JP2006229688A - Image processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To generate packets with the smallest amount of data by selecting coding data/raw data, referring to the amount of coding data. <P>SOLUTION: The image processing device comprises an image input means which inputting image information; a division means for dividing into a predetermined unit the image information inputted from the image input means; two or more encoding means for carrying out, respectively coding processing of two or more data of a predetermined unit divided by the division means, while the image information divided by the division means, having two or more data; a comparison means for measuring the data amount of data of before and after being coded by the encoding means; and a data output means for selecting which of the data before coding or the data after coding and for outputting the selected data, on the basis of the comparison result of the comparison means. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus that divides image data read by an image reading apparatus or image data acquired via a network into arbitrary units, performs encoding processing, and stores the encoded image data in a memory. . In particular, it has auxiliary information indicating the state of the image data in addition to the image data. The auxiliary information is divided into arbitrary units in the same manner as the image data, and the divided auxiliary information is encoded. The present invention relates to an image processing apparatus.

  2. Description of the Related Art Conventionally, when storing image data in an image processing apparatus, the image processing apparatus is provided with an image processing apparatus in order to reduce the memory capacity for storing the image data and incidental information representing the state of the image data (hereinafter referred to as image additional information). In general, encoding processing is performed on the captured image data and image additional information, and the data amount is reduced, and then stored in a memory inside the image processing apparatus.

  At this time, the image data and the image additional information are subjected to encoding processing in consideration of the respective characteristics. For example, when image data is degraded, JPEG encoding processing is performed to increase the compression rate even when image degradation occurs, and image additional information that is meaningful for each bit is lost when it is decoded. In order to prevent this, a process such as performing a pack bits encoding process which is one of the run length encodings is performed.

These image data and image additional information are divided into predetermined units (hereinafter referred to as packets), and the divided image data and image additional information are subjected to the above encoding process for each packet (see, for example, Patent Document 1).
JP 2000-244925 A

  However, in the above conventional example, since the image data and the image additional information are encoded according to a predetermined setting, for example, if the image data and the image additional information are encoded for a packet in which a complex image exists, the compression rate is increased. In some cases, the amount of data before encoding does not increase.

  The present invention has been made paying attention to the above points, and is an image that generates a packet with the smallest data amount by referring to the encoded data amount and selecting encoded data / data before encoding. An object is to provide a processing apparatus.

  When the amount of encoded data is larger than the amount of data before encoding, the data before encoding (hereinafter referred to as raw data) is output to the subsequent stage instead of the encoded data. The amount of data to be performed can be made smaller than the amount of encoded data.

  However, when a plurality of encoding processes are performed on a plurality of data in a packet, the image additional information is encoded even if the encoded data of the image data is larger than the raw data amount as a result of encoding the image data. If the encoded data amount is small, the encoded data amount (added value of the encoded data amount of the image data and the encoded data amount of the image additional information) as the packet unit including the image data and the image additional information is the raw data amount ( There is a possibility that the state is smaller than the sum of the raw data amount of the image data and the raw data amount of the image additional information.

  On the other hand, as a result of encoding the image additional information, even if the encoded data of the image additional information is larger than the raw data amount, if the encoded data amount of the image data is small, the image data and the image additional information are included. As a packet unit, the amount of encoded data (added value of encoded data amount of image data and encoded data amount of image additional information) is raw data amount (addition of raw data amount of image data and raw data amount of image additional information) The value may be less than (value).

  At this time, as the compression rate of the packet, the data amount of the encoded data is smaller than the data amount of the raw data, so that it is possible to output the encoded data for both the image data and the image additional information. Then, in order to minimize the data amount of the packet to be output, the image data and the image additional information are independently compared between the encoded data amount and the raw data amount, and the encoded data amount is larger than the raw data amount. Is configured to output raw data.

