JP5534722B2 - Image processing apparatus and processing method thereof - Google Patents

Description

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

  Recording apparatuses that record information such as characters and images on a recording medium are known. As a recording method, for example, there is an ink jet recording method in which recording is performed using ink. In recent years, multi-function machines (hereinafter referred to as “ink-jet multi-function machines”) that employ such an ink jet recording method have become widespread. A feature of the inkjet multifunction device is that both copying to plain paper and direct printing to photographic paper can be realized by a single device.

  Now, with the improvement in recording speed, it is required to increase the image processing speed. Image processing is realized by hardware or software. Hardware scale and cost are in principle proportional to each other, and in order to improve functionality and reduce product price to increase customer satisfaction, it is necessary to improve its processing power while keeping hardware costs low. is there. Therefore, a hardware technique for flexibly changing the process according to the specification application has been proposed (see Patent Document 1).

  In general, processing such as character processing is performed in copying. The character processing is performed, for example, in order to suppress an adverse effect on a halftone dot portion due to edge enhancement of the character portion. Therefore, in this process, it is necessary to determine a character area and a halftone dot area and execute image processing in accordance with each area (see Patent Document 2).

  Further, when recording a photographic image (so-called photographic printing), a dodging process or the like is performed. In dodging, it is necessary to determine the subject and the other areas and execute processing according to each area in order to reduce the adverse effects on the background while making the brightness of the subject in the backlighting state appropriate (patent) Reference 3). Similarly, uniform processing is not performed on the entire image in red-eye correction processing for removing red eyes at the time of flash photography, noise correction processing for removing dark portion noise at the time of photography, or the like. That is, it is necessary to carry out correction processing for a specific problem occurrence place while suppressing the problem on a portion with good image quality (see Patent Documents 4 and 5).

JP 2006-178546 A JP 2002-77623 A Japanese Patent Laid-Open No. 9-18704 JP-A-11-355591 JP 2004-15322 A

  In the case of speeding up the above-described image processing (that is, image processing corresponding to an area) at the time of copying, photo printing, or the like, a dedicated circuit is generally provided for each. For this reason, for example, in an apparatus having a copy function and a photo printing function such as an inkjet multifunction peripheral, both circuits need to be mounted, and an increase in cost is inevitable.

The present invention has been made in view of the above-described problems, and provides a technique capable of performing various processes using a plurality of processing units that sequentially process processing data including image data representing an image. Objective.

In order to solve the above problems, an image processing apparatus according to one embodiment of the present invention inputs image data representing an image, or processing data including image data representing an image and attribute data representing processing content of the image data. And a plurality of processing means for sequentially processing the processing data input by the input means, wherein at least one of the plurality of processing means is a process input by the input means. Processing of image data included in the image data, processing for newly setting attribute data for processing means subsequent to the processing means using the result obtained by processing the image data, and processing the image data; And processing data can be output as processed data.

According to the present invention, various processes can be performed using a plurality of processing units that sequentially process processing data including image data representing an image for each of the plurality of processing units.

1 is a perspective view showing a schematic configuration of an image processing apparatus 100. FIG. FIG. 2 is a diagram showing an example of a control unit (not shown) of the image processing apparatus 100 shown in FIG. 1. The figure which shows an example of a structure of the image processing module 4 shown in FIG. The figure which shows an example of the format of the image data for a process concerning this embodiment. The figure which shows an example of a structure of a submodule. The figure which shows an example of the structure which provided the attribute value transmission wrapper layer. FIG. 2 is a schematic configuration diagram for explaining the first embodiment. 8 is a flowchart showing an example of the flow of image processing in the configuration shown in FIG. FIG. 5 is a schematic configuration diagram for explaining a second embodiment. FIG. 5 is a schematic configuration diagram for explaining a third embodiment. FIG. 6 is a schematic configuration diagram for explaining a fourth embodiment. FIG. 6 is a schematic configuration diagram for explaining a fifth embodiment.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an image processing apparatus and a processing method thereof according to the present invention will be described in detail with reference to the accompanying drawings. In the following description, an image processing apparatus adopting an ink jet recording method will be described as an example. However, the present invention is not limited to this. For example, an electrophotographic method using toner as a color material may be adopted.

  The image processing apparatus may be, for example, a single function printer having only a recording function, or may be, for example, a multi-function printer having a plurality of functions such as a recording function, a FAX function, and a scanner function. Good. In addition, for example, a manufacturing apparatus for manufacturing a color filter, an electronic device, an optical device, a minute structure, and the like by a predetermined recording method may be used.

  In the following description, “recording” is not limited to the case where significant information such as characters and figures is formed, and it does not matter whether it is significant. Further, it also represents a case where an image, a pattern, a pattern, a structure, or the like is widely formed on a recording medium or a medium is processed regardless of whether or not it is manifested so that a human can perceive it visually.

  “Recording medium” represents not only paper used in general recording apparatuses but also cloth, plastic film, metal plate, glass, ceramics, resin, wood, leather, and the like that can accept ink. .

  Further, “ink” should be interpreted widely as in the definition of “recording”. Therefore, by being applied on the recording medium, it is used for forming an image, pattern, pattern, etc., processing the recording medium, or processing the ink (for example, coagulation or insolubilization of the colorant in the ink applied to the recording medium). Represents a liquid that can be provided.

  FIG. 1 is a perspective view showing a schematic configuration of an image processing apparatus 100 according to an embodiment of the present invention.

  The image processing apparatus 100 includes, for example, a recording unit, a reading unit, and a control unit (not shown). For example, an ink jet serial printer is built in the recording unit, and, for example, a CCD type flat head scanner is built in the reading unit.

  The reading unit optically scans the recording medium and reads an image formed on the recording medium. The reading unit is generally provided at the top of the recording unit. The reading unit is provided with an upper lid 66 for pressing the document. The upper lid 66 is rotatable by a hinge (not shown). On the other hand, the recording unit is provided with a paper discharge unit 67 for discharging the recording medium after recording to the outside of the apparatus, and a paper feeding unit 68 on which recording media such as recording paper used for recording are stacked.

  As shown in FIG. 1, the recording unit includes, for example, a head carriage 51, a recording head 52, ink tanks 53 to 56, a shaft 57, a conveyance roller 59, a platen 50, a head carriage motor (not shown), and a conveyance motor (not shown). ) And the like.

  The head carriage 51 includes, for example, a recording head 52, an ink tank 53 containing black ink, an ink tank 54 containing cyan ink, an ink tank 55 containing magenta ink, and an ink tank 56 containing yellow ink. . The head carriage 51 moves in the arrow B direction along the shaft 57.

  The recording head 52 is configured by, for example, an inkjet method that ejects ink using thermal energy. The recording head 52 is provided with a recording element (hereinafter referred to as a heater) composed of a heating resistor element and a heater driving circuit. A heater is provided corresponding to each nozzle. In an ink jet recording system using a heating resistor element, a voltage is applied to the heating resistor element, and ink near the heating resistor element is boiled instantaneously. Then, ink droplets are ejected by an abrupt foaming pressure generated by a phase change of ink at the time of boiling. The ink ejection method is not limited to this, and any method such as a method using a piezo element, a method using an electrostatic element, or a method using a MEMS element may be adopted.

  In the recording operation, the image processing apparatus 100 draws the uppermost recording sheet 58 in the sheet feeding unit 68 into a U shape (arrow C) using the conveyance roller 59 and a drive motor. When the leading edge of the recording paper 58 reaches below the recording head 52, the head carriage 51 reciprocates in the direction of arrow B, and ink droplets are ejected from the recording head 52 onto the recording paper 58. Thereby, recording for one scan is performed. The image processing apparatus 100 transports the recording paper 58 by a length corresponding to one scan by the transport roller 59, and then scans the head carriage 51 again to perform recording. Recording is performed by repeating this operation. When the recording on the recording medium is completed, the image processing apparatus 100 discharges the recording paper 58 onto the paper discharge unit 67. Thereby, the recording operation ends.

  FIG. 2 is a diagram illustrating an example of a control unit (not shown) of the image processing apparatus 100 illustrated in FIG.

  The CPU 1 comprehensively controls processing in the image processing apparatus 100. The ROM 2 stores programs and the like. The RAM 7 is used as an image development area, a work area for program execution, and the like. The special application integrated circuit (ASIC) 30 generates a control signal for controlling the recording head 52, for example. The ASIC 30 is provided with m image processing modules 4-1 to 4 -m (where m is an integer equal to or greater than 1), and performs various types of image processing (hereinafter collectively referred to as the image processing module 4). ). Although details will be described later, each of the image processing modules is provided with m configuration registers 101-1 to 101-m (m is an integer of 1 or more). Called the registration register 101).

  The arbiter 5 arbitrates access to the RAM 7. The RAM interface 6 is an interface with the RAM 7. The head interface 8 is an interface with the recording head 52. In the present embodiment, the case where the RAM 7 is shared and used by the CPU 1 and each of the image processing modules 4-1 to 4 -m will be described as an example, but a RAM dedicated to the CPU 1 may be provided separately. Good.

  Here, the operation of the image processing apparatus 100 when an image to be processed is received will be described. When the image processing apparatus 100 receives an image at the I / O port 9, the image processing apparatus 100 stores the received image in the RAM 7 via the bus bridge 3, the arbiter 5, and the RAM interface 6. This process is executed by the CPU 1 following a program stored in the ROM 2, for example.

  When the image is stored, the image processing apparatus 100 sets parameters in the configuration register (hereinafter simply referred to as a register) 101-1 of the image processing module 4-1, in the CPU 1. As the parameter is set, the image processing module 4-1 starts operation. The register 101 holds parameters for controlling the operation of the image processing module. The register 101 holds, for example, an address range in the RAM 7 in which data is stored, an address range in the RAM 7 that is a storage destination of processed data, and the like.

  When a parameter is set in the register 101-1, the image processing module 4-1 reads data from the RAM 7 or writes data to the RAM 7 in accordance with the set parameter (in this case, an address range). The image processing module 4-1 performs processing according to this parameter. When the processing ends, the image processing module 4-1 generates an interrupt and notifies the CPU 1 to that effect.

  Receiving the interrupt notification, the CPU 1 analyzes the interrupt factor. Here, if the notification is sent along with the end of the reading process, the CPU 1 sets the reading address range of the next data in the register 101-1, and causes the image processing module 4-1 to continue the processing. If it is a notification sent along with the end of the writing process, the CPU 1 sets an address range as a storage destination of the processed data in the register 101-1, and processes the image processing module 4-1. To continue. At this time, the CPU 1 sets a parameter in the register 101-2 of the image processing module 4-2, and causes the image processing module 4-2 to start operation.

