JP2004120135A - Image processor, and control method therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor which is easily structured and designed and can change parameters with respect to compression in units of pixel blocks such as a tile. <P>SOLUTION: The image processor includes: a division section 2112 for dividing image data received from an image input section such as a scanner 2070 in units of prescribed pixel blocks; an image compression section 2106 for applying compression coding to the divided pixel block; and an image storage section 2002 for storing compression coded information subjected to compression coding, and the division section, the image compression section, and the image storage section make communication with each other by using packet data. Then the division section 2112 discriminates the attribute of the pixel block in the case of dividing the received image data into the prescribed pixel block and outputs information with respect to the produced attribute as attached information together with data of the pixel block as packet data. Further, the image compression section determines the parameters for compression coding according to the attribute information in the packet data outputted from the division section and applies compression coding to the data of the pixel block included in the packet data. The storage section stores the packet data compressed by the image compression section. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image processing apparatus and a control method therefor, and more particularly, to an image processing apparatus that compresses and encodes image data and a control method therefor.
[0002]
[Prior art]
Conventionally, when an image processing apparatus stores image data and image information accompanying the image data in an internal memory of the processing apparatus from an image input apparatus such as an image reading apparatus, the image data and the image information captured by the image processing apparatus are In general, the image is divided into tiles, and the amount of data is reduced by performing JPEG compression processing for each tile, and then the image is stored in a memory.
[0003]
This is done not only to reduce the memory capacity for storing image data and image information, but also to reduce the amount of data transferred between the image processing device and the memory and between the memory and the printing device when printing. Is
[0004]
In some cases, data communication of compressed tile image data is performed in a packet format in which one tile corresponds to one packet. The packet is composed of a header part and an image data part. The header part stores information such as the data size of the packet, the position of the tile in the entire image, and the content of the image processing. Tile image data or uncompressed tile image data is stored.
[0005]
[Problems to be solved by the invention]
However, in the conventional example, setting has not been changed for a tile at a specific position. In the conventional example, compression is performed for each page, and a compression parameter is set in the image compression circuit at the top of the page.
[0006]
It is desirable to change the compression parameters in accordance with the characteristics of the image. For example, for tiles on which characters are written, it is desirable to set a compression parameter that does not compress high frequency components so much that a fine image is obtained. For tiles on which natural images are written, increase the image compression ratio of high frequency components. It is desirable to set the compression parameter set as above and obtain an image having a small data size. In a conventional image processing apparatus, when images having different characteristics such as a character image tile and a natural image tile are mixed in one page, it is impossible to change the image compression parameter according to the characteristics of each tile image. However, there is a problem that it is difficult to perform image compression with a high compression ratio while maintaining image quality.
[0007]
SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and provides an image processing apparatus capable of changing parameters related to compression in units of pixel blocks such as tiles and a control method thereof while facilitating the apparatus configuration and apparatus design. Is what you do.
[0008]
Another object of the present invention is to provide an image processing apparatus capable of compressing an image to a target size and a control method thereof, in addition to the above objects.
[0009]
[Means for Solving the Problems]
In order to solve such a problem, for example, an image processing apparatus of the present invention has the following configuration.
[0010]
A division unit that divides image data input from the image input unit into predetermined pixel block units, an image compression unit that compresses and encodes the divided pixel blocks, and an image storage unit that stores compression-encoded compression encoded information, An image processing device that communicates as packet data between the division unit, the image compression unit, and the image storage unit,
The dividing unit, when dividing the input image data into predetermined pixel blocks, determines the attribute of the pixel block, and outputs information on the generated attribute as additional information as packet data together with the data of the pixel block,
The image compression unit determines a parameter when performing compression encoding according to the attribute information in the packet data output from the division unit, and compression-encodes data of a pixel block included in the packet data,
The storage unit stores the packet data compressed by the image compression unit.
[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
[0012]
[Description of hardware]
FIG. 10 shows a block configuration of a digital multifunction peripheral according to the present embodiment. In the figure, reference numeral 2000 denotes a controller (Controller Unit) to which the image input / output control device and the image processing device of the embodiment can be applied.
[0013]
Here, reference numeral 2150 denotes a system control unit that controls the entire digital multifunction peripheral. An image processing unit 2149 performs predetermined image processing on input image data, and details thereof will be described later. An image ring 2008 connects the system processing unit 2150 and the image processing unit 2149 in a ring shape.
[0014]
Reference numeral 2012 denotes an operation unit for performing various settings and operation instruction operations, and includes a touch panel, a liquid crystal display, various switches and buttons. 2002 is a RAM, and 2003 is a ROM. Reference numeral 2143 denotes a general-purpose PCI (Peripheral Component Interconnect Bus) bus, reference numeral 2004 denotes an external storage device (for example, a large-capacity storage device such as a hard disk or MO), and reference numeral 2144 denotes a disk controller. Reference numeral 2050 denotes a modem for connecting to a public line. Reference numeral 2146 denotes a PHY / PMD which is connected to the LAN 2011. The digital multifunction peripheral can communicate with an external device via the modem 2050 and the PHY / PMD. The printer 2095, the image memory 2123, and the scanner 2070 are connected to the image processing unit 2149.
[0015]
Here, the configuration of the controller 2000 from the viewpoint of hardware will be described. A controller in a digital multifunction peripheral such as a digital copying machine is configured as a system LSI on a single semiconductor substrate for each of a system control unit 2150 and an image processing unit 2149. The semiconductor substrate described here can be called an IC chip. That is, in this embodiment, the controller 2000 is a single printed circuit board, and the system control unit 2150 and the image processing unit 2149 are mounted on this printed circuit board as an IC chip. However, the present invention is not limited to this. For example, the system control unit 2150 and the image processing unit 2149 may be mounted as IC chips on separate printed boards. Further, the system control unit 2150 and the image processing unit 2149 may be configured on the same semiconductor substrate.
[0016]
Next, FIG. 11 shows a detailed overall configuration mainly for describing the internal configuration of the system control unit 2150 and the image processing unit 2149.
[0017]
The controller 2000 is connected to a scanner 2070 (having an ADF) as an image input device and a printer 2095 as an image output device. On the other hand, the controller 2000 is connected to a LAN 2011 or a public line WAN 2051 so that image information and device information can be obtained. This is a controller for performing input / output and image development of PDL data.
[0018]
The CPU 2001 is a processor that controls the entire system. In this embodiment, an example using two CPUs is shown. These two CPUs are connected to a common CPU bus 2126 and further to a system bus bridge 2007.
[0019]
The system bus bridge 2007 is a bus switch, and includes a CPU bus 2126, a RAM controller 2124, a ROM controller 2125, an IO bus 1 (2127), an IO bus 2 (2129), an image ring interface 1 (2147), and an image ring interface 2 ( 2148) is connected.
[0020]
The RAM 2002 is a system work memory for the operation of the CPU 2001 and is also an image memory for temporarily storing image data. The RAM 2002 is controlled by a RAM controller 2124.
[0021]
The ROM 2003 is a boot ROM, and stores a system boot program. It is controlled by the ROM controller 2125.
[0022]
The IO bus 1 (2127) is a kind of internal IO bus, and is connected to a controller of a USB bus which is a standard bus, a USB interface 2138, a general-purpose serial port 2139, an interrupt controller 2140, and a GPIO interface 2141. IO bus 1 (2127) includes a bus arbiter (not shown).
[0023]
An operation unit I / F 2006 is an interface unit with the operation unit UI 2012 and outputs image data to be displayed on the operation unit 2016 to the operation unit 2012. In addition, it plays a role of transmitting information input by the user of the system from the operation unit 2012 to the CPU 2001.
[0024]
The IO bus 2 (2129) is a kind of internal IO bus, and the general-purpose bus interfaces 1 and 2 (2142) and the LAN controller 2010 are connected. IO bus 2 (2142) includes a bus arbiter (not shown).
[0025]
General-purpose bus interfaces 1 and 2 (2142) are bus bridges composed of two identical bus interfaces and supporting a standard IO bus. In the present embodiment, an example in which the PCI bus (2143) is adopted has been described.
[0026]
The external storage device 2004 in the embodiment is a hard disk drive (hereinafter, HDD) for storing system software and image data, and is connected to one PCI bus 2143 via a disk controller 2144.
