stored in the first area of the storage means is sequentially read out in a predetermined division unit, supplied to the image processing means, and the image data processed by the image processing means is stored in the storage means. In the process of sequentially storing in the second area, a work area for predetermined image processing by the image processing means is defined as an area in which processed image data in the second area is not stored and the first area. Image data transfer means for securing the read image data in the area to the area where the read image data is stored, based on the use status of the storage means;
With
The image data transfer unit is configured to wait until the size of the area in which the read image data in the first area is stored is equal to or larger than the size of the work area required for the predetermined image processing. A work area for performing the image processing is secured in an area in which the processed image data in the second area is not stored, and the read image data in the first area is stored At a predetermined timing after the size of the area becomes equal to or larger than the size of the work area required for the predetermined image processing, a work area for performing the predetermined image processing is read in the first area. An image processing apparatus that switches to an area in which image data has been stored . The image data transfer means
After the size of the area in which the read image data in the first area is stored exceeds the size of the work area required for the predetermined image processing, the processed area in the second area is processed. The work area for performing the predetermined image processing at a predetermined timing until the size of the area in which image data is not stored becomes smaller than the size of the work area required for the predetermined image processing is the first area. The image processing apparatus according to claim 1 , wherein the image processing apparatus switches to an area in which read image data in one area is stored. The image data transfer means
For performing the predetermined image processing at a timing when the size of the area in which the read image data in the first area is stored is equal to or larger than the size of the work area required for the predetermined image processing. The image processing apparatus according to claim 2 , wherein the work area is switched to an area in which the read image data in the first area is stored. Storage means having at least a first area for storing original image data and a second area for storing image data after image processing on the original image;
Image processing means for performing predetermined image processing;
The original image data stored in the first area of the storage means is sequentially read out in a predetermined division unit, supplied to the image processing means, and the image data processed by the image processing means is stored in the storage means. In the process of sequentially storing in the second area, a work area for predetermined image processing by the image processing means is defined as an area in which processed image data in the second area is not stored and the first area. Image data transfer means for securing the read image data in the area to the area where the read image data is stored, based on the use status of the storage means;
With
Wherein the image processing means, said first have rows first image processing on the original image data stored in the region at a predetermined division unit, the image data after the first image processing is performed There line a second image processing at a predetermined division unit relative,
The working area is an image processing apparatus which is a working area for temporarily storing the results of the first image processing. Original image data stored in a first area in a memory having at least a first area for storing original image data and a second area for storing image data after image processing for the original image, Sequentially reading in predetermined division units;
Performing predetermined image processing on original image data sequentially read in the predetermined division unit;
The work area for the predetermined image processing includes an area in which the processed image data in the second area is not stored and an area in which the read image data in the first area is stored. And securing based on the usage status of the memory,
Sequentially storing the image data subjected to the predetermined image processing in the second area of the memory via the work area;
Including
The predetermined image processing is performed until the size of the area in which the read image data in the first area is stored is equal to or larger than the size of the work area required for the predetermined image processing. A work area is secured in an area in which the processed image data in the second area is not stored, and the size of the area in which the read image data in the first area is stored is the predetermined area. The read-out image data in the first area was stored as the work area for performing the predetermined image processing at a predetermined timing after the size of the work area required for the image processing became larger. An image processing method characterized by switching to a region . On the computer,
Original image data stored in a first area in a memory having at least a first area for storing original image data and a second area for storing image data after image processing for the original image, A function for sequentially reading in a predetermined division unit,
A function of performing predetermined image processing on the original image data sequentially read in the predetermined division unit;
The work area for the predetermined image processing includes an area in which the processed image data in the second area is not stored and an area in which the read image data in the first area is stored. A function to be secured based on the usage status of the memory,
A function of sequentially storing the image data that has undergone the predetermined image processing in the second area of the memory via the work area
And execute
The predetermined image processing is performed until the size of the area in which the read image data in the first area is stored is equal to or larger than the size of the work area required for the predetermined image processing. A work area is secured in an area in which the processed image data in the second area is not stored, and the size of the area in which the read image data in the first area is stored is the predetermined area. The read-out image data in the first area was stored as the work area for performing the predetermined image processing at a predetermined timing after the size of the work area required for the image processing became larger. A program characterized by switching to an area . A storage unit having at least a first region for storing original image data and a second region for storing image data after image processing on the original image; and an image processing unit for performing predetermined image processing. An image processing method executed by the image processing apparatus,
The original image data stored in the first area of the storage means is sequentially read out in a predetermined division unit, supplied to the image processing means, and the image data processed by the image processing means is stored in the storage means. In the process of sequentially storing in the second area, a work area for predetermined image processing by the image processing means is defined as an area in which processed image data in the second area is not stored and the first area. Including an image data transfer process that secures the read image data in the area based on the usage status of the storage means,
The image processing means performs first image processing on the original image data stored in the first area in a predetermined division unit, and applies the image data after the first image processing is performed. On the other hand, the second image processing is performed in a predetermined division unit,
The work area is a work area for temporarily storing the result of the first image processing.
An image processing method. A computer of an image processing apparatus having storage means having at least a first area for storing original image data and a second area for storing image data after image processing on the original image;
Image processing means for performing predetermined image processing;
The original image data stored in the first area of the storage means is sequentially read out in a predetermined division unit, supplied to the image processing means, and the image data processed by the image processing means is stored in the storage means. In the process of sequentially storing in the second area, a work area for predetermined image processing by the image processing means is defined as an area in which processed image data in the second area is not stored and the first area. Image data transfer means for securing the read image data in the area to the area where the stored image data is stored based on the use status of the storage means
Function as
The image processing means performs first image processing on the original image data stored in the first area in a predetermined division unit, and applies the image data after the first image processing is performed. On the other hand, the second image processing is performed in a predetermined division unit,
The work area is a work area for temporarily storing the result of the first image processing.
A program characterized by that.

Images