  As the parameters are set, the image processing module 4-2 starts processing in the same manner as the image processing module 4-1. Then, when the above-described reading process or writing process ends, the image processing module 4-2 generates an interrupt and notifies the CPU 1 to that effect. Thereafter, the same processing as described above is performed, and for example, the operation of the image processing module 4-3 is started. In this way, in the image processing apparatus 100, when the processing by a certain image processing module 4 is completed, the next image processing module 4 is activated, and the processing data is successively transferred to each image processing module 4. Thereby, a pipeline with the image processing module 4 as a unit is realized.

  In one image processing module 4, after the execution of the image processing instructed by the CPU 1 is started, the image processing module 4 ends the image processing and the output of the processed data becomes possible. The image processing module 4 can also perform processing in parallel. Specifically, data can be input from the RAM 7 to the other image processing module 4 and data processed by the other image processing module can be output to the RAM 7. Furthermore, since the image processing module 4 performs image processing independently of the other image processing modules 4, the plurality of image processing modules 4 can be operated in parallel, and a plurality of image processing can proceed simultaneously. it can.

  When the processing proceeds to the image processing module 4-m-1 as described above and bitmap data having a predetermined size or more is generated, the CPU 1 activates the recording unit. The image processing module 4-m sends bitmap data to the recording unit in accordance with the synchronization signal of the recording unit. Thereby, an image is recorded on the recording medium.

  FIG. 3 is a diagram showing an example of the internal configuration of the image processing module 4 shown in FIG.

  The image processing module 4 includes a read buffer 10, submodules 11-1 to 11 -n (referred to collectively as submodule 11), a write buffer 12, an arbiter 13, and a read address generation unit 14. It is comprised and comprises. Further, the image processing module 4 is provided with an interrupt control unit 15, a write address generation unit 16, a switch matrix 17, a register 101, a clock distribution unit 102, and a power supply unit 103.

  In the register (configuration register) 101, parameters related to operations of the switch matrix 17, the read address generation unit 14, the write address generation unit 16, and the like are set. The clock distribution unit 102 distributes the clock pulse clk of the apparatus to each unit in the image processing module 4. The clock distribution unit 102 can stop the supply of the clock clk to an arbitrary submodule 11 by setting the register 101. The power supply unit 103 controls power supply to each unit in the image processing module 4. The power supply unit 103 can stop power supply to an arbitrary submodule by setting the register 101. In the switch matrix 17, data is sequentially input to the plurality of submodules 11 in a predetermined order. Accordingly, a series of image processing to be performed on the data to be processed is performed using the plurality of submodules 11.

  The order in which the switch matrix 17 inputs the processing image data to the plurality of submodules 11 is determined in advance. This order can be fixedly determined, for example, as read buffer 10, submodule 1, submodule 2,..., Submodule n, and write buffer 12. As a result, the switch matrix 17 can play its role if the processing image data is moved between the submodules in an always determined order, so that the configuration of the switch matrix 17 can be simplified. Even if the input order of the processing image data between the sub-modules in the image processing module 4 is fixed, if the CPU 1 changes the order of the contents of the processing given to each sub-module, one image Various image processing can be realized by the processing module 4.

  Here, consider a case where different types of sub-modules 11 are mixed in one image processing module 4. Submodules 1 (11-1) to n-1 (11-n-1) are submodules having an (adaptive) image processing unit 413 shown in FIG. 5 to be described later, and submodule n (11-n). Is a (static) image processing unit 413 shown in FIG. In this case, there may be a disadvantage that the order in which the (static) image processing unit 413 processes the processing image data is fixed. However, this inconvenience can be solved by using a plurality of image processing modules 4 in combination. For example, in the first image processing module 4, the (adaptive) image processing unit 413 and the (static) image processing unit 413 perform the first half of image processing including the processing using the (static) image processing unit 413. carry out. Then, the processing image data that has been processed using the (static) image processing unit 413 is transferred to the second image processing module 4. In the second image processing module 4 that has received the processing image data, the (adaptive) image processing unit 413 performs the latter half of the image processing. In the second image processing module 4, the (adaptive) image processing unit 413 performs the latter half of the image processing and then directly uses the (static) image processing unit 413 without using the (static) image processing unit 413. Transfer data.

  On the other hand, even in the case where different types of sub-modules are mixed in one image processing module 4, it can be designed to change the processing order between the sub-modules. This is because the CPU 1 sets the order information instructing the order of processing between the sub-modules in the register 101, and the switch matrix 17 refers to the order information to control the input order of the processing image data. Can be realized. In this case, the configuration of the switch matrix 17 is complicated as compared with the case where the processing order between the submodules is fixed. However, in this case, since each of the plurality of image processing modules 4 can be configured to perform separate image processing, the image processing speed is higher than when one image processing is performed using the plurality of image processing modules 4. Improvement can be expected.

  As described above, the CPU 1 sets parameters in the register 101 of the image processing module 4. When a read start address and a read end address are set by this parameter setting, the read address generation unit 14 outputs a read enable signal Ren. At this time, the read address generation unit 14 refers to the parameter set in the register 101 and outputs the read address Rad and the read access unit Rn from the RAM 7. If Rn is a fixed value, Rn may be output to the register 101 without referring to the parameter. Further, when the write start address and the write end address are set, the write address generation unit 16 outputs the write enable signal Wen. At this time, the read address generation unit 14 refers to the parameter set in the register 101 and outputs the read address Wad and the read access unit Wn from the RAM 7. If Wn is a fixed value, Wn may be output to the register 101 without referring to the parameter.

  The arbiter 13 detects the free capacity Rp of the read buffer 10 and the read enable signal Ren from the read address generation unit 14. As a result of detection, it is assumed that the read address is valid (Ren = 1) and there is an empty area in the read buffer 10 (Rp ≧ Rn (Rn: read access unit from the RAM 7)). In this case, the arbiter 13 issues a read request (PREQ = 1, PNRW = 0, PNUM = Rn, PADD = Rad) to the arbiter 5.

  On the other hand, if the data storage amount Wp of the write buffer 12 is equal to or larger than a predetermined size (Wp ≧ Wn (Wn: write access unit to the RAM 7)), the arbiter 13 detects the write enable signal Wen of the write address generation unit 16. The arbiter 13 issues a write request to the arbiter 5. That is, if the write address is valid (Wen = 1) and the write buffer 12 can output data (Wp ≧ Wn), the arbiter 13 A write request (PREQ = 1, PNRW = 1, PNUM = Wnum, PADD = Wad) is issued to the arbiter 5.

  When receiving the request signal PREQ from the image processing module 4, the arbiter 5 determines whether it is a read request or a write request based on the signal PNRW (0: read, 1: write). The arbiter 5 detects a transfer size (number of transfer words) based on the signal PNUM and detects an address based on the signal PADD. If there is no other request (request from the CPU 1 or another image processing module 4), the arbiter 5 starts access to the corresponding data in the RAM 7, and receives the reception signal PACK as the request source image processing module 4. Return to. On the other hand, if there is another request, the arbiter 5 arbitrates the processing corresponding to each request, and executes the processing in order according to the priority order.

  When the arbiter 13 receives the reception signal PACK from the arbiter 5, it determines whether it is a read request or a write request. As a result of the determination, if it is a read request, the arbiter 13 returns a reception signal Rack to the request source read address generation unit 14. The read address generation unit 14 that has received the reception signal Rack generates an address in the RAM 7 as a data read destination and outputs it to the Rad. If the requested address (the address generated by the read address generation unit 14) is the read end address set in the register 101, the read address generation unit 14 resets the read enable signal Ren. Further, the read address generation unit 14 outputs a read end signal Rend to the interrupt control unit 15.

  On the other hand, when the reception signal PACK from the arbiter 5 is a write request, the arbiter 13 returns the reception signal Wack to the write address generation unit 16 that is the request source. Upon receiving the reception signal Wack, the write address generation unit 16 generates an address in the RAM 7 that is a data writing destination. If the requested address is the write end address set in the register 101, the write address generation unit 16 resets the write enable signal Wen and outputs the write end signal Wend to the interrupt control unit 15. To do.

  Here, the interrupt control unit 15 sets a read end interrupt mask and a write end interrupt mask based on the parameters set in the register 101. If the setting of each interrupt mask is enabled, the interrupt control unit 15 generates an interrupt signal INT according to the read end signal Rend or the write end signal Wend, and notifies the CPU 1 of the generated signal.

  Receiving the interrupt signal INT, the CPU 1 reads the status of the interrupt control unit 15 and determines the interrupt factor. As a result of the determination, if the interrupt factor is the end of reading, the CPU 1 resets the read end interrupt mask and cancels the interrupt. In the case of continuing the process, the CPU 1 resets the read start address and the read end address in the register 101, and accordingly, the read address generation unit 14 outputs the read enable signal Ren. Thereafter, the interrupt control unit 15 sets a read end interrupt mask.

  On the other hand, as a result of reading the status of the interrupt control unit 15, if the interrupt factor is the write end, the CPU 1 resets the write end interrupt mask and releases the interrupt. When the processing is continued, the CPU 1 resets the write start address and the write end address in the register 101, and the write address generation unit 16 outputs a write enable signal Wen and the like accordingly. Thereafter, the interrupt control unit 15 sets a write end interrupt mask.

  Here, when data is read from the RAM 7, the arbiter 5 returns a RAM data valid signal PVALID to the requesting image processing module 4. In the request source image processing module 4, the arbiter 13 returns a data valid signal Rvalid to the read buffer 10. The read buffer 10 stores the data included in the RAM data output signal PDIN while the data valid signal Rvalid is being output. As a result, data read from the RAM 7 based on the address indicated by the signal PADD (= Rad) (the address generated by the read address generation unit 14) is stored in the read buffer 10.

  On the other hand, when data is written to the RAM 7, the arbiter 5 returns a RAM data valid signal PVALID to the image processing module 4 as a request source in accordance with the write timing of the RAM 7. In the request source image processing module 4, the arbiter 13 returns a data valid signal Wvalid to the write buffer 12. The write buffer 12 outputs write data using the RAM data input signal PDOUT while the data valid signal Wvalid is output. As a result, the data in the write buffer 12 is stored at the address in the RAM 7 (the address generated by the write address generator 16) indicated by the signal PADD (= Wad).