[0027]
The LAN controller 2010 is connected to the LAN 2011 via the MAC circuit 2145 and the PHY / PMD circuit 2146, and inputs and outputs information. The Modem 2050 connects to the public line 2051 and inputs and outputs information.
[0028]
An image ring interface 1 (2147) and an image ring interface 2 (2148), which are packet transfer means, connect the system bus bridge 2007 and the image ring 2008 for transferring image data at high speed, and transfer the tiled data to the RAM 2002 and the RAM 2002. A DMA controller that transfers data between the processing units 2149.
[0029]
The image ring 2008, which is also a packet transfer unit, is configured by a combination of a series of unidirectional connection paths. The image ring 2008 is connected to the command processing unit 2104, the status processing unit 2105, and the tile bus 2127 in the image processing unit 1 (2149) via the image ring interface 3 (2101) and the image ring interface 4 (2102). You.
[0030]
Although the format of a packet serving as a communication unit will be described later, the information exchanged includes not only image data but also commands and interrupt requests. If dedicated signal lines are allocated to exchange each of these signals, the circuit (chip size) and the size of the LSI package increase, and the cost increases. In this regard, as described above, all the information necessary for operating the system between the system control unit 215 and the image processing unit 219 is in the form of information communication in a packet format, so that each circuit is designed independently. As a result, the degree of freedom is further increased, and each circuit can be made versatile.
[0031]
The command processing unit 2104 is connected to the register setting bus 2109 and, in addition to the connection to the image ring interface, transmits a register setting request issued from the CPU 2001 via the image ring to the corresponding block connected to the register setting bus 2109. Write to Further, based on a register read request issued by the CPU 2001, information is read from the corresponding register via the register setting bus and transferred to the image ring interface 4 (2102).
[0032]
The status processing unit 2105 monitors information of each image processing unit, generates an interrupt bucket for issuing an interrupt to the CPU 2001, and outputs the interrupt bucket to the image ring interface 4 (2012).
[0033]
To the tile bus 2107, in addition to the above blocks, functional blocks such as an image input interface 2112, an image output interface 2113, and a plurality of rectangular image processing units 2116 to 2119 and 2030 are connected.
[0034]
In the present embodiment, as the rectangular image processing unit, a multi-level conversion unit 2119, a binarization unit 2118, a color space conversion unit 2117, an image rotation unit 2030, and a resolution conversion unit 2116 are mounted. The scanner 2070 is connected to the image processing unit 2149.
[0035]
An image input interface 2112 in the image processing unit 2149 receives raster image data subjected to correction image processing by a scanner 2170 described later as input, and performs the following processing.
[0036]
First, for each pixel, it is determined whether the pixel is in a character / line image or in a halftone image, and the like, and added to image additional information (hereinafter referred to as a Z flag) 3003 provided for each pixel. Set to 1 if it is a character / line drawing, otherwise set to 0. In the embodiment, as will be described later, the image is compressed in a rectangular pixel block (32 × 32 pixel block), in other words, in a tile unit. Next, the Z flag of each pixel constituting the pixel block is compressed. , The attribute (Q-Table sel data, Char-Flag) in the target pixel block is determined. When the attribute for the pixel block is determined in this way, the resulting data is output to the subsequent tile bus 2107 (the details of how to generate a Q-Table sel will be described later). In the embodiment, the data between the processing units is performed in packet units. At this time, the data of the pixel block is in a packet format including at least header information (including Q-Table sel data) + tile data, and a clock. Output in synchronization with.
[0037]
The image output interface 2113 in the image processing unit 2149 receives the rectangular data from the tile bus 2107 as input, converts the structure into a raster image, changes the clock rate, and outputs the raster image to the printer 2095.
[0038]
The image rotation unit 2030 rotates image data, and the resolution conversion unit 2116 changes the resolution of the image by performing processing such as interpolation / thinning. The color space conversion unit 2117 performs color space conversion of color and grayscale images. The binarization unit 2118 binarizes the multi-valued color or grayscale image by error diffusion processing or dither processing. Further, the multi-level conversion unit 2119 converts a binary image into multi-level data. For example, when focusing on a certain binary pixel, the pixel of interest is converted into two or more bits by referring to the surrounding pixel group. This is converted into multi-valued pixel data.
[0039]
The memory control unit 2122 is connected to the memory bus 2108, and writes and reads image data to and from the image memory 2123 and refreshes the image data when necessary according to a request from each image processing unit. Perform the operation. In this embodiment, an SDRAM is used as an image memory.
[0040]
Next, error processing of the digital multifunction peripheral according to the present embodiment will be described.
[0041]
The image input interface 2112 has the following two errors: an overrun error that occurs when the operation of the image input interface 2112 is slower than the data transmitted by the scanner 2170 and the image input interface 2112 cannot receive data; A synchronization error that occurs when an incorrect vertical synchronization signal or horizontal synchronization signal is received from 2170 is detected. The image input interface 2112 has the following six interrupts: a scanner normal operation end interrupt that occurs when one page of data is normally read from the scanner 2170, and an image input interface 2112 that outputs one page of data. Is output, or when the image input interface 2112 is forcibly stopped from outside, or when the image input interface 2112 detects an error, an image input interface operation end interrupt that occurs and one tile Interrupt that occurs every time the output of one minute of data is completed, a tile start interrupt that occurs each time the output of one tile of data is started, and occurs when a rising edge of the vertical synchronization signal output from the scanner 2170 is detected. Vertical A rising interrupt of a signal and a falling interrupt of a vertical synchronizing signal generated when a falling of the vertical synchronizing signal is detected, and also generated when data corresponding to the number of lines set in the register is received from the scanner 2170. Generate a mark line interrupt. The image input interface 2112 detecting the error or the interrupt notifies the status processing unit 2105 of the error or the interrupt using the interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error or the interrupt using the interrupt bucket.
[0042]
The image output interface 2113 has the following two errors: the operation of the image output interface 2113 is slower than the data to be sent to the printer 2095, an underrun error that occurs when the printer 2095 cannot receive the data, and the Detects a synchronization error that occurs when an incorrect vertical or horizontal synchronization signal is received. The image output interface 2113 includes the following six interrupts: a printer normal operation end interrupt that occurs when one page of data is normally transmitted to the printer 2095, and an image output interface 2113 that outputs one page of data. , Or when the image output interface 2113 is forcibly stopped from the outside, or when the image output interface 2113 detects an error, an image output interface operation end interrupt that occurs and one tile Tile end interrupt that occurs each time the output of one minute of data ends, tile start interrupt that occurs each time the output of one tile of data starts, and the rising edge of the vertical synchronizing signal that occurs when the rising edge of the vertical synchronizing signal is detected Interrupts, and Straight synchronizing signal the vertical synchronizing signal falling interrupt that occurs when the fall is detected, and generates a mark line interrupt generated when receiving the number of data lines that is set in the register from the printer 2095. The image output interface 2113 that has detected the error or the interrupt notifies the status processing unit 2105 of the error or the interrupt using the interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error or the interrupt using the interrupt bucket.
[0043]
Here, the illegal vertical synchronizing signal or the horizontal synchronizing signal described above will be described for reference. The image output interface 2113 and the image input interface 2112 count the number of rises of the horizontal synchronization signal in the period from the rise of the vertical synchronization signal to the rise. When the normal vertical sync signal and horizontal sync signal are input, the number of rises of the horizontal sync signal in the period from the rise to the rise of the vertical sync signal does not change, but when an incorrect vertical sync signal or horizontal sync signal is input The number of rises of the horizontal synchronization signal in the period from the rise of the vertical synchronization signal to the rise changes. Here, the rising of the vertical synchronizing signal and the horizontal synchronizing signal has been described, but it goes without saying that the falling may be the falling.
[0044]
The image rotation unit 2030 has the following three errors, namely, a packet header other than the color formats (YMCK, RGBα, RGB, YUV, Lab, Y, M, C, K, ND, and BK) that can be processed by the image rotation unit 2030. ImageType error that occurs when an undefined ImageType is written, and the X coordinate or Y coordinate written in the packet header is larger than the X size or Y size of one page written in the image rotation unit 2030 An X coordinate error or a Y coordinate error occurring in the case, and a ProcessInstructionMode (0 ° rotation, 90 ° rotation, 180 ° rotation, 270 ° rotation, and the like, respectively) that can be processed by the image rotation unit 2030 in the ProcessInstructionMode written in the packet header Mirror image) Detects a ProcessInstructionMode error that occurs when an undefined ProcessInstructionMode is written. The image rotation unit 2030 that has detected the error notifies the status processing unit 2105 of the error using an interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error using an interrupt bucket.
[0045]
The resolution conversion unit 2116 includes the following six errors, a PacketType error that occurs when the PacketTypeID written in the packet header is other than a data packet, and a color format (YMCK, RGBα, RGB) that can be processed by the resolution conversion unit 2116 in the packet header. , YUV, Lab, Y, M, C, K, ND, BK), an ImageType error that occurs when an undefined ImageType is written, and the PageID written in the packet header is different from the PageID being processed. The page ID error that occurs in the case where the error occurs, the JobID error that occurs when the JobID written in the packet header is different from the JobID being processed, and the X coordinate or the Y coordinate written in the packet header are converted to the resolution conversion unit 21. 6 detects the X coordinate error or Y coordinate error that occurs when not a order that you expect. The resolution conversion unit 2116 that has detected the error notifies the status processing unit 2105 of the error using an interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error using an interrupt bucket. In addition, the resolution conversion unit 2116 generates a resolution conversion normal operation end interrupt that occurs when data of one page of resolution conversion is normally transmitted. The resolution conversion unit 2116 detecting the error or the interrupt notifies the status processing unit 2105 of the error or the interrupt using the interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error or the interrupt using the interrupt bucket.
[0046]
The color space conversion unit 2117 outputs the following three errors, namely, a color format (YMCK, RGBα, RGB, YUV, Lab, Y, M, C, K, ND, and BK) that can be processed by the color space conversion unit 2117 in the packet header. An ImageType error that occurs when an undefined ImageType other than the above is written, a PacketType error that occurs when the PacketTypeID written in the packet header is other than a data packet, and a color that can be processed by the color space conversion unit 2117 in the packet header Detects a ProcessInstructionMode error that occurs when a ProcessInstructionMode other than a format (YMCK, RGBα, RGB, YUV, Lab, Y, M, C, K, ND, and BK) is written. You. The color space conversion unit 2117 that has detected the error notifies the status processing unit 2105 of the error using an interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error using an interrupt bucket.
[0047]
The binarizing unit 2118 occurs when the following five errors, that is, ImageTypes other than the color formats (Y, M, C, K, and ND) that can be processed by the binarizing unit 2118 are written in the packet header. An ImageType error, and a PacketType error that occurs when the PacketTypeID written in the packet header indicates something other than a data packet, and a PageID error that occurs when the PageID written in the packet header is different from the PageID being processed. Also, a JobID error that occurs when the JobID written in the packet header is different from the JobID being processed, and the X-coordinate or the Y-coordinate written in the packet header of one page written in the binarization unit 2118 From X size or Y size Detecting the X coordinate error or Y coordinate error that occurs when larger. The binarizing unit 2118 that has detected the error notifies the status processing unit 2105 of the error using an interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error using an interrupt bucket.
[0048]
The multi-level conversion unit 2119 states that the following three errors, that is, an ImageType error that occurs when an ImageType other than black is written in the packet header, and a PacketTypeID written in the packet header indicate a data packet other than the data packet A PacketType error that occurs in this case and a ProcessInstructionMode error that occurs when the ProcessInstructionMode written in the packet header is other than the ProcessInstructionMode that can be processed by the multi-value coding section 2119 are detected. The multi-level converting unit 2119 that has detected the error notifies the status processing unit 2105 of the error using an interrupt signal, and the status processing unit 2105 notifies the CPU 2001 of the error using an interrupt bucket.
[0049]
Upon detecting the error or the interrupt, the CPU 2001 reads the status processing unit 2105 and the error or interrupt factor register of each image processing unit, and performs a process suitable for each factor. In the case of an error, the CPU 2001 resets the image processing unit 2149 and returns from the error.
[0050]
[Entire system]
Next, FIG. 9 shows a configuration diagram of an entire network system including the digital multifunction peripheral according to the present embodiment.
[0051]
Reference numeral 1001 denotes the digital multifunction peripheral according to the embodiment described above, which is controlled by a controller to which the image output control device and the image processing device of the present invention can be applied.
[0052]
As described above, the digital multifunction peripheral 1001 has a scanner and a printer, and can flow an image read from the scanner to a local area network 1010 (hereinafter, LAN), and can print out an image and print data received from the LAN by the printer.
[0053]
Further, the image read from the scanner can be transmitted to the PSTN or ISDN 1030 by a facsimile transmission unit as the modem 2050, and the image received from the PSTN or ISDN can be printed out by a printer. A database server 1002 manages a binary image and a multi-valued image read by the digital multifunction peripheral 1001 as a database.
[0054]
Reference numeral 1003 denotes a database client of the database server 1002, which can browse and search image data stored in the database 1002.
[0055]
An electronic mail server 1004 can receive an image read by the digital multifunction peripheral 1001 as an electronic mail attachment. Reference numeral 1005 denotes an e-mail client, which is capable of receiving and browsing the e-mail received by the e-mail server 1004 and transmitting the e-mail.
[0056]
Reference numeral 1006 denotes a WWW server that provides an HTML document to the LAN, and the digital multifunction peripheral 1001 can print out an HTML document provided by the WWW server.
[0057]
A router 1011 connects the LAN 1010 with the Internet / intranet 1012 by a router. Devices similar to the aforementioned database server 1002, WWW server 1006, e-mail server 1004, and digital multifunction peripheral 1001 are connected to the Internet / intranet as 1020, 1021, 1022, and 1023, respectively. On the other hand, the digital multifunction peripheral 1001 is capable of transmitting and receiving to and from the FAX device 1031 via the PSTN or ISDN 1030.
[0058]
A printer 1040 is also connected to the LAN, and is configured to be able to print out an image read by the digital multifunction peripheral 1001.
[0059]
[Tile image (packet) format]
Next, the format of a packet processed in the controller 2000 to which the image input / output device and the image processing device of the present embodiment can be applied will be described.
[0060]
In the controller 2000 according to the present embodiment, image data, commands from the CPU 2001, interrupt information issued from each block, and the like are transferred in a packetized format. In the present embodiment, three different types of packets are used: a data packet shown in FIG. 12, a command packet shown in FIG. 13, and an interrupt packet shown in FIG. Hereinafter, each packet will be described.
[0061]
<Data packet (FIG. 12)>
First, a data packet will be described with reference to FIG. In the present embodiment, an example has been described in which image data is divided into image data (ImageData + padding) 3002 in units of 32 pixels × 32 pixels and handled.
[0062]
The required header information (header) 3001 and additional image information (Zdata + padding) 3003 are added to the image in tile units to form a data packet (DataPacket). Hereinafter, information included in the header information 3001 will be described.
[0063]
The packet type is distinguished by a packet type (PcktType) 3004 in the header information 3001. A chip ID (Chip ID) 3005 indicates the ID of a chip that is a target for transmitting a packet. A data type (DataType) 3006 indicates a data type. A page ID (PageID) 3007 indicates a page, and a job ID (JobID) 3008 stores an ID for management by software.
[0064]
The tile number is represented by YnXn, which is a combination of the tile coordinates 3009 in the Y direction and the tile coordinates 3010 in the X direction. Data packets may be compressed image data or uncompressed. In the present embodiment, a case of non-compression has been described. The distinction between the case of compression and the case of non-compression is indicated by a compression flag (CompressFlag) 3017.
[0065]
The process instruction (Process Instruction) 3011 is set left-justified in the processing order, and each processing unit (such as the above-described rectangular image processing unit) shifts the post-processing process instruction 3011 to the left by 8 bits. The process instruction 3011 stores eight sets of a unit ID (Unit ID) 3019 and a mode (Mode) 3020. A unit ID 3019 specifies each processing unit, and a mode 3020 specifies an operation mode in each processing unit. Thereby, one packet can be processed continuously by eight units.