  The read buffer 10 outputs the valid signal valid_out when all the data necessary for the processing of the submodule 11 is stored, and resets the valid signal valid_out otherwise. Further, when the holding request signal stall_in from the submodule 11 is not input, the read buffer 10 sequentially outputs the stored data to data_out in synchronization with the clock. If the holding request signal stall_in from the submodule 11 is input, the data output to data_out is not updated. Here, the holding request signal stall_out output from the submodule 11 is input to the holding request signal stall_in of the read buffer 10. On the other hand, the submodule 11 outputs the valid signal valid_out when the processed data can be output to the write buffer 12, and resets the valid signal valid_out otherwise. Also, if the holding request signal stall_in is not input from the read buffer 10, the submodule 11 sequentially outputs the stored data to data_out in synchronization with the clock. If the holding request signal stall_in is input from the read buffer 10, the data output to data_out is not updated. Here, the holding request signal stall_out output from the write buffer 12 is input to the holding request signal stall_in of the submodule 11. Data transmission / reception between the submodules 11 is also performed in the same manner as the procedure between the submodule 11, the read buffer 10, and the write buffer 12 described above.

  The submodule 11 receives data from the read buffer 10 or the other submodule 11 via data_in while the valid signal valid_out is output from the read buffer 10 or the other submodule 11. While outputting the valid signal valid_out, the submodule 11 outputs data to the write buffer or another submodule 11 via data_in. If the data cannot be received, the submodule 11 outputs a holding request signal stall_out. As a result, the holding request signal stall_in of the read buffer 10 or other submodule 11 that is the data transmission source is set, and the output of the read buffer 10 is held. When the write buffer 12 or other submodule 11 cannot receive data, the write buffer 12 or other submodule 11 outputs a holding request signal stall_out. As a result, the hold request signal stall_in of the submodule 11 that is the data transmission source is set, and the output of the submodule 11 is held. The order of data input / output among the read buffer 10, submodule 11, and write buffer 12 is managed by the switch matrix according to the setting of the register 101. Then, according to the order, data_in and data_out, valid_in and valid_out, stall_in and stall_out are connected, and data is input and output. If it is not necessary to rearrange the input data or output data, the read buffer 10 and the write buffer 12 may be configured by, for example, a FIFO.

  As described above, the inside of the image processing module 4 is configured by one or more submodules 11. Data is exchanged between the sub-modules 11 by the above-described operation (handshaking based on the valid signal valid and the hold request signal stall (_in of a certain module is connected to _out of another module)). These submodules 11 are configured by paying attention to hardware resources necessary for various image processing. As a result, the connection relationship, processing sequence, calculation accuracy, and the like of each submodule can be changed as appropriate, so that a compact and flexible image processing apparatus that can handle various processing modes can be constructed. Further, since the clock to the unused submodule 11 can be stopped and the power supply to the unused submodule 11 can be stopped, not only the processing speed and processing accuracy but also power consumption can be suppressed.

  If the image processing module 4 is configured to be compatible with various formats, division processing (multi-sequence processing) can be realized. In this case, even a process that is not supported by the above-described hardware resource can be configured to be executed by another module, a CPU, or the like. Basically, it is desirable to unify the formats of input data and output data among the read buffer 10, each submodule 11, and the write buffer 12. When the format is unified, the input / output data format of data among the read buffer 10, each submodule 11, and the write buffer 12 can be changed even if the processing order of submodules is changed or a specific submodule is not used. No change will occur.

  Here, FIG. 4 is a diagram showing an example of an input / output data format according to the present embodiment. This format (hereinafter referred to as processing image data) is shared and used, for example, between the submodule 11 and the switch matrix 17 described above.

  The processing image data 301 has a format having no attribute value. This format is the same as the conventional format. This processing image data 301 is composed of R (red), G (green), and B (blue) pixel value fields (3011 to 3013) each having a 32-bit pixel value, and is 96 bits in total.

  The processing image data 302 has a format having three 8-bit attribute value fields (3024 to 3026) and three 24-bit pixel value fields (3021 to 3023). Here, in the present embodiment, various setting values (attribute values) are set in the attribute value field. The attribute value is not fixed in type or role, and a free value is set. For example, it is possible to set a value indicating what processing is performed in which image processing sub-module, and a value indicating what processing is performed with reference to which of the attribute value fields. .

  The processing image data 303 has a format having six 8-bit attribute value fields (3034 to 3039) and three 16-bit pixel value fields (3031 to 3033). The processing image data 304 has a format having three 16-bit attribute value fields (3044 to 3046) and three 16-bit pixel value fields (3041 to 3043). The processing image data 305 has two 16-bit attribute value fields (3054 and 3055), two 8-bit attribute value fields (3056 and 3057), and three 16-bit pixel value fields (3051 to 3053). It is.

  In the present embodiment, the processing image data 301 to 305 are all configured in a 96-bit format. Therefore, if the dynamic range of the attribute value is increased (the number of bits in the attribute value field is increased), the dynamic range of the pixel value is decreased (the number of bits in the pixel value field is reduced), and the attribute value and the pixel value are Has a trade-off relationship.

  Based on this, it can be said that the configuration of the processing image data 301 is a format that can maintain the highest accuracy of the pixel value. The configuration of the processing image data 302 can be said to be a format in which attribute values can be used while maintaining high accuracy of pixel values. The configuration of the processing image data 303 includes the number of attribute values held by each pixel. It can be said that this is a form further increased. It can be said that the configuration of the processing image data 304 has a format in which the attribute value has the same dynamic range as the pixel value, and realizes control with a more accurate attribute value. The configuration of the processing image data 305 can be said to have a format that has attribute values of two types of sizes and can be used by combining large and small attribute values as necessary.

  By using the processing image data as described above, different image processing can be performed in various use cases. Specifically, by using the attribute value, different image processing can be performed using the same resource. Thereby, flexibility of image processing and reduction of circuit size can be achieved.

  The above is the description of the format of the processing image data. However, the format configuration described above is merely an example, and the present invention is not limited to this. For example, in the above description, the case where each pixel value is 32 bits and the overall size is 96 bits has been described as an example, but the size may be other than this. For example, each pixel value may be 16 bits and the overall size may be 48 bits. The number of bits of each pixel value does not need to be the same, and each pixel value may have a non-uniform size, for example, R: 10 bits, G: 12 bits, B: 10 bits. Note that the format (processing image data) shown in FIG. 4 is created by the CPU 1 (which may be the image processing module 4), for example. That is, in the CPU 1, image data representing an image corresponds to each pixel of the image, and includes a pixel value field that holds the pixel value of each pixel and an attribute value field that holds an attribute value. A function for editing image data is provided.

  Next, the configuration of a submodule that performs image processing (hereinafter referred to as an image processing submodule) will be described with reference to FIG. The image processing submodule is realized as the submodule 11 shown in FIG. In addition, the same code | symbol is attached | subjected to the same thing as the structure shown in FIG. 3 mentioned above.

  The input data processing method setting unit 401 refers to the parameter set in the register 101 and sets a processing method for the input attribute value and the input pixel value. Note that the processing image data including the input attribute value and the input pixel value is input by a Data_in control signal from the switch matrix 17.

  The output data processing method setting unit 402 refers to the parameter set in the register 101 and sets a processing method for the output attribute value and the output pixel value. Note that the processing image data including the output attribute value and the output pixel value is output to the switch matrix 17 by the Data_out control signal. The CPU 1 sets the processing values for the input processing image data and the output processing image data by setting the setting values related to the read buffer 10, each submodule 11-n, and the write buffer 12 in the configuration register 101. Done. The setting method is not particularly limited. For example, the signal lines from each of the image processing modules 4 shown in FIG. Then, the CPU 1 and each register 101 of the image processing module 4 may be connected and set directly from the CPU 1 to the configuration register 101. Alternatively, the CPU 1 and the switch matrix 17 of each of the image processing modules 4 may be connected, and the setting may be performed via the switch matrix 17 by executing a register setting command. It should be noted that the method of setting all the various settings described below is not particularly limited.

  The input attribute value processing unit 403 edits the input attribute value included in the input processing image data. The Bit expander 404 extends the input pixel value included in the input processing image data to a format suitable for the processing of the image processing unit 413. The processing in the input attribute value processing unit 403 and the bit expander 404 is performed based on setting information (first setting information) set by the input data processing method setting unit 401.

  The first attribute value management unit 405 causes the delay unit 406 to hold the input attribute value processed by the input attribute value processing unit 403. When outputting the output attribute value, the output pixel value is output to the output attribute value processing unit 407 at a predetermined timing. This output timing needs to match the electrical configuration of the image processing unit 413.

  For example, assume that the image processing unit 413 performs the following processing.

Input pixels: 3 consecutive input pixels (pixel (X-1), pixel (X), pixel (X + 1))
Output pixel: Position of pixel X Output value: Sum of 3 pixels Pixel value (X-1) + pixel value (X) + pixel value (X + 1)
Attribute value: No particular processing, output attribute value (X) = input attribute value (X)
In this case, the output attribute value (X) is used to calculate the output value, but the final output value is determined at the timing when the pixel (X + 1) is input. Therefore, in the first attribute value management unit 405, the attribute value (X) already held in the delay unit 406 at the timing when the output pixel value (X) is sent from the image processing unit 413 to the output attribute value processing unit 407. Is output to the output attribute value processing unit 407. That is, the attribute value (X) that is the attribute value of the output pixel value (X) is delayed until the calculation of the output pixel value (X) using the pixel (X + 1) input after the pixel (X) is completed. Hold at 406. Thereafter, similarly to this, the value that becomes the attribute value of the output pixel value (X) is held in the delay unit 406 until the output timing of the output pixel value (X) comes.

As described above, the delay unit 406 holds attribute values (input attribute value, output attribute value). The delay unit 406 is provided for delaying input / output of attribute values. The processing by the delay unit 406 will be described with a specific example.
For example, assume that the image processing unit 413 performs the following processing.
Input pixels: 3 consecutive input pixels (pixel (X-1), pixel (X), pixel (X + 1))
Output pixel: Position of pixel X Output value: Divide the sum of three pixels by the attribute value (pixel value (X-1) + pixel value (X) + pixel value (X + 1)) / attribute value (X)
Attribute value: No particular processing, output attribute value (X) = input attribute value (X)
In this case, the image processing unit 413 needs the input attribute value (X) at the timing when the input pixel value (X + 1) is input. Therefore, the delay unit 406 delays input of the input attribute value (X) to the image processing unit 413.