[0066]
A packet length (PacketByteLength) 3012 indicates the total number of bytes of the packet. The image data length (ImageDataByteLength) 3015 represents the number of bytes of image data, the Z data length (ZdataByteLength) 3016 represents the number of bytes of image additional information, and the image data offset (ImageDataOffset) 3013 and the Z data offset (ZdataOffset) 3014 are respectively Indicates the offset from the beginning of the data packet. Reference numeral 3015 denotes the number of bytes (Image Data Byte length) of the image data, and reference numeral 3016 denotes the number of bytes of Z-Data (Z Data byte length). Reference numeral 3017 denotes a flag (Compress Flag) indicating whether or not the image data in the packet is compressed, 3029 indicates a flag (Char-Flag) indicating whether or not the image is a character area, and 3030 indicates a target pixel block (tile). ) Is quantization table selection data (Q-Table Sel) for determining a quantization step when compressing.
[0067]
Next, a packet table (Packet Table) will be described with reference to FIG. Each packet is managed by a packet table 6001. The components of the packet table 6001 are as follows. When 5 bits are added to the values in the table, a packet start address (PacketAddressPointer) 6002 and a packet length (Byte Length) 6005 are obtained. Also, here, the relationship of Packet Address Pointer 27 bits + 5b00000 = packet start address, Packet Length 11 bits + 5b00000 = packet length holds. The packet table 6001 and the chain table (ChainTable) 6010 are not divided.
[0068]
The packet table 6001 is always arranged in the scanning direction, and Yn / Xn = 000/00000/0001/002 /. . . . It is arranged in the order. The entry of the packet table 6001 uniquely indicates one tile. The next entry of Yn / Xmax is Yn + 1 / X0.
[0069]
If the packet has exactly the same data as the immediately preceding packet, the packet is not written in the memory, and the same start address (PacketAddressPointer) and packet length (PacketLength) as the first entry are stored in the packet table entry. . One packet data is represented by two table entries. In this case, the repeat flag (RepeatFlag) 6003 of the second table entry is set.
[0070]
When the packet is divided into a plurality of pieces by the chain DMA, a division flag (DividFlag) 6004 is set, and a chain table number (ChainTableNo) 6006 of a chain block including the head of the packet is set.
[0071]
The entry of the chain table 6010 is composed of a chain block address (ChainBlockAddress) 6011 and a chain block length (ChainBlockLength) 6012, and 0 is stored in the last entry of the table for both the address and the data length.
[0072]
<Command packet (FIG. 13)>
Next, a command packet (Command Packet) will be described with reference to FIG.
[0073]
The command packet is for accessing the register setting bus 2109. By using the command packet, the CPU 2001 can also access the image memory 2123.
[0074]
In a chip ID (Chip ID) 4004 in the header 4001, an ID indicating an image processing unit to which a command packet is to be transmitted is stored. A page ID (Page ID) 4007 and a job ID (Job ID) 4008 store a page ID and a job ID for management by software.
[0075]
The packet ID (PacketID) 4009 is represented in one dimension. Therefore, only the X-coordinate (X-coordinate) 4009 of the data packet (Data Packet) is used. The packet length (PacketByteLength) 4010 is fixed at 128 bytes.
[0076]
The packet data section (Command) 4002 can store a maximum of 12 commands using a set of an address (Address) 4011 and data (Data) 4012 as one command. The command type of write or read is indicated by a command type (CmdType) 4005, and the number of commands is indicated by a command number (Cmdnum) 4006.
[0077]
<Interrupt packet (Fig. 14)>
Finally, an interrupt packet (Interrupt Packet) will be described with reference to FIG.
[0078]
The interrupt packet is for notifying an interrupt from the image processing unit 2149 to the CPU 2001. When transmitting the interrupt packet, the status processing unit 2105 must not transmit the interrupt packet until the CPU 2001 permits the transmission.
[0079]
The packet length (PacketByteLength) 5006 is fixed at 128 bytes. The packet data section (IntData) 5002 stores status information (ModuleStatus) 5007 of each internal module (each rectangular image processing section, input / output interface, etc.) of the image processing section. The status processing unit 2105 collects status information of each module in the image processing unit, and can collectively send the status information to the system control unit 2150.
[0080]
The chip ID (ChipID) 5004 stores an ID indicating the system control unit 2150 that is the transmission destination of the interrupt packet, and the interrupt chip ID (IntChipID) 5005 stores the ID indicating the image processing unit that is the transmission source of the interrupt. You.
[0081]
[Tile compression unit]
FIG. 1 shows a block diagram of the tile compression section 2106 in FIG.
[0082]
In FIG. 1, reference numeral 101 denotes a tile bus interface unit, which acquires a data packet input from the tile bus 2107, temporarily stores data input to an internal buffer, and a header information holding unit 102 connected at a subsequent stage. , The header information, the image data, and the image additional information data are distributed and output to the first compression processing unit 103 and the second compression processing unit 104, respectively.
[0083]
When a data packet is input from the tile bus 2107 to the tile bus interface unit 101, first, the header information of the data packet is input, and the tile bus interface unit 101 refers to the value set in the header information to check the header information. Preparations are made for input of image data and image additional information to be input later.
[0084]
The tile bus interface unit 101 analyzes header information (3001) input from the tile bus. Here, as a result of analyzing the header information, if the values set in the header information are inconsistent or set to values beyond the range which can be taken, the tile bus interface unit 101 interrupts the register setting unit 109 described later. After outputting a signal to notify that the header information is inconsistent, the acquisition of data packets from the tile bus is stopped. This state in which no data packet is obtained continues until a reset signal (not shown) is input.
[0085]
If there is no inconsistency in the header information (3001), the header information is output to the header information holding unit 102 connected at the subsequent stage, and subsequently, acquisition of image data and image additional information from the tile bus 2107 is started.
[0086]
The acquired image data and image additional information are output to the first compression processing unit 103 and the second compression processing unit 104 with reference to the ImageType (3006) of the header information.
[0087]
For example, when the ImageType in the header information is Bk, which is image data representing one pixel with one bit, the image data obtained from the tile bus is output to the second compression processing unit 104. At this time, the image additional information is discarded by the tile bus interface unit 101.
[0088]
When ImageType represents a value other than Bk (8-bit ND, 8-bit RGB data or multi-bit (multi-valued) image data), the image data is transmitted to the first compression processing unit 103, and the image additional information is transmitted to the second compression processing unit. Output to 104. However, when Ztype (3020) indicates that there is no image additional information, it is considered that there is no image additional information, and the image additional information is not subjected to compression processing in the second compression processing unit.
[0089]
Reference numeral 102 denotes a header information holding unit which holds the header information in an internal buffer and stores the stored header information while the first compression processing unit 103 and the second compression processing unit 104 compress the image data and the image additional information. The information necessary for compression processing is output to the first compression processing unit and the second compression processing unit.
[0090]
In the header information holding unit 102, the header information input from the tile bus interface unit 101 and the header information stored in the buffer of the header information holding unit before the header information is input (one in the tile compression unit 2106). The header information of the data packet that has been processed immediately before is compared with the header information input from the tile bus interface unit 101, and the comparison result is output to the packet generation unit 105 using the bus 111.
[0091]
When the compression processing is completed by the first compression processing unit 103 and the second compression processing unit 104, the header information held in the header information holding unit 102 is output to the packet generation unit using the bus 110 in response to a request from the packet generation unit 105. You.
[0092]
Reference numeral 103 denotes a first compression processing unit for compressing image data, and in this embodiment, performs compression according to the JPEG method. In addition, the first compression processing unit 103 includes, in the data packet, a pixel at a corresponding position between the image data input from the tile bus interface unit 101 and the image data that has been subjected to the compression processing by the first compression processing unit immediately before. All the pixels are compared, and the comparison result is output to the packet generation unit 105 using the bus 113.
[0093]
The encoded image data compressed by the first compression processing unit 103 is output to the packet generation unit using the bus 112 in response to a request from the packet generation unit 105.
[0094]
When the ImageType (3006) in the header information indicates that the image data is 1-bit image data (binary image data), the image data is compressed by the second compression processing unit. Does not work.
[0095]
Reference numeral 104 denotes a second compression processing unit for compressing the image additional information or the 1-bit image data. In the present embodiment, the second compression processing unit performs a compression method using no information loss, specifically, a packbits method. In addition, the second compression processing unit 104 performs first compression on the image additional information or image data input from the tile bus interface unit 101 and the image additional information or image data that has been subjected to compression processing by the second compression processing unit immediately before. Similar to the processing unit 103, all the pixels included in the data packet are compared for the image additional information or the image data for the pixel at the corresponding position, and the comparison result is output to the packet generation unit 105 using the bus 114.
[0096]
The encoded image additional information or image data compressed by the second compression processing unit 104 is output to the packet generation unit using the bus 113 in response to a request from the packet generation unit 105.
[0097]
Reference numeral 105 denotes a packet generation unit, which acquires header information, image data, and image additional information from the header information holding unit 102, the first compression processing unit 103, and the second compression processing unit 104, and compresses the image data and image additional information. Generate a data packet.
[0098]
The header information generated by the packet generation unit 105 includes a comparison result of each unit output from the header information holding unit 102, the first compression processing unit 103, and the second compression processing unit 104, and the first compression processing unit 103 The image data amount after the encoding by the second compression processing unit 104, the data amount of the image additional information, the data amount of the data packet, and the compression processing by the first compression processing unit 103 and the second compression processing unit 104 Information such as parameters used at that time is reflected.
[0099]
As described above, when the compression processing by the first compression processing unit 103 and the second compression processing unit 104 is completed to generate encoded data, and when each information is reflected in the header information, the data packet is transferred to the image ring output unit 106. Is output to
[0100]
An image ring output unit 106 outputs a data packet sent from the packet generation unit 105 to the image ring interface 4 (2102) in synchronization with the image ring interface 4 (2102).
[0101]
Reference numeral 109 denotes a register setting unit for setting a location related to processing inside the tile compression unit 2106, and sets a predetermined value in the register setting unit 109 so that the tile compression unit 2106 performs a predetermined compression process.
[0102]
These settings are sent from the system control unit 2150 to the command processing unit 2104 of the image processing unit 2149 using the command packet shown in FIG. 13, and sent from the command processing unit to the tile compression unit 2106 via the register setting bus 2109.
[0103]
The value set in the register setting unit 109 is sent to the first compression processing unit 103 and the second compression processing unit 104, and both compression processing units perform a process determined by referring to those set values.
[0104]
It should be noted that not only the value is set to the register setting unit using the command packet, but also the setting value held by the register setting unit can be output to the system control unit 2150 using the command packet.
[0105]
The register setting unit 109 has a register for storing header information. When an interrupt signal is input from the tile bus interface unit 101, the register setting unit 109 fetches the header information from the tile bus interface 101 and sets it in the register. An interrupt signal for notifying that an interrupt has occurred to the unit 2105 and a status signal indicating an error state are output.
[0106]
Reference numeral 108 denotes a register setting bus interface unit, which is a block for performing format conversion and handshaking so that the register setting unit 109 can acquire a setting value input from the register setting bus 2109.
[0107]
[First compression processing unit]
FIG. 2 is a block diagram of the first compression processing unit 103. In the drawing, reference numeral 201 denotes a first data buffer for storing image data sent from the tile bus interface unit 101, and when a predetermined amount of data is sent, the first data buffer 201 is sent to a DCT conversion unit 202 connected at a subsequent stage. Data is output in a predetermined order.
[0108]
Note that the first data buffer is input with the ImageType (3006) of the header information from the header information holding unit 102, so that the image data is input to the first data buffer and the first data buffer is transferred to the DCT conversion unit 202. The order of the data to be output to and the number of data of the image data are controlled by ImageType.
[0109]
Further, the image data is sent from the tile bus interface unit to the first data buffer, and the image data is stored in a predetermined location in the first data buffer, and at the same time the image data is stored in a location where the image data is stored. The stored image data (this is image data corresponding to the position of the transmitted image data in the image data of the data packet processed immediately before by the first compression processing unit) is compared with a data comparison described later. Output to the unit 211.
[0110]
Reference numeral 202 denotes a DCT transform unit that, when image data is input in a predetermined order from the first data buffer 201, performs discrete cosine transform to convert the image data into frequency component data. Also, at this time, the DCT transform unit 202 outputs the value of the DC component generated by performing the discrete cosine transform together with a latch signal to a thumbnail generating unit 212 described later.
[0111]
Reference numeral 203 denotes a quantization unit that performs quantization on the frequency component data output from the DCT transformation unit 202 using a predetermined quantization value. Note that values used for quantization are input from a quantization table described later. The quantization table used for quantization in the DCT transform unit 202 is determined by referring to the header information. Details of a method for determining the quantization table to be used will be described later.
[0112]
Reference numeral 204 denotes a Huffman encoding unit which performs Huffman encoding on the data output from the quantization unit 203 to generate encoded data.
[0113]
Reference numeral 205 denotes a second data buffer, which is a buffer for storing the encoded data encoded by the Huffman encoding unit 204.
[0114]
When the input of the encoded data from the Huffman encoding unit 204 ends, the second data buffer 205 outputs the data amount of the encoded image data to the packet generation unit 105 as Data Byte Length, and a data counter 214 described later. Output to
[0115]
Thereafter, the stored encoded data is output according to a request from the packet generation unit 105.
[0116]
A data comparison unit 211 compares the image data input from the tile bus interface unit 101 with the image data stored in the first data buffer 201. Here, the image data input from the first data buffer 201 to the data comparison unit 211 is not the image data of the data packet currently input from the tile bus interface unit 101 but is input from the tile bus interface unit 101. This is image data that has been subjected to compression processing by the first compression processing unit before the data packet. Therefore, the data comparison unit 211 compares the image data currently input from the tile bus interface unit 101 to the first compression processing unit with the image data processed by the first compression processing unit before that. .
[0117]
By performing the above operation over all the image data input from the tile bus interface unit 101, the data comparison unit 211 compares all the image data included in the data packet with the image data of the previous data packet. be able to.
[0118]
When the comparison is completed for one data packet, all the image data of one data packet is stored in the first data buffer 201. The image data thus stored in the first data buffer 201 is compared with the next image data by the data comparison unit 211 when the image data of the next data packet is input from the tile bus interface unit 101. Is performed.
[0119]
The result of performing the comparison operation on all image data of one packet input from the tile bus interface unit 101 is output to the packet generation unit 105 as a comparison result Compare result.
[0120]
Reference numeral 212 denotes a thumbnail generation unit that acquires a DC component generated when performing DCT conversion output from the DCT conversion unit 202 and generates a thumbnail value based on a plurality of DC components. The acquisition of the DC component from the DCT converter 202 is performed with reference to a latch signal output in synchronization with the DC component. The generated thumbnail value is output as Thumbnail Data to the packet generation unit 105, and is stored in Thumbnail Data (3021) of the header information in a predetermined format.
[0121]
Reference numeral 210 denotes a quantization table unit for storing a quantization value for performing quantization in the quantization unit 203. The quantization table means that when the DCT transform unit 202 performs the discrete cosine transform and divides the frequency component data by a predetermined value for each frequency component to perform quantization, the quantization unit removes each frequency component. A group of data that provides a quantization value for each frequency component generated by the DCT transform unit 202 is referred to as a quantization table.
[0122]
The quantization unit 203 performs quantization by obtaining a quantization value corresponding to frequency data to be quantized from an arbitrary quantization table.
[0123]
The quantization table unit of the present embodiment has four of the above quantization tables, selects a predetermined quantization table according to a selection signal input from a quantization table selection unit described later, and sends a quantization value to the quantization unit 203. Output.
[0124]
A quantization table selection unit 213 outputs a plurality of (four in the present embodiment) quantization tables stored in the quantization table unit by outputting a quantization table selection signal to the quantization table unit 210. Select a predetermined table from. The quantization unit 203 performs quantization using the quantization table selected by the quantization table selection unit 213.