  The output attribute value processing unit 407 edits the output attribute value output from the image processing unit 413. The Bit compressor 408 performs Bit compression on the output pixel value included in the output processing image data into an output format. The processing in the output attribute value processing unit 407 and the Bit compressor 408 is performed based on setting information (second setting information) set by the output data processing method setting unit 402.

  The input attribute value information setting unit 409 refers to the parameter set in the register 101 and defines which attribute value among the input attribute values held in the delay unit 406 is used during image processing by the image processing unit 413. Setting information (third setting information) to be set. That is, information indicating which of the pixel values of the plurality of pixel value fields and the attribute values of the plurality of attribute value fields is associated with and input to the image processing unit 413 is set.

  The output attribute value information setting unit 410 refers to the parameter set in the register 101 and sets setting information (fourth setting) that defines which attribute value field the output attribute value held in the delay unit 406 is set to. Information). That is, information indicating which of the plurality of attribute value fields is associated with each of the plurality of pixel value fields processed by the image processing unit 413 is set. In other words, the output attribute value information setting unit 410 sets the attribute value field of the output attribute value processed by the image processing unit 413.

  The second attribute value management unit 411 manages input / output of attribute values in the image processing unit 413. This management is performed based on the setting information set by the input attribute value information setting unit 409 and the output attribute value information setting unit 410. For example, the second attribute value management unit 411 extracts the input attribute value from the delay unit 406 and outputs it to the image processing unit 413. Also, the second attribute value management unit 411 causes the delay unit 406 to hold the output attribute value output from the image processing unit 413, for example.

  The parameter setting unit 412 refers to the parameters set in the register 101 and sets processing parameters specific to image processing. The image processing unit 413 performs image processing corresponding to the attribute value for each pixel (adaptive image processing). Specifically, predetermined image processing is performed using input pixel values, input attribute values, parameters, and the like input from the Bit expander 404, the second attribute value management unit 411, the parameter setting unit 412, and the like. Also, the image processing unit 413 outputs the output pixel value to the Bit compressor 408 as necessary, and outputs the output attribute value to the second attribute value management unit 411 as necessary. Note that the image processing unit 413 according to the present embodiment processes each pixel value (R / G / B) in 32 bits for both input and output. Therefore, when the attribute value is used, the bit expander 404 performs expansion from “16 bits” or “24 bits” to “32 bits”. Conversely, in the bit compressor 408, compression from “32 bits” to “16 bits” or “24 bits” is performed. That is, the number of bits is matched (in this case, the ratio between the number of bits in the attribute value field and the number of bits in the pixel value field is changed). Note that such a bit relationship is not necessarily required, and the number of image processing bits in the image processing unit 413 may be smaller than the number of bits of pixel values extracted by the input attribute value processing unit 403. In this case, for example, the positions of the bit expander 404 and the bit compressor 408 shown in FIG. Also, as a method of bit expansion and compression, bit shift may be used, or multiplication and division may be performed.

  As described above, in this embodiment, the Bit expander 404 and the Bit compressor 408 are provided, and even if the number of bits of the pixel value field (that is, the pixel) held in the processing image data is different, the image At the input / output stage of the processing unit 413, the number of bits is made constant. Thereby, image processing using the same resource (image processing unit) can be realized regardless of the number of bits of the attribute value and the number of bits of the pixel value.

  The configuration of the image processing submodule has been described above. However, in the image processing submodule, all the above-described configurations are not necessarily essential. For example, when one pixel is input, predetermined image processing is performed and one pixel is output, the delay unit 406 is not necessary. At this time, the first attribute value management unit 405 and the second attribute value management unit 411 may be configured integrally. Further, in the above-described image processing submodule, setting of attribute values is also changed, but a module for performing image processing using only attribute values may be configured. In this case, the output attribute value information setting unit 410 is not necessary. Similarly, when a module that does not use the attribute value for image processing and only changes the setting of the attribute value is configured, the input attribute value information setting unit 409 becomes unnecessary.

  Here, in the configuration shown in FIG. 3, a plurality of submodules are arranged. If sub-modules that do not support attribute values (hereinafter referred to as sub-modules that do not support attribute values) are mixed, the attribute values may be handled as pixel values in that module. In this case, a correct output result cannot be obtained. A configuration for dealing with this will be described with reference to FIG.

  An attribute value transmission wrapper layer 110 is provided between the attribute value non-corresponding submodule 11 and the switch matrix 17. The attribute value transmission wrapper layer 110 is located in an intermediate layer between the switch matrix 17 and the attribute value non-corresponding submodule 11 (that is, functions as an intermediate processing module) and plays a role of transmitting attribute values. Thereby, the attribute value non-corresponding submodule 11 can behave as if it is an attribute value corresponding submodule. In addition, since each component shown to the code | symbol 401-408 of the attribute value transmission wrapper layer 110 fulfill | performs the function equivalent to each component which demonstrated FIG. 5, the description is abbreviate | omitted.

  The parameter setting unit 412 refers to the parameters set in the register 101 and sets processing parameters specific to image processing. The image processing unit 413 performs uniform image processing for each pixel (static image processing). Specifically, predetermined image processing is performed using input pixel values, parameters, and the like input from the bit expander 404, the parameter setting unit 412, and the like. Further, the image processing unit 413 outputs the output pixel value to the Bit compressor 408 as necessary. Of the control signals between the switch matrix 17 and the attribute value non-corresponding submodule, Valid_in, Stall_in, Stall_out, and Valid_out are directly connected to each other without using the attribute value transmission wrapper layer 110. Also good.

  Data_in from the attribute value transmission wrapper layer 110 to the attribute value incompatible sub-module 11 includes only the input pixel value. This is because the attribute value is removed in the input attribute value processing unit 403 and the bit expander 404 of the attribute value transmission wrapper layer 110. Further, Data_out from the attribute value non-corresponding submodule 11 to the attribute value transmission wrapper layer 110 includes only the output pixel value.

  With the above configuration, it is possible to realize a mechanism for transmitting attribute values without changing the image processing unit 413. Here, if the image processing unit 413 performs 32-bit calculation processing, the bit width of the pixel value is reduced by the attribute value, but the accuracy of the image processing itself is not changed. Further, by providing the attribute value transmission wrapper layer 110, it is possible to configure the image processing module 4 in which the attribute value compatible submodule and the attribute value non-compatible submodule are mixed. Thus, using a common format (that is, processing image data), both high-precision image processing with a large number of bits (static) and adaptive image processing with a relatively small number of bits. Can be realized.

Next, setting information set by the input data processing method setting unit 401 will be described using “Table 1”. That is, the input attribute value processing unit 403 and the bit expander 404 execute various processes based on the setting information of “Table 1”. Note that the setting values shown in “Table 1” are merely examples, and the present invention is not limited thereto.
(Table 1)

[Get_Mode10]
This setting is suitable when the processing image data has the format 301 shown in FIG. In this case, the processing image data (that is, 96 bits) are all configured by a pixel value field (pixel value field 32 bits × 3). In this setting, the input attribute value processing unit 403 sets all 0 (standard value) in the attribute value field. In this case, since the pixel value is 32 bits, the Bit expander 404 does not operate.

[Get_Mode11]
This setting is suitable when the processing image data has the format 302 shown in FIG. In this case, the processing image data (ie, 96 bits) is composed of a pixel value field of 24 bits × 3 and an attribute value field of 8 bits × 3. In this setting, the input attribute value processing unit 403 sets (overwrites) all 0 (standard value) in the attribute value field. As a result, the image processing unit 413 performs static image processing instead of adaptive image processing. In this case, since the pixel value is 24 bits, the Bit expander 404 performs Bit expansion from “24 bits” to “32 bits”. The setting of “Get_Mode 11” is used in, for example, a module that generates processing contents of other submodules as attribute values and plays a role of setting them. For example, a sub-module that executes “Get_Mode 11” performs a process of extracting an image feature (determination of whether or not it is an edge portion or the like) and outputs it together with the original pixel value. In this case, in the submodule that executes “Get_Mode 11”, the image processing unit 413 extracts image features based on the pixel value in the pixel value field. Then, the determination result is generated as an attribute value, and the generated attribute value is set in the attribute value field. In addition, a pixel value that has not been processed or corrected is set in the pixel value field. Thereafter, the output attribute value processing unit 407 outputs processing image data. This processing image data is transferred to the subsequent submodule via the switch matrix 17. Then, in the subsequent submodule, using the attribute value set in the attribute value field of the received processing image data, the processing content for the pixel value is determined and the image processing is adaptively executed. In addition, in the setting of “Get_Mode11”, the input attribute value processing unit 403 sets 0 in the attribute value field, so that the propagation of the attribute value can be stopped. If only the propagation of the attribute value is to be stopped, the processing image data in which 0 is set in the attribute value field may be output. Therefore, in the output side of the attribute value, that is, in the output attribute value processing unit 407, This process (0 padding) may be performed. Further, when the propagation of the attribute value is stopped and the newly generated attribute value is set in the attribute value field, the input attribute value processing unit 403 does not have to set the attribute value to 0. Specifically, the image processing unit 413 performs image processing without using the attribute value, and the newly generated attribute value is sent to the attribute value output side, that is, the output attribute value processing unit 407 in the attribute value field. Set to. Then, the output attribute value processing unit 407 outputs processing image data. When “Get_Mode 11” is set, the process is closed by the sub module alone, so that the possibility of human error is reduced.

[Get_Mode12]
This setting is suitable when the processing image data has the format 302 shown in FIG. In this setting, the input attribute value processing unit 403 sets the attribute value included in the input processing image data as it is. The setting of “Get_Mode12” is used when adaptive image processing is performed using attribute values, for example.

[Get_Mode13]
This setting is substantially the same operation as “Get_Mode 11” described above. This is suitable when the processing image data has the formats 303 to 305 shown in FIG. The difference from “Get_Mode 11” is the number of bits of the attribute value and the pixel value. When “Get_Mode 13” is set, the degree of bit expansion by the bit expander 404 is changed from “16 bits” to “32 bits”.

[Get_Mode14]
This setting is substantially the same operation as the above-described Get_Mode 12. This is suitable when the processing image data has the formats 303 to 305 shown in FIG. The difference from “Get_Mode12” is the number of bits of the attribute value and the pixel value. When “Get_Mode 14” is set, the degree of bit expansion by the bit expander 404 is changed from “16 bits” to “32 bits”.