[0125]
The mode (3025) and Q-Table Sel (3030) from the header information holding unit 202 for the quantization table selection unit 213, and the Q-Table ID (3028) for storing in the header information from the register setting unit 109 are determined. A register value Reg-QTField for indicating the correspondence between the register values Reg-Qmode and Q-TableID and the quantization table used by the DCT conversion unit 202, and a counter value and a register value of the data counter from a counter value comparison unit described later. The comparison result count-over is input. The quantization table selection unit 213 selects a Q-TableID to be stored in header information and a quantization table used when performing quantization in the quantization unit 203 by referring to these inputs.
[0126]
A data counter 214 acquires Data Byte Length and adds it to the internal counter each time encoded image data for one data packet is output from the second data buffer to the packet generator. By repeating the above operation, the data counter 214 holds the total amount of encoded image data sent from the tile compression unit 2106 to the RAM 2002 via the system bus bridge 2007 in the internal counter.
[0127]
The internal counter of the data counter 214 is cleared by a counter reset signal output from the header information holding unit 102. As will be described later, the header information section 102 compares the header information of the immediately preceding data packet with the Page ID (3007) and the Job ID (3008) of the header information currently being compressed by the first compression processing section 103, and When one or both of the JobIDs have different values in both header information, a counter reset signal is output.
[0128]
With the above configuration, the counter is cleared in the data counter 214 when a different page or a data packet of a different job is input. Therefore, the internal counter of the data counter 214 stores the total amount of encoded image data in the same page of the same job.
[0129]
A counter value comparing unit 215 compares the internal counter value of the data counter 214 with the register setting value Reg-JPEGCountValue for counter value output from the register setting unit 109, and selects the result as count_over to select the quantization table. Output to the unit 213.
[0130]
The details of the operations of the data counter 214, the counter value comparison unit 215, and the quantization table selection unit 213 will be described later.
[0131]
[Second compression processing unit]
FIG. 4 is a block diagram of the second compression processing unit 104. In the following description, an example in which image additional information is input from the tile bus interface unit 101 will be described. Unless otherwise specified, the same operation is performed when 1-bit image data is input.
[0132]
In FIG. 4, reference numeral 401 denotes a first data buffer for storing the image additional information transmitted from the tile bus interface unit 101, and a pack bit encoding unit connected to the subsequent stage when a predetermined amount of data is transmitted. Data is output to 402 in a predetermined order.
[0133]
Further, the image additional information is sent from the tile bus interface unit to the first data buffer, and the image additional information is stored in a predetermined location in the first data buffer, and at the same time the image additional information is stored. The image additional information stored up to that point (this is the image additional information corresponding to the position of the transmitted image additional information in the image additional information of the data packet processed immediately before by the second compression processing unit). Is output to a data comparison unit 405 described later.
[0134]
Reference numeral 402 denotes a PackBits encoding unit, which performs encoding processing according to the PackBits method on the image additional information output from the first data buffer 401 to generate encoded data.
[0135]
Reference numeral 403 denotes a second data buffer, which is a buffer for storing the encoded data encoded by the PackBits encoding unit 402.
[0136]
When the input of the encoded data from the PackBits encoding unit 402 is completed, the second data buffer 403 outputs the data amount of the encoded image additional information to the packet generation unit 105 as Data Byte Length, and outputs data to be described later. Output to the counter 406.
[0137]
Thereafter, the stored encoded data is output according to a request from the packet generation unit 105.
[0138]
A data comparison unit 405 compares the image additional information input from the tile bus interface unit 101 with the image additional information stored in the first data buffer 401. Here, the image additional information input to the data comparison unit 405 from the first data buffer 401 is not the image additional information of the data packet currently input from the tile bus interface unit 101, but is input from the tile bus interface unit 101. This is image additional information that has been subjected to compression processing by the second compression processing unit before the data packet that has been processed. Therefore, the data comparison unit 405 compares the image additional information currently input from the tile bus interface unit 101 to the second compression processing unit with the image additional information previously processed by the second compression processing unit. ing.
[0139]
By performing the above operation over all the image additional information input from the tile bus interface unit 101, the data comparison unit 405 compares all the image additional information included in the data packet with the image additional information of the previous data packet. A comparison can be made.
[0140]
When the comparison is completed for one data packet, all the image additional information of one data packet is stored in the first data buffer 401. It should be noted that the image additional information stored in the first data buffer 401 in this manner is used by the data comparing unit 405 when the image additional information of the next data packet is input from the tile bus interface unit 101. Information and comparisons are made.
[0141]
The result of performing the above-described comparison operation on all pieces of image additional information of one packet input from the tile bus interface unit 101 is output to the packet generation unit 105 as a comparison result Compare result.
[0142]
The above described Data Byte Length and Compare result are sent to the packet generator 105 via the bus 115 as information for generating header information in the packet generator 105 (see FIG. 1).
[0143]
Reference numeral 406 denotes a data counter that acquires Data Byte Length and adds it to the internal counter each time encoded image additional information for one data packet is output from the second data buffer to the packet generation unit 105. By repeating the above operation, the data counter 406 holds the total amount of the encoded image additional information sent from the tile compression unit 2106 to the RAM 2002 via the system bus bridge 2007 in the internal counter.
[0144]
The internal counter of the data counter 406 is cleared by a counter reset signal output from the header information holding unit 102. As will be described later, the header information section 102 compares the header information of the immediately preceding data packet with the Page ID (3007) and the Job ID (3008) of the header information currently being compressed by the first compression processing section 103, and When one or both of the JobIDs have different values in both header information, a counter reset signal is output.
[0145]
With the above configuration, the data counter 406 is cleared when a different page or a data packet of a different job is input. Therefore, the internal counter of the data counter 214 stores the total amount of encoded image data in the same page of the same job.
[0146]
A counter value comparison unit 407 compares the internal counter value of the data counter 406 with a register setting value Reg-PackCountValue for counter value output from the register setting unit 109, and uses the result as a count_over using the bus 115. Output to the packet generation unit 105.
[0147]
[Header information holding unit]
FIG. 3 is a block diagram of the header information holding unit 102. The header information unit 102 includes a header buffer 301 and a header comparison unit 302.
[0148]
The header buffer 301 holds the header information sent from the tile bus interface unit 101 while the first compression processing unit 103 and the second compression processing unit 104 compress the image data and the image additional information. And a header information necessary for compressing the image data and the image additional information to the first compression processing unit 103 and the second compression processing unit 104. In the present embodiment, ImageType for determining the amount of input image data and Mode and Q-Table Sel as parameters for determining a quantization table used for JPEG compression are provided to the first compression processing unit 103. Is output.
[0149]
Further, ImageType for determining the data amount of the input image additional information is output to the second compression processing unit 104.
[0150]
Further, the following contents are output to the header comparing unit for comparing the header information.
ImageType (3006)
Page_ID (3007)
Job_ID (3008)
Mode (3025)
Source_id (3018)
Char_Flag (3029)
Q-Table Sel (3030)
Z_DataType (3020)
The data comparing unit 302 compares predetermined information of the header information input from the tile bus interface unit 101 with the header information stored in the header buffer 301. Here, the header information input from the header buffer 301 to the data comparison unit 302 is not the header information of the data packet currently input from the tile bus interface unit 101, but the data packet input from the tile bus interface unit 101. Is the header information stored in the header buffer before. Therefore, the data comparison unit 302 compares the header information currently input to the header holding unit from the tile bus interface unit 101 with the header information previously stored in the header holding unit.
[0151]
The header comparison unit 302 outputs the comparison result Compare result to the packet generation unit 105.
[0152]
Further, the data comparison unit 302 outputs a counter reset signal to the data counter 214 in the first compression processing unit and the data counter 406 in the second compression processing unit. The counter reset signal is performed by referring to the Page ID and the Job ID among the header information input for comparison of the header information.
[0153]
That is, the page ID input from the tile bus interface unit 101 is compared with the page ID input from the header buffer 301, and similarly, the job IDs are compared. As a result, if one or both values of the page ID and the job ID are different, the data comparison unit 302 Outputs a counter reset signal.
[0154]
As a result, the data counter 214 in the first compression processing unit and the data counter 406 in the second compression processing unit receive data packets of different pages continuously and data packets of different Jobs continuously. In this case, a counter reset signal is output.
[0155]
[Quantization table selector]
FIG. 5 shows a block diagram of the quantization table selection unit 231 shown in FIG.