Next, setting information set by the output data processing method setting unit 402 will be described using “Table 2”. That is, the output attribute value processing unit 407 and the Bit compressor 408 execute various processes based on the setting information of “Table 2”. The set values shown in “Table 2” are merely examples, and the present invention is not limited to these.
(Table 2)

[Put_Mode20]
This setting is suitable when the processing image data to be output is in the format 301 shown in FIG. In this case, all the processing image data (that is, 96 bits) is composed of a pixel value field (pixel value field 32 bits × 3). With this setting, the output attribute value processing unit 407 does not set an attribute value. In addition, since the pixel value is 32 bits, the Bit compressor 408 does not operate.

[Put_Mode21]
This setting is suitable when the processing image data to be output is in the format 302 shown in FIG. In this case, the processing image data (ie, 96 bits) is composed of a pixel value field of 24 bits × 3 and an attribute value field of 8 bits × 3. In this setting, the output attribute value processing unit 407 sets (overwrites) all 0 (standard value) in the attribute value field. As a result, the processing image data in which the attribute value is invalid is output toward the other submodule. Also, since the pixel value is 32 bits, the Bit compressor 408 performs Bit compression from “32 bits” to “24 bits”. This “Put_Mode 21” is set, for example, by itself (the submodule) performing adaptive image processing using attribute values, but in other submodules, static image processing that does not use attribute values is performed. Used for implementation. In the setting of “Put_Mode 21”, since the attribute value field is set to 0, the propagation of the attribute value can be stopped. Note that this processing (0 padding) may be performed on the attribute value input side, that is, the input attribute value processing unit 403. When “Put_Mode 21” is set, it is possible to reduce the probability of adverse effects occurring in other attribute value incompatible submodules.

[Put_Mode22]
This setting is substantially the same operation as “Put_Mode 21” described above. This is suitable when the processing image data has the format 302 shown in FIG. In this setting, the output attribute value processing unit 407 sets the attribute value using the attribute value output from the image processing unit 413 as it is. Also, since the pixel value is 32 bits, the Bit compressor 408 performs Bit compression from “32 bits” to “24 bits”. The setting of “Put_Mode 22” is used in, for example, a module that generates processing contents of other submodules as attribute values and plays a role of setting them.

[Put_Mode23]
This setting is substantially the same operation as “Put_Mode 21” described above. This is suitable when the processing image data has the formats 303 to 305 shown in FIG. The difference from “Put_Mode 21” is the number of bits of the attribute value and the pixel value. When “Put_Mode 23” is set, the bit compression degree by the bit compressor 408 is changed from “32 bits” to “16 bits”.

[Put_Mode24]
This setting is substantially the same operation as “Put_Mode 22” described above. This setting is suitable when the processing image data has the formats 303 to 305 shown in FIG. The difference from “Put_Mode 22” is the number of bits of the attribute value and the pixel value. When “Put_Mode 24” is set, the bit compression degree by the bit compressor 408 is changed from “32 bits” to “16 bits”.

  As described above, by using the setting information shown in “Table 2”, it is possible to control the use / non-use of attribute values in the attribute value corresponding submodule. Thereby, various image processing can be realized using the same resource (image processing unit). In addition, since it is possible to freely design which field in the processing image data is used for the attribute value or the pixel value, the submodule can be used for various purposes while realizing the flexibility of the image processing design.

Next, setting information set by the input attribute value information setting unit 409 will be described using “Table 3”. The second attribute value management unit 411 executes various processes based on the setting information in “Table 3” before starting the image processing. Here, setting information when using the format 302 (three attribute value fields) shown in FIG. 4 will be described as an example.
(Table 3)

[In_Mode30]
With this setting, attribute values are not used for image processing. That is, the attribute value is not input to the image processing unit 413 in the second attribute value management unit 411. In this case, the submodule operates as an attribute value non-corresponding submodule. This setting is suitable, for example, when the image processing unit 413 performs static image processing. There are various methods for causing the image processing unit 413 to perform static image processing. For example, all the attribute values transmitted to the image processing unit 413 may be set to 0 (standard value). Even when “Get_Mode11” shown in “Table 1” is set, the attribute value field can be set to 0. However, when “In_Mode30” is set, the attribute value itself received from another submodule is changed to 0. Because of this, attribute value propagation continues.

[In_Mode31]
This setting is suitable when adaptive image processing is performed on all the pixel value fields (3 channels) using only the attribute value field A0 among the plurality of attribute value fields (A0 to A2). When “In_Mode 31” is set, the image processing is performed without using the other attribute value fields (A1, A2). When “In_Mode 31” is set, adaptive image processing can be performed on all channels (RGB) of the pixel value field using the same attribute value. Here, A0 is the attribute value field 3024 of the format 302 shown in FIG. 4, A1 is the attribute value field 3025 of the format 302 shown in FIG. 4, and A2 is the attribute value field of the format 302 shown in FIG. 3026. In the following first to fifth embodiments, A0 to A2 indicate attribute value fields 3024 to 3026.

[In_Mode32]
This setting is suitable when adaptive image processing is performed for all the pixel value fields (3 channels) using only the attribute value field A1 among the plurality of attribute value fields (A0 to A2). When “In_Mode 32” is set, the image processing is performed without using other attribute value fields (A0, A2). When “In_Mode 32” is set, adaptive image processing can be performed on all channels (RGB) of the pixel value field using the same attribute value.

[In_Mode33]
This setting is suitable when adaptive image processing is performed for all the pixel value fields (3 channels) using only the attribute value field A2 among the plurality of attribute value fields (A0 to A2). When “In_Mode 33” is set, the image processing is performed without using other attribute value fields (A0, A1). When “In_Mode 33” is set, adaptive image processing can be performed on all channels (RGB) of the pixel value field using the same attribute value.

[In_Mode34]
This setting is suitable when adaptive image processing is performed using the attribute value field A0 as the pixel value R, the attribute value field A1 as the pixel value G, and the attribute value field A2 as the pixel value B.

  As described above, by performing processing using “In_Mode 30 to 34” selectively, adaptive image processing can be performed using at least one piece of information in a plurality of attribute value fields. The set values shown in “Table 3” are merely examples, and the present invention is not limited to these. For example, when the pixel value includes brightness (Y) information and color (Cb / Cr) information such as Y, Cb, and Cr instead of R, G, and B, the attribute value field A0 regarding brightness is included. For color, adaptive image processing may be performed using the attribute value field A1. In this case, for example, an attribute value used by another submodule may be set in the attribute value field A2.

Next, setting information set by the output attribute value information setting unit 410 will be described using “Table 4”. The second attribute value management unit 411 executes various processes based on the setting information in “Table 4” after the image processing is completed. Here, setting information when using the format 303 (six attribute value fields) shown in FIG. 4 will be described as an example.
(Table 4)

[Out_Mode 40]
This setting is suitable when the attribute value is not set by itself (the submodule). That is, the output attribute value remains as the input attribute value, and both are the same value.

[Out_Mode41]
This setting is suitable when only the attribute value field A0 is newly set among the plurality of attribute value fields (A0 to A2). When “Out_Mode 41” is set, the other attribute value fields (A1, A2) are not updated.

[Out_Mode42]
This setting is suitable when only the attribute value field A1 is newly set out of the plurality of attribute value fields (A0 to A2). When “Out_Mode 42” is set, the other attribute value fields (A0, A2) are not updated.

[Out_Mode43]
This setting is suitable when only the attribute value field A2 is newly set among the plurality of attribute value fields (A0 to A2). When “Out_Mode 43” is set, the other attribute value fields (A0, A1) are not updated.

[Out_Mode44]
This setting is suitable when all of the plurality of attribute value fields A0 to A2 are newly set.

  As described above, by selectively using “Out_Mode 40 to 44”, it is possible to change the setting of at least one piece of information in the plurality of attribute value fields. As a result, adaptive image processing can be performed in other submodules. The set values shown in “Table 4” are merely examples, and the present invention is not limited to these. For example, when the pixel value includes brightness (Y) information and color (Cb / Cr) information such as Y, Cb, and Cr instead of R, G, and B, the attribute value field A0 regarding brightness is included. As for colors, different attribute values may be set for the attribute value field A1. In this case, for example, an attribute value used by another submodule may be set in the attribute value field A2.

  Next, with reference to the first to seventh embodiments, an operation example in the image processing submodule and a setting example of setting information associated therewith will be described with some examples.

(Embodiment 1)
FIG. 7 shows an outline in the case where image processing is performed using each of the three attribute values transmitted from the outside in the three attribute value corresponding sub-modules. Here, a case where backlight correction, red-eye correction, and noise correction are performed during photo printing will be described as an example. Note that the processing image data to be used has a format 302 (three attribute value fields) shown in FIG.

  Here, first, processing performed in each submodule will be described. Backlight correction is a process for brightening a dark part (for example, gradation conversion using a one-dimensional lookup table), and it is necessary to perform a correction process only on the dark part. Red-eye correction is a process for blackening red (for example, color conversion using a three-dimensional lookup table), and it is necessary to perform a correction process only on the red-eye area. The noise correction is a process for blurring the noise generation area (for example, smoothing using a two-dimensional planar filter), and it is necessary to perform the correction process only on the noise generation area. Note that these correction processes may be realized by any method.

The relationship between the submodule and the correction process is as follows.
Backlight correction processing sub-module 11-1: Backlight correction processing (gradation conversion processing using a one-dimensional lookup table: no correction at attribute value 0, correction at attribute value 1)
Red-eye correction processing sub-module 11-2: Red-eye correction processing (color conversion processing using a three-dimensional lookup table: no correction at attribute value 0, correction at attribute value 1)
Noise correction processing sub-module 11-3: Noise correction processing (smoothing processing using a two-dimensional plane filter: no correction at attribute value 0, correction at attribute value 1)
Note that the presence or absence of correction is merely an example, and the value of the attribute value (8-bit: 0 to 255) may be used as the correction strength. Further, instead of designating the presence or absence of correction according to the attribute value, a lookup table or a filter coefficient referred to by the attribute value may be switched. In that case, for example, the first look-up table is used as a table without correction, and the second look-up table is provided as a table with correction. Furthermore, a third look-up table may be provided to correct in the reverse direction. In this case, for example, processing such as blurring, no correction, edge enhancement, and the like can be switched according to the attribute value.