[0156]
The quantization table selector 231 includes one selector 502 and two lookup tables (LUTs) 501 and 503.
[0157]
The selector 502 selects a signal that refers to the LUT 501. In FIG. 5, one of the Q-Table Sel (3030) of the header information and the comparison result count_over output from the counter value comparison unit 215 is selected and input to the LUT 501. In the present embodiment, count_over is selected.
[0158]
The LUT 501 outputs a 4-bit Q-TableID (3028) with reference to the Mode (3025) 3 bits and the value of the count_over1 bit output from the selector 502. The value of the Q-TableID output from the LUT 501 uses Reg-Qmode [63: 0] input from the register setting unit 109.
[0159]
FIG. 6 is a diagram illustrating a relationship between an input value (address) and an output value (stored data) in the LUT 501.
[0160]
For example, when the Mode_000 is input to the LUT 501 and the count_over = 0 is input because the internal counter value of the data counter 214 is equal to or less than the register value Reg-JpegCountValue, the LUT 501 receives the Reg-Qmode in the Reg-Qmode [63: 0]. The value input from [3: 0] is output as Q-TableID (3028). An example of the value of the output Q-TableID (3028) is shown on the right side of FIG. 6, in which case 0000 is output as the Q-TableID.
[0161]
Here, the compression processing of the data packet progresses for a while by the first compression processing unit, and when the internal counter value of the data counter 214 exceeds the register value Reg-JpegCountValue, the counter value comparison unit 215 outputs count_over = 1. Note that the value of the register value Reg-JpegCountValue is set from the UI (operation unit) 2012, but may be automatically determined based on the document size.
[0162]
Then, as shown in FIG. 6, Mode = 000 and count_over = 1 are input to the LUT 501, so that the value 1000 of Reg-Qmode [35:32] is output as the Q-TableID.
[0163]
The Q-TableID output from the LUT 501 by the above operation is sent to the packet generation unit 105 via the bus 113, and is reflected in the Q-TableID (3028) of the header information.
[0164]
The LUT 503 outputs a 2-bit quantization table selection signal SEL to the quantization table unit 210 with reference to the value of the Q-TableID output from the LUT 501. The value of SEL output from the selector 502 uses Reg-QTField [31: 0] input from the register setting unit 109.
[0165]
FIG. 8 is a diagram for describing the configuration of Reg-QTField [31: 0]. Reg-QTField [31: 0] is composed of a QTField composed of a set of 2 bits, and 16 bits are stored in one register. It has a set of QTFields, and the selected QTField is the SEL.
bit [31:30] QTField used when Q-Table ID = 1111 is input
bit [29:28] Q-Field used when Q-Table ID = 1110 is input
bit [27:26] QTField used when Q-Table ID = 1101 is input
bit [25:24] QTField used when Q-Table ID = 1100 is input
bit [23:22] QTField used when Q-Table ID = 1011 is input
bit [21:20] QField used when Q-Table ID = 1010 is input
bit [19:18] QTField used when Q-Table ID = 1001 is input
bit [17:16] QTField used when Q-Table ID = 1000 is input
bit [15:14] QTField used when Q-Table ID = 0111 is input
bit [13:12] QTField used when Q-Table ID = 0110 is input
bit [11:10] QTField used when Q-Table ID = 0101 is input
bit [9: 8] QTField used when Q-Table ID = 0100 is input
bit [7: 6] QTField used when Q-Table ID = 0011 is input
bit [5: 4] QTField used when Q-Table ID = 0010 is input
bit [3: 2] QTField used when Q-Table ID = 0001 is input
bit [1: 0] QTField used when Q-Table ID = 0000 is input
FIG. 7 shows the relationship between the value of the Q-TableID (input address) input to the LUT 503 and the value of the SEL (storage data) output.
[0166]
For example, when Q = TableID = 0000 is input to the LUT 503, the LUT 503 outputs a value input from Reg-QTField [1: 0] in Reg-QTField [31: 0] as SEL. The right side of FIG. 7 shows an example of the output SEL value and the quantization table selected by the quantization table unit 210 at that time. In FIG. 7, when Q = TableID = 0000 is input, 00 is output as SEL, and the quantization table Q0 is selected.
[0167]
Also, as shown in FIG. 6, when count_over = 1 is input to the LUT 501 and the Q-TableID output from the LUT 501 changes to 1000, the SEL value output from the LUT 503 also changes from 00 to 10 and the quantization table Q2 is selected.
[0168]
The quantization table section 210 uses the quantization table selected by the SEL output from the quantization table selection section 213, and outputs a quantization value according to a request from the quantization section 203.
[0169]
According to the above description, when the data amount reaches the threshold value Reg-JobCountValue while the image of one page is compressed and stored in the RAM 2002, the quantization table newly set after that is set. , And compression is performed, and image input continues. However, already-compressed encoded data (data stored in the RAM 2002) before reaching the threshold value has been compression-encoded using the conventional quantization table, so that the data is read out and the packet is read out. The data is decompressed by the tile decompression unit 2103 via the image ring 2008, and recompression is performed using the newly determined quantization table.
[0170]
Further, in the present embodiment, the method of selecting the parameters of the quantization table has been described. However, the present invention is not limited to the quantization table, and other parameters such as the Huffman encoding unit 204 have exactly the same configuration. The parameter can be selected by a tile at an arbitrary position. For example, if the target pixel block is luminance data including RGB format and ND format, a Huffman coding table for luminance is used, and if density data including CMYK format and gray scale is used, color difference Is encoded by using a Huffman encoding table for encoding. The determination of the type (color space) of the pixel block of interest may be made in accordance with the Image Type included in the header of the data packet.
[0171]
Further, in the present embodiment, the parameter selection at the time of image compression has been described. However, this is not limited to the compression, and the same effect can be expected at the time of decompression with the same configuration as the present invention.
[0172]
In this embodiment, the quantization table is selected based on the values of the two tables (registers) corresponding to the information in the packet header. However, this is limited to the two tables (registers). Instead, the quantization table may be selected based on one or three or more tables (registers).
[0173]
Here, the scanner 2070 has a character determination function of evaluating a scanned image and adding information (character flag) indicating whether each pixel is an image constituting character data to each pixel. The configuration and algorithm for character determination are described in, for example, Japanese Patent No. 2872285.
[0174]
Also, information (Q-Table Sel (3030)) indicating whether each tile is a tile including character data is included in the header information of each tile. The Q-Table Sel is added by the scanner 2070 based on the character flag added for each pixel, and the addition condition can be selected from the following two conditions.
[0175]
Condition 1:
If all 1024 character flags in one tile, that is, in 32 × 32 pixels, are not “0”, that is, not a character, Q-Table Sel (3030) is set to “0”.
[0176]
In other cases, that is, when any one of 1024 character flags in one tile, that is, 32 × 32 pixels is “1”, that is, a character, the Q-Table Sel (3030) is set to “1”. . Condition 1 can be represented by the following logic circuit.
Q-Table Sel = (character flag of pixel 1) or (character flag of pixel 2) or ... or_ (character flag of pixel 1024)
Condition 2:
If any one of 1024 character flags in one tile, that is, 32 × 32 pixels, is not “0”, that is, if it is not a character, Q-Table Sel (3030) is set to “0”.
When all the 1024 character flags in one tile, that is, in 32 × 32 pixels are “1”, that is, all the characters are characters, the Q-Table Sel (3030) is set to “1”. In other cases, that is, condition 2 can be expressed by the following logic circuit.
Q-Table Sel = (character flag of pixel 1) and (character flag of pixel 2) and ... and_ (character flag of pixel 1024)
When Q-Table Sel (3030) = '0', Reg-QTField [31: 0] and Q-TableID are set so that a Q-Table for a photograph having a high compression ratio is selected. Image compression that emphasizes data saving can be performed.
[0177]
When Q-Table Sel (3030) = '1', image quality is set by setting Reg-QTField [31: 0] and Q-TableID so that a Q-Table for characters with high resolution is selected. Important image compression can be performed.
[0178]
Here, for the sake of explanation, the conditions 1 and 2 have been described using a simple logic circuit. However, the number of pixels whose character flag is “1” is counted, and the magnitude of the counter value and a preset threshold value are determined. By switching the Q-Table based on the result of the comparison, it is possible to set the condition with higher accuracy in accordance with the image.