  Here, an example of the flow of image processing (for example, during photo printing) in the configuration shown in FIG. 7 will be described with reference to FIG.

  The image processing apparatus 100 first determines in the CPU 1 whether or not the photo to be processed is a backlight photo, and specifies a dark part to be corrected (S101). When the dark part is specified, the image processing apparatus 100 determines in the CPU 1 whether or not red eyes have occurred in the image to be processed, and specifies the red eye area to be corrected (S102). Further, the image processing apparatus 100 determines in the CPU 1 whether or not shooting noise is generated in the image to be processed, and specifies a noise generation area to be corrected (S103).

When the area to be processed is specified, the image processing apparatus 100 sets an attribute value in the CPU 1 (S104). Various methods are conceivable as the attribute value setting method, and any method may be used. For example, you may write directly in the photograph (image) in memory (RAM7). Here, the setting contents of the attribute value are shown below. This attribute value is set for each pixel according to the detection results of the processes in S101 to S103 described above.
Attribute value field A0 Backlight correction attribute value (0: correction unnecessary, 1: correction required)
Attribute value field A1 Red-eye correction attribute value (0: correction unnecessary, 1: correction required)
Attribute value field A2 Noise correction attribute value (0: correction unnecessary, 1: correction required)

  Next, when the attribute values are set, the image processing apparatus 100 performs operation settings of the modules (image processing module 4 and submodule 11) in the CPU 1 (S105). At this time, the contents shown in the lower part of FIG. In addition, in the process of S105, setting of the one-dimensional and three-dimensional lookup tables used in the submodules 11-1 to 11-3, setting of a two-dimensional filter coefficient, and the like are also performed.

  When this setting is completed, the image processing apparatus 100 activates the image processing module 4 and the sub modules 11-1 to 11-3 in the CPU 1 (S106). As a result, each of the submodules performs various image processing using the attribute values individually. Details of the image processing performed in the submodules 11-1 to 3 will be described below.

[Backlight compensation processing sub-module]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode31 Image processing based on backlight correction attribute value A0 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (backlight correction) Combined with output pixel value) and output
[Red-eye correction processing submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode32 Image processing based on red-eye correction attribute value A1 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (red-eye correction) Combined with output pixel value) and output
[Noise correction processing submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode33 Image processing based on noise correction attribute value A2 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (noise removal) Combined with output pixel value) and output

(Embodiment 2)
Next, a case where noise removal correction is performed by individually switching the degree of blurring for each of the pixel values RGB will be described with reference to FIG. Note that the processing image data to be used has a format 302 (three attribute value fields) shown in FIG.

The relationship between the submodule and the correction process is as follows.
Backlight correction processing sub-module 11-1 Does not operate Red-eye correction processing sub-module 11-2 Does not operate Noise correction processing sub-module 11-3 Noise correction processing (smoothing processing using a two-dimensional plane filter. Attribute value A0 is 0 and R correction is not performed , Attribute value A0 is 1 and R is corrected, attribute value A1 is 0 and G is not corrected, attribute value A1 is 1 and G is corrected, attribute value A2 is 0 and B is not corrected, attribute value A2 is 1 and B is corrected )

Since the flow of processing is the same as that in FIG. 8 described above, only the differences will be described here. The difference is that the process starts from S103. In the process of S104, the attribute value is set for each pixel as follows according to the detection result of the process of S103.
Attribute value field A0 Noise correction attribute value R (0: R correction unnecessary, 1: R correction required)
Attribute value field A1 Noise correction attribute value G (0: G correction unnecessary, 1: G correction required)
Attribute value field A2 Noise correction attribute value B (0: B correction unnecessary, 1: B correction required)

  In the setting process in S105, the contents shown in the lower part of FIG. 9 are set as the setting information related to the submodules 11-1 to 11-3. In addition, in the process of S105, the setting of the two-dimensional filter coefficient used in the submodule (noise correction processing submodule) 11-3, the non-use setting of the submodules 11-1 and 11-2, and the like are also performed. In the activation process in S106, the submodule 11-3 performs various image processes using attribute values according to the settings shown below.

[Noise correction processing submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode34 Image processing based on attribute values A0 to A2 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (noise removal) Combined with output pixel value) and output

  When configured as described above, only the noise correction processing submodule 11-3 performs image processing based on all the attribute values in the attribute value field. At this time, the other submodules 11-1 and 11-2 do not perform image processing (attribute value unused). Needless to say, in this case, the backlight correction processing submodule 11-1 and the red-eye correction processing submodule 11-2 can be made to perform static image processing that does not use attribute values.

  In the noise correction processing, it is determined whether or not correction is performed for each of the RGB channels. However, the user may instruct whether or not to perform correction, or the image may be corrected. It may be determined automatically based on the analysis result. In the case of performing automatically, for example, it may be performed based on the processing results of S101 and S102 shown in FIG. Specifically, when it is determined that the necessity for backlight correction and red-eye correction is low, noise correction may be performed on each of the RGB channels. That is, when the number of correction processes to be corrected by a series of processes in one image processing module 4 is large, all the channels (RGB) may be corrected using one attribute value for each correction. Further, when the number of correction processes to be corrected by a series of processes in one image processing module 4 is small, each channel may be corrected more finely using each attribute value. For example, as an example of a case where the number of correction processes to be corrected by a series of processes in one image processing module 4 is large, three corrections of backlight correction, red-eye correction, and noise correction are performed. In this case, for backlight correction, all channels (RGB) are corrected using one attribute value A0, for red-eye correction, all channels (RGB) are corrected using one attribute value A1, and noise correction is performed. Corrects all channels (RGB) using one attribute value A2. On the other hand, as an example of a case where the number of correction processes to be corrected by a series of processes in one image processing module 4 is small, for example, only noise correction is performed. In this case, the correction can be made more finely by correcting the R channel using the attribute value A0, correcting the G channel using the attribute value A1, and correcting the B channel using A2.

(Embodiment 3)
Next, as a use case different from the case described above with reference to FIG. 10, the image area is separated from the character area and the photograph area at the time of copying, the color correction processing for the photograph is performed in the photograph area, and the character processing is performed in the character area. A case where processing is performed will be described. Note that the processing image data to be used has a format 302 (three attribute value fields) shown in FIG.

  Here, first, processing performed in each submodule will be described. Image area separation is a process for determining a character area and a photographic area (eg, determination by comparing with a threshold value using a local density change). Color correction is a process for performing color correction suitable for a photograph (for example, color conversion using a three-dimensional lookup table), and it is necessary to perform a correction process only on a photograph area. Character processing is processing for sharpening characters (eg, edge enhancement processing using a two-dimensional plane filter), and correction processing needs to be performed only on the character area. Note that these correction processes may be realized by any method.

The relationship between the submodule and the correction process is as follows.
Image Area Separation Processing Submodule 11-0 Image Area Separation Process (Comparison processing between the pixel value of the target pixel and the pixel values of surrounding pixels around the target pixel: the attribute value is set to 0 for the character area, and the attribute value for the photo area Is set to 1)
Color correction processing sub-module 11-2 Color correction processing (color conversion processing using a three-dimensional lookup table: no correction at attribute value 0, correction at attribute value 1)
Character processing sub-module 11-3 Character processing (smoothing using a two-dimensional plane filter: correction with attribute value 0, no correction with attribute value 1)
Here, the code 11-0 is used for the image area separation processing submodule. This is provided for convenience in order to clearly show that the backlight correction processing submodule 11-1 in FIGS. 7 and 8 performs the correction processing, but is a submodule that performs image area separation processing different from the correction processing. It is a code. That is, the image area separation processing submodule 11-0 is a submodule corresponding to the submodule 11-1 shown in FIG.

  Although the names are different, the color correction processing sub-module and the above-described red-eye correction processing sub-module can be realized by the same sub-module, and the character processing processing sub-module and the above-described noise correction processing sub-module are realized by the same module. it can. However, it is necessary to change and set parameters and attribute values. That is, in the same submodule, different image processing can be performed when printing a photo and when copying.

  Since the flow of processing is the same as that in FIG. 8 described above, only the differences will be described here. The difference is that the process starts from S105. Further, in the setting process in S105, the contents shown in the lower part of FIG. 10 are set as the setting information related to the submodules 11-0, 11-2, and 11-3. In addition, in the process of S105, setting of a three-dimensional lookup table and a two-dimensional filter coefficient used in the submodules 11-2 and 11-3 are also performed. Further, in the activation process in S106, each of the submodules performs various image processes using the attribute values according to the settings shown below.

[Image area separation processing submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode30 Attribute value unused Attribute value output processing Out_Mode41 Update image area separation attribute value A0 Output data processing Put_Mode22 Set attribute values A0 to A2 (image area separation output pixels Combined with value) and output
[Color correction processing submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode31 Image processing based on image area separation attribute value A0 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (red-eye Combined with the corrected output pixel value) and output
[Text Processing Submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode31 Image processing based on image area separation attribute value A0 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (noise) Combined with the removal output pixel value) and output

  When configured as described above, the attribute value of the same attribute value field can be shared by a plurality of submodules. Therefore, attribute values can be used effectively. For example, in a submodule not shown in FIG. 10, processing based on attribute value fields (for example, A1 and A2) that are not used by the submodule shown in FIG. 10 can be performed. Of course, the attribute value fields A0 and A1 are set in the image area separation processing submodule 11-0, and image processing is performed using the set attribute values in the other submodules 11-2 and 11-3. You may comprise as follows.

  In this way, the assignment method, usage method, and setting method of attribute value fields in the processing image data can be set in units of sub-modules, so flexible image processing is performed while reducing the number of attribute values in the entire system. can do.

(Embodiment 4)
Next, the case where the attribute value compatible sub-module and the attribute value non-compatible sub-module are mixed will be described. FIG. 11 shows two attribute value corresponding sub-modules and one attribute value non-corresponding sub-module. Here, a case will be described in which image processing is performed in an attribute value non-corresponding submodule after image processing is performed using attribute values transmitted from the outside in two attribute value corresponding submodules.

  Here, each submodule performs processing using the format 302 (three attribute value fields) shown in FIG. However, the format 301 shown in FIG. 4 is input to the noise correction processing submodule 11-4 that does not support attribute values after the attribute value removal processing.