[0179]
Here, for the sake of explanation, it has been described that the condition determination is made in 1024 of one tile, that is, 32 × 32 pixels. However, the present invention is not limited to this. With -TableSel, it is also possible to select a Q-Table for each JPEG processing unit.
[0180]
Note that, in the above-described embodiment, the description has been given mainly of compressing the image read from the scanner 2070 and storing the image in the RAM 2002. However, the apparatus according to the present embodiment is connected to the network as described above, For example, a personal computer or the like also functions as a printer. Therefore, in this case, as the data flow, the CPU 2001 interprets the print information (PDL data) received from the LAN controller 2010, compresses it even when it is expanded, and stores it in the RAM 2002. The compression process at this time is transmitted to the image processing unit 2149 via the image ring 2008 in the form of a packet (the rendering unit interface 2110 performs the same processing as the image input interface 2112), and transmits the compression result (packet). The data is received via the image ring 2008 and stored in the RAM 2002. While the compression process is being performed, the compressed image data stored in the RAM 2002 (it can be determined whether or not the data is compressed by checking the compression flag 3017 in the header) is again processed through the image ring 2008. The packet is transmitted to the unit 2149, decompressed by the tile decompression units 2103 (1) and 2103, and printed by the printer 2095 via the printer image processing unit 2115.
[0181]
As described above, according to the present embodiment, it is possible to select a quantization table for each tile based on quantization table selection information written in a packet header transmitted for each tile, and By setting a compression parameter corresponding to a feature for each tile, it is possible to perform image compression at a high compression rate without deteriorating image quality.
[0182]
【The invention's effect】
As described above, according to the present invention, it is possible to change parameters related to compression in units of pixel blocks such as tiles while facilitating the device configuration and device design. Further, it becomes possible to compress the image to the target size.
[Brief description of the drawings]
FIG. 1 is a block diagram of a tile compression unit 2106 according to the present embodiment.
FIG. 2 is a block diagram of a first compression processing unit 103;
FIG. 3 is a block diagram of a header information holding unit 102;
FIG. 4 is a block diagram of a second compression processing unit 104;
FIG. 5 is a block diagram of a quantization table selection unit 213.
FIG. 6 is a diagram illustrating a relationship between an input value and an output value in an LUT 501.
FIG. 7 is a diagram illustrating a relationship between an input value and an output value in an LUT 503.
FIG. 8 is a diagram showing a format of output data Reg-QTField [31: 0] of a register setting unit 109.
FIG. 9 is a configuration diagram of an entire network system including the digital multifunction peripheral according to the embodiment.
FIG. 10 is a block diagram of a controller of the digital multi-function peripheral in the embodiment.
FIG. 11 is a block diagram illustrating details of a controller of the digital multi-function peripheral in the embodiment.
FIG. 12 is a diagram showing a format of a data packet used in the controller 2000.
FIG. 13 is a diagram showing a format of a command packet used in the controller 2000.
FIG. 14 is a diagram showing a format of an interrupt packet used in the controller 2000.
FIG. 15 is a diagram showing the contents of a packet table used in the controller 2000.

Claims (19)

A division unit that divides image data input from the image input unit into predetermined pixel block units, an image compression unit that compresses and encodes the divided pixel blocks, and an image storage unit that stores compression-encoded compression encoded information, An image processing device that communicates as packet data between the division unit, the image compression unit, and the image storage unit,
The dividing unit, when dividing the input image data into predetermined pixel blocks, determines the attribute of the pixel block, and outputs information on the generated attribute as additional information as packet data together with the data of the pixel block,
The image compression unit determines a parameter when performing compression encoding according to the attribute information in the packet data output from the division unit, and compression-encodes data of a pixel block included in the packet data,
The image processing apparatus, wherein the storage unit stores the packet data compressed by the image compression unit. The image processing apparatus according to claim 1, wherein the image input unit is an image reading device. 2. The image processing apparatus according to claim 1, wherein the image input unit is a controller device that generates image information from a page description language. The image processing apparatus according to claim 1, wherein the image data and the additional information are compressed by the image compression unit using different compression methods. 2. The image processing apparatus according to claim 1, wherein the image compression processing of the image compression unit is JPEG compression, and the division unit divides the image into an integral multiple of a unit of JPEG compression. 6. The image processing apparatus according to claim 5, wherein the image compression unit has a plurality of quantization tables, and changes a compression ratio by selecting a quantization table to be used according to a parameter. The image processing apparatus according to claim 6, wherein the compression unit has a plurality of Huffman tables, and changes a compression ratio by selecting a Huffman table indicated by the parameter. A division unit that divides image data input from the image input unit into predetermined pixel block units, an image compression unit that compresses and encodes the divided pixel blocks, and an image storage unit that stores compression-encoded compression encoded information, A method for controlling an image processing device that communicates as packet data between the division unit, the image compression unit, and the image storage unit,
When dividing the input image data into predetermined pixel blocks, the division unit determines an attribute of the pixel block, and outputs information on the generated attribute as additional information as packet data together with the data of the pixel block. And
The image compression unit determines a parameter when performing compression encoding according to attribute information in the packet data output from the division unit, and compression-encodes data of a pixel block included in the packet data. ,
The method according to claim 1, wherein the storage unit stores the packet data compressed by the image compression unit. An image processing apparatus for compressing and storing image data,
Image dividing means for dividing the input image into predetermined pixel blocks, determining the attribute of the pixel block, and outputting information on the determined attribute as packet data together with the data of the pixel block as additional information,
Compression means for receiving the packet data of the pixel block obtained by the division and compressing and encoding the packet data;
Counting means for counting the amount of data compressed and compressed by the compression means and determining whether or not the amount has reached a predetermined amount;
Storage means for storing compression-encoded data as packet data,
The compression means,
An image processing apparatus, wherein compression encoding is performed based on additional information in packet data output from the image dividing unit and a parameter determined based on a determination result by the counting unit. The image processing apparatus according to claim 9, wherein the image divided by the image dividing unit is an image read by a document reading device. The image processing apparatus according to claim 9, wherein the image divided by the image dividing unit is an image generated from a page description language. The image processing apparatus according to claim 9, wherein the image compression unit performs JPEG compression, and the parameter is for determining a quantization step. 13. The image processing apparatus according to claim 12, wherein the size of the pixel block is an image size that is an integral multiple of an image processing unit (8 × 8 pixels) at the time of JPEG compression. 14. The image according to claim 12, wherein the compression unit includes a plurality of quantization tables each having a different quantization step for a high-frequency component with respect to a coefficient after DCT performed when performing JPEG compression. Processing equipment. The image processing apparatus according to claim 14, wherein the additional information includes information indicating whether the pixel is part of image data forming a character. The compression means has at least two of a first quantization table and a second quantization table, and the first quantization table has less image deterioration of high frequency components than the second quantization table, The image processing apparatus according to claim 15, wherein a compression ratio of the component is low. The additional information indicates that all the target pixel blocks are not characters, or indicates that at least one character data is mixed, or that the number of character data pixels is equal to or greater than the threshold number. 17. The image processing apparatus according to claim 16, wherein when there is, the first quantization table is selected. The additional information indicates that all the pixel blocks of interest are characters, or that one or more character data are not mixed, or that the number of character data pixels is equal to or less than the threshold number. 17. The image processing apparatus according to claim 16, wherein the second quantization table is selected when the second quantization table is set. A method of controlling an image processing device that compresses and stores image data,
An image dividing step of dividing the input image into predetermined pixel blocks, determining the attribute of the pixel block, and outputting information on the determined attribute as packet data together with the data of the pixel block as additional information;
A compression step of receiving packet data of the pixel block obtained by division and compressing and encoding the packet data;
A counting step of counting the amount of data compressed and encoded in the compression step, and determining whether or not the amount has reached a predetermined amount;
Storing the compression-encoded data as packet data in predetermined storage means,
The compression step includes:
A control method for an image processing apparatus, comprising: performing compression encoding based on additional information in packet data output from the image dividing step and a parameter determined based on a determination result by the counting unit.

Images