The relationship between the submodule and the correction process is as follows.
Backlight correction processing sub-module 11-1 Backlight correction processing (gradation conversion processing using a one-dimensional lookup table: no correction at attribute value 0, correction at attribute value 1)
Red-eye correction processing sub-module 11-2 Red-eye correction processing (color conversion processing using a three-dimensional lookup table: no correction at attribute value 0, correction at attribute value 1)
Noise correction processing submodule 11-4 Noise correction processing (smoothing processing using a two-dimensional planar filter: processing is performed uniformly on all pixels with the same filter coefficient)

Since the flow of processing is the same as that in FIG. 8 described above, only the differences will be described here. As a difference, after the process of S102, the process proceeds to S104 without performing the process of S103. Here, the setting contents of the attribute value in the process of S104 are shown below. This attribute value is set for each pixel in accordance with the detection results of the processes in S101 and S102 described above.
Attribute value field A0 Backlight correction attribute value (0: correction unnecessary, 1: correction required)
Attribute value field A1 Red-eye correction attribute value (0: correction unnecessary, 1: correction required)
Attribute value field A2 Not set

  Further, in the setting process in S105, the contents shown in the lower part of FIG. 11 are set as the setting information related to the submodules 11-1, 11-2, and 11-4. In addition, in the process of S105, setting of one-dimensional and three-dimensional lookup tables used in the submodules 11-2 and 11-3, setting of a two-dimensional filter coefficient, and the like are also performed. Further, in the activation process in S106, each of the submodules performs various image processes using the attribute values according to the settings shown below.

[Backlight compensation processing sub-module]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode31 Image processing based on backlight correction attribute value A0 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Attribute values A0 to A2 are backlit correction output pixels Join with value
[Red-eye correction processing submodule]
Input data processing Get_Mode128bit Extracts attribute values A0 to A2 Attribute value input processing In_Mode32 Image processing based on red-eye correction attribute value A1 Attribute value output processing Out_Mode40 No update of attribute value Output data processing Put_Mode20 Red-eye correction output pixel without using attribute value Output the value by extending each channel (32 bits)
[Noise correction processing submodule]
Static image processing Input RGB 32bit pixel value, perform filter processing uniformly on entire image, then output RGB 32bit filter output pixel value

  The reason why the red-eye correction processing submodule 11-2 expands to 32 bits without using attribute values is to adapt to the input format of the noise correction processing submodule 11-4 that does not support attribute values. . Here, the attribute value non-corresponding sub-module 11-4 as the noise correction sub-module has been described as a sub-module that processes the format 301 (RGB each channel 32 bits) shown in FIG. Not limited to this, if the input format of the noise correction processing submodule 11-4 that does not support attribute values is 24 bits for each channel, the red-eye correction processing submodule performs output data processing in “Put_Mode 21”. That is, all 8 bits of the attribute value area are set to “0”. Furthermore, if the input format of the noise correction processing submodule 11-4 that does not support attribute values is 16 bits for each channel, the red-eye correction processing submodule performs output data processing with “Put_Mode23”. That is, all 16 bits of the attribute value area are set to “0”. In the present embodiment, the processing image data subjected to noise correction by the attribute non-corresponding submodule 11-4 is output to the write buffer 12, but the attribute noncorresponding submodule 11-4 has a function of rewriting the attribute value field. Absent. Therefore, when it is not necessary to output the attribute value used in the image processing module 4 of this embodiment to the outside of the image processing module 4, before inputting the processing image data to the attribute non-corresponding submodule 11-4. This is because the attribute value in the attribute value field is deleted.

  According to the above configuration, all the attribute values are set to “0” in the attribute value-compatible submodule that performs processing immediately before the attribute value non-compatible submodule. Therefore, the attribute value non-corresponding submodule need not have a function of rewriting the attribute value, and the circuit configuration of the image processing module 4 can be simplified. As a result, even when the attribute value corresponding sub-module and the attribute value non-corresponding sub-module are mixed in the image processing module 4, the image processing can be performed without any problem.

(Embodiment 5)
Next, with reference to FIG. 12, a case where the attribute value corresponding sub-module and the attribute value non-corresponding sub-module are mixed and the attribute value transmission wrapper layer 110 is provided will be described. Here, a case will be described in which image processing is performed using attribute values transmitted from the outside in two attribute value-compatible sub-modules after image processing is performed in the attribute value-incompatible sub-module.

  Here, each submodule performs processing using the format 302 (three attribute value fields) shown in FIG. However, the format 301 shown in FIG. 4 is input to the backlight correction processing submodule 11-5 that does not support attribute values by providing the attribute value transmission wrapper layer 110.

The relationship between the submodule and the correction process is as follows.
Backlight correction processing sub-module 11-5 Backlight correction processing (gradation conversion processing using a one-dimensional lookup table: the same processing is uniformly performed on all pixels)
Red-eye correction processing sub-module 11-2 Red-eye correction processing (color conversion processing using a three-dimensional lookup table: no correction at attribute value 0, correction at attribute value 1)
Noise correction processing sub-module 11-3 Noise correction processing (smoothing using a two-dimensional plane filter: no correction at attribute value 0, correction at attribute value 1)

Since the flow of processing is the same as that in FIG. 8 described above, only the differences will be described here. The difference is that the process starts from S102. Further, in the process of S104, attribute values are set for each pixel as follows according to the detection results of the processes of S102 and S103.
Attribute value field A0 Not set Attribute value field A1 Red-eye correction attribute value (0: correction unnecessary, 1: correction required)
Attribute value field A2 Noise correction attribute value (0: correction unnecessary, 1: correction required)

  In the setting process in S105, the contents shown in the lower part of FIG. 12 are set as the setting information related to the submodules 11-2 and 11-3 and the attribute value transmission wrapper layer 110. In addition, in the processing of S105, setting of one-dimensional and three-dimensional lookup tables and two-dimensional filter coefficients used in the submodules 11-2 and 11-3 and the attribute value transmission wrapper layer 110 are also performed. Further, in the activation process in S106, each of the submodules performs various image processes using the attribute values according to the settings shown below.

[Attribute value transmission wrapper layer 110]
Input data processing Get_Mode12 Attribute values A0 to A2 are extracted Output data processing Put_Mode22 Attribute values A0 to A2 are set (combined with backlight correction output pixel values) and output
[Red-eye correction processing submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode32 Image processing based on red-eye correction attribute value A1 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (red-eye correction) Combined with output pixel value) and output
[Noise correction processing submodule]
Input data processing Get_Mode12 Extract attribute values A0 to A2 Attribute value input processing In_Mode33 Image processing based on noise correction attribute value A2 Attribute value output processing Out_Mode40 No attribute value update Output data processing Put_Mode22 Set attribute values A0 to A2 (noise removal) Combined with output pixel value) and output

  According to the above configuration, by providing the attribute value transmission wrapper layer 110, there is no problem even when the attribute value corresponding sub-module and the attribute value non-corresponding sub-module are mixed in the image processing module 4. Image processing can be performed.

  In this way, the assignment method, usage method, and setting method of attribute value fields in the processing image data can be set in units of sub-modules, so flexible image processing is performed while reducing the number of attribute values in the entire system. can do. Furthermore, sub-modules corresponding to attribute values can be arranged together with existing sub-modules not corresponding to attribute values.

(Embodiment 6)
Here, a case will be described in which processing is performed in each submodule using the format 303 (six attribute value fields) shown in FIG.

Here, the setting information set by the input attribute value information setting unit 409 will be described using “Table 5”. The second attribute value management unit 411 executes various processes based on the setting information in “Table 5” before starting the image processing. The set values shown in “Table 5” are merely examples, and the present invention is not limited to these. In this embodiment, A0 is the attribute value field 3034 of the format 303 shown in FIG. 4, A1 is the attribute value field 3035 of the format 303 shown in FIG. 4, and A2 is the attribute value of the format 303 shown in FIG. Field 3036. A3 is the attribute value field 3037 of the format 303 shown in FIG. 4, A4 is the attribute value field 3038 of the format 303 shown in FIG. 4, and A5 is the attribute value field 3039 of the format 303 shown in FIG. It is.
(Table 5)

[In_Mode50]
With this setting, attribute values are not used for image processing. That is, the attribute value is not input to the image processing unit 413 in the second attribute value management unit 411. In this case, the submodule operates as an attribute value non-corresponding submodule.

[In_Mode 51-56]
This setting is adaptive to all the pixel value fields (3 channels) using any one of the attribute value fields (A0 to A5) (A5 for In_Mode 51 and A5 for In_Mode 56). This is suitable when performing various image processing. When “In_Mode 51 to 56” is set, the image processing is performed without using other attribute value fields.

[In_Mode57]
This setting is adaptive using the attribute value field A0 as the pixel value R, the attribute value field A2 as the pixel value G, and the attribute value field A4 as the pixel value B among the plurality of attribute value fields (A0 to A5). This is suitable for image processing. When “In_Mode 57” is set, the image processing is performed without using other attribute value fields (A1, A3, A5).

[In_Mode58]
This setting is adaptive using the attribute value field A1 as the pixel value R, the attribute value field A3 as the pixel value G, and the attribute value field A5 as the pixel value B among the plurality of attribute value fields (A0 to A5). This is suitable for image processing. When “In_Mode 58” is set, the image processing is performed without using other attribute value fields (A0, A2, A4).

  As described above, by performing processing using “In_Mode 50 to 58” selectively, adaptive image processing can be performed using at least one piece of information in a plurality of attribute value fields. For example, the following usage examples are given.

1) RGB common image processing using different attribute values in 6 submodules 2) RGB individual image processing using 3 attribute values each in 2 submodules 3) 3 sub In the module, RGB common image processing is performed using different attribute values, and further, in each sub module, RGB individual image processing is performed using three attribute values.

Next, setting information set by the output attribute value information setting unit 410 will be described using “Table 6”. The second attribute value management unit 411 executes various processes based on the setting information of “Table 6” after the end of the image processing. Note that the setting values shown in “Table 6” are merely examples, and the present invention is not limited thereto.
(Table 6)

[Out_Mode60]
This setting is suitable when the attribute value is not set by itself (the submodule). That is, the output attribute value remains as the input attribute value, and both are the same value.

[Out_Mode 61-66]
This setting is suitable when only one of the attribute value fields (A0 to A5) is newly set (A5 for Out_Mode 61 and A5 for Out_Mode 66). When “Out_Mode 61 to 66” is set, other attribute value fields are not updated.

[Out_Mode67]
This setting is suitable when only the attribute value fields A0, A2, and A4 are newly set among the plurality of attribute value fields (A0 to A5). When “Out_Mode 67” is set, the other attribute value fields (A1, A3, A5) are not updated.

[Out_Mode68]
This setting is suitable when only the attribute value fields A1, A3, and A5 are newly set among the plurality of attribute value fields (A0 to A5). When “Out_Mode 68” is set, the other attribute value fields (A0, A2, A4) are not updated.

  As described above, by selectively using “Out_Mode 60 to 68”, it is possible to change the setting of at least one piece of information in the plurality of attribute value fields. For example, the following usage examples are given.

1) Case where different attribute values are set in each of the six submodules 2) Case where three attribute values are set in each of the two submodules 3) Each of the three submodules sets different attribute values In addition, the case where three attribute values are set in one submodule

  Thus, when the attribute value field increases, the flexibility of image processing can be further increased. Even when the attribute value field is increased, the data amount (96 bits) of the processing image data does not change. Therefore, a design that balances the number of bits to be assigned as pixel values (tone gradation) and the number of bits to be assigned as attribute values (flexibility of adaptive image processing) without greatly changing the scale of the entire system. Can be done.

(Embodiment 7)
Here, a case will be described in which processing is performed in each submodule using the format 305 (four attribute value fields) shown in FIG. In the present embodiment, A0 is the attribute value field 3054 of the format 305 shown in FIG. 4, and A1 is the attribute value field 3055 of the format 305 shown in FIG. Further, A2 is the attribute value field 3056 of the format 305 shown in FIG. 4, and A3 is the attribute value field 3057 of the format 305 shown in FIG. In the format 305, as described above, the attribute value field A0 and the attribute value field A1 have a 16-bit width, while the attribute value field A2 and the attribute value field A3 have an 8-bit width. The attribute value field A2 and the attribute value field A3 may be handled as consecutive 16-bit width attribute values, or may be handled as different attribute values of 8 bits each. Furthermore, the attribute value fields A0 and A2 and the attribute value fields A1 and A3 may be handled as attribute values having a width of 24 bits.

  When configured in this manner, the balance between the number of attribute values (type of adaptive image processing) and the number of bits of each attribute value (fineness of adaptive image processing) is maintained while the format of the processing image data is fixed. Therefore, more flexible image processing can be realized.

Here, the setting information set by the input attribute value information setting unit 409 will be described using “Table 7”. The second attribute value management unit 411 executes various processes based on the setting information in “Table 7” before starting the image processing. Note that the setting values shown in “Table 7” are merely examples, and the present invention is not limited thereto.
(Table 7)

[In_Mode70]
With this setting, attribute values are not used for image processing. That is, the attribute value is not input to the image processing unit 413 in the second attribute value management unit 411. In this case, the submodule operates as an attribute value non-corresponding submodule.

[In_Mode71]
This setting is suitable when adaptive image processing is performed on all the pixel value fields (3 channels) using only the attribute value field A0 among the plurality of attribute value fields (A0 to A3). When “In_Mode 71” is set, image processing is performed without using other attribute value fields (A1, A2, A3).

[In_Mode72]
This setting is suitable when adaptive image processing is performed on all the pixel value fields (3 channels) using only the attribute value field A1 among the plurality of attribute value fields (A0 to A3). When “In_Mode 72” is set, the image processing is performed without using other attribute value fields (A0, A2, A3).

[In_Mode73]
This setting uses an attribute value field A2 and A3 as one attribute value (16 bits) among a plurality of attribute value fields (A0 to A3), and is an image adaptive to all pixel value fields (3 channels). This is suitable when processing is performed. When “In_Mode 73” is set, the image processing is performed without using other attribute value fields (A0, A1).

[In_Mode74]
In this setting, among a plurality of attribute value fields (A0 to A3), the attribute value field A0 is a pixel value R, the attribute value field A1 is a pixel value G, and the attribute value fields A2 and A3 are one attribute value (16 bits). Used for pixel value B. The use of such attribute values is suitable for performing adaptive image processing.

[In_Mode75]
This setting uses an attribute value field A0 and A2 as one attribute value (24 bits) among a plurality of attribute value fields (A0 to A3), and is an image adaptive to all the pixel value fields (3 channels). This is suitable when processing is performed. When “In_Mode 75” is set, the image processing is performed without using the other attribute value fields (A1, A3).

[In_Mode76]
This setting uses an attribute value field A1 and A3 as one attribute value (24 bits) among a plurality of attribute value fields (A0 to A3), and is an image adaptive to all pixel value fields (3 channels). This is suitable when processing is performed. When “In_Mode 76” is set, the image processing is performed without using other attribute value fields (A0, A2).

[In_Mode77]
This setting uses the attribute value field A2 as one attribute value (8 bits) among a plurality of attribute value fields (A0 to A3), and performs adaptive image processing for all the pixel value fields (3 channels). It is suitable for implementation. When “In_Mode 77” is set, the image processing is performed without using other attribute value fields (A0, A1, A3).

[In_Mode78]
This setting uses the attribute value field A3 as one attribute value (8 bits) among a plurality of attribute value fields (A0 to A3), and performs adaptive image processing for all the pixel value fields (3 channels). It is suitable for implementation. When “In_Mode 78” is set, the image processing is performed without using other attribute value fields (A0, A1, A2).

  As described above, by performing processing using “In_Mode 70 to 78” selectively, adaptive image processing can be performed using at least one piece of information in a plurality of attribute value fields.

Next, setting information set by the output attribute value information setting unit 410 will be described using “Table 8”. The second attribute value management unit 411 executes various processes based on the setting information in “Table 8” after the image processing is completed. The set values shown in “Table 8” are merely examples, and the present invention is not limited to these.
(Table 8)

[Out_Mode80]
This setting is suitable when the attribute value is not set by itself (the submodule). That is, the output attribute value remains as the input attribute value, and both are the same value.

[Out_Mode81]
This setting is suitable when only the attribute value field A0 is newly set among a plurality of attribute value fields (A0 to A3). When “Out_Mode 81” is set, the other attribute value fields (A1, A2, A3) are not updated.

[Out_Mode82]
This setting is suitable when only the attribute value field A1 is newly set among the plurality of attribute value fields (A0 to A3). When “Out_Mode 82” is set, the other attribute value fields (A0, A2, A3) are not updated.

[Out_Mode83]
This setting is suitable when the attribute value fields A2 and A3 are handled as one attribute value (16 bits) and are newly set for the field. When “Out_Mode 83” is set, the other attribute value fields (A0, A1) are not updated.

[Out_Mode84]
This setting is suitable for newly setting all of the plurality of attribute value fields (A0 to A3). In this case, the attribute value fields A2 & A3 are handled as one attribute value (16 bits).

[Out_Mode85]
This setting is suitable when the attribute value fields A0 and A2 are handled as one attribute value (24 bits) and are newly set for the field. When “Out_Mode 85” is set, the other attribute value fields (A1, A3) are not updated.

[Out_Mode86]
This setting is suitable when the attribute value fields A1 and A3 are handled as one attribute value (24 bits) and are newly set for the field. When “Out_Mode 86” is set, the other attribute value fields (A0, A2) are not updated.

[Out_Mode87]
This setting is suitable when only the attribute value field A2 is newly set among the plurality of attribute value fields (A0 to A3). When “Out_Mode 87” is set, the other attribute value fields (A0, A1, A3) are not updated.

[Out_Mode88]
This setting is suitable when only the attribute value field A3 is newly set among the plurality of attribute value fields (A0 to A3). When “Out_Mode 88” is set, the other attribute value fields (A0, A1, A2) are not updated.

  By performing processing using “Out_Mode 80 to 88” selectively as described above, adaptive image processing can be performed using at least one piece of information in a plurality of attribute value fields.

  The above is an example of a typical embodiment of the present invention, but the present invention is not limited to the embodiment described above and shown in the drawings, and can be appropriately modified and implemented without departing from the scope of the present invention. .

Claims (12)

Input means for inputting image data representing an image or processing data including image data representing an image and attribute data representing the processing content of the image data;
A plurality of processing means for sequentially processing the processing data input by the input means,
At least one processing means of the plurality of processing means is configured to process the image data included in the processing data input by the input means, and use the result obtained by the processing of the image data. Image processing characterized in that it is possible for subsequent processing means to newly set attribute data for processing the image data and to output processing data as processed data apparatus. When the processing data input by the input means does not include attribute data, or when it is set not to use the attribute data, the processing means for processing input by the input means The image processing apparatus according to claim 1, wherein predetermined processing is performed on image data included in the data. 3. The image processing according to claim 1, wherein at least one processing unit of the plurality of processing units includes a determination unit that determines attribute data to be processed by the processing unit. apparatus. The attribute data included in the processing data input by the input means includes a plurality of attribute data having different processing contents,
The image processing apparatus according to claim 3, wherein the determining unit determines a usage method for each of the plurality of attribute data. The image data included in the processing data input by the input unit is color image data, and the attribute data includes processing contents for each color. Image processing device. The plurality of processing units include a first processing unit to an Nth processing unit (N is a natural number of 2 or more), and the processed data processed by each processing unit is sequentially piped to the Nth processing unit. The image processing apparatus according to claim 1, wherein line processing is performed. The processing unit can set the attribute data when the processing data input by the input unit does not include the attribute data. The image processing apparatus described. The processing means sets the attribute data to 0 or deletes the attribute data when the processing means to be processed next does not correspond to the attribute data. 8. The image processing device according to any one of items 7. The number of bits assigned to the image data and the attribute data in the processing data input by the input means is a predetermined number of bits,
The number of bits to which the image data and the attribute data in the processed data output by the processing unit are allocated is the same number of bits as the predetermined number of bits. The image processing apparatus according to item. The image processing apparatus according to claim 9, wherein the processing unit includes a changing unit that changes a ratio between the image data and the attribute data in the predetermined number of bits. The image processing apparatus according to claim 1, wherein the processing data input by the input unit includes a plurality of image data and a plurality of attribute data. A processing method of an image processing apparatus having a plurality of processing means,
Input image data representing an image or processing data including image data representing an image and attribute data representing the processing content of the image data,
Sequentially processing the input processing data by the plurality of processing means;
In at least one of the plurality of processing means, the processing subsequent to the processing means using the processing of the image data included in the input processing data and the result obtained by the processing of the image data. And a processing method for newly setting attribute data for processing the image data, and processing data can be output as processed data.

Images