With the above configuration, in the image processing apparatus of the present invention,
Image data and image additional information are both encoded data Image data is encoded data, image additional information is raw data Image data is raw data, image additional information is encoded data Image data and image additional information are both raw data A packet having a minimum data amount is always generated, and data having a minimum data amount for one page is generated.

  The present invention has been described above. If further described, in order to solve the above-described problem, the present invention is configured as follows (1).

(1) Image input means for inputting image information,
A dividing unit for dividing the image information input from the image input unit into predetermined units;
The image information divided by the dividing means has a plurality of data.
A plurality of encoding means for encoding each of a plurality of data in a predetermined unit divided by the dividing means;
Comparison means for comparing the amount of data before and after being encoded by the encoding means;
Data output means for selecting and outputting either the encoded data or the data before encoding according to the comparison result of the comparison means;
An image processing apparatus comprising:

  According to the present invention, in an image processing apparatus that encodes a plurality of types of data, that is, image data and image additional information, the encoded data amounts of the image data and the image additional information are independently determined as raw data data before the encoding process. When the amount of encoded data is larger than the amount of raw data, the raw data is used instead of the encoded data, so that a packet having the minimum amount of data is always generated. Data with a minimum amount of data is generated.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to examples.

[First embodiment]
The apparatus of the present invention and its operation will be described in detail below.

[hardware]
Overall Configuration FIG. 12 shows an overall configuration diagram.
Controller Unit (2000) is connected to Scanner (2070), which is an image input device, and Printer (2095), which is an image output device. This is a controller for inputting / outputting image information and device information and developing images of PDL data.

  A 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 connected to a system bus bridge (2007).

  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), a sub bus switch (2128), and an IO bus 2 (2129). ), The image ring interface 1 (2147) and the image ring interface 2 (2148) are connected.

  The sub bus switch (2128) is a second bus switch to which the image DMA1 (2130), the image DMA2 (2132), the font expansion unit (3134), the sort circuit (2135), and the bitmap trace unit (2136) are connected. Then, the memory access request output from these DMAs is arbitrated to connect to the system bus bridge.

  A RAM (2002) is a system work memory for operating the CPU (2001), and is also an image memory for temporarily storing image data. In this embodiment, which is controlled by the RAM controller (2124), an example in which a direct RDRAM is employed is shown.

  A ROM (2003) is a boot ROM, which stores a system boot program. It is controlled by the ROM controller (2125).

  The image DMA1 (2130) is connected to the image compression unit (3131), controls the image compression unit (2131) based on information set through the register access ring (2137), and is on the RAM (2002). In this embodiment, in which uncompressed data is read, compressed, and compressed data is written back, an example in which JPEG is adopted as a compression algorithm is shown.

  The image DMA2 (2132) is connected to the image expansion unit (2133), controls the image expansion unit (2133) based on the information set through the register access ring (2137), and is on the RAM (2002). In this embodiment, in which compressed data is read, decompressed, and data is written back after decompression, JPEG is used as the decompression algorithm.

  The font decompression unit (2134) is a compressed font data stored in the ROM (2003) or RAM (2002) based on the font code included in the PDL data transferred from the outside via the LAN interface (2010) or the like. Stretching. In this embodiment, an example in which the FBE algorithm is adopted is shown.

  The sort circuit (2135) is a circuit that rearranges the order of the objects in the display list generated at the stage of developing the PDL data.

  The bitmap trace circuit (2136) is a circuit that extracts edge information from bitmap data.

  The IO bus 1 (2127) is a kind of internal IO bus, and includes a standard USB bus controller, a USB interface (2138), a general-purpose serial port (2139), an interrupt controller (2140), and a GPIO interface (2141). Connected. The IO bus 1 includes a bus arbiter (not shown).

  An operation unit I / F (2006) is an operation unit (UI) (2012) and an interface unit, and outputs image data to be displayed on the operation unit (2012) to the operation unit (2012). Also, it plays a role of transmitting information input by the system user from the operation unit (2012) to the CPU (2001).

  The IO bus 2 (2129) is a kind of internal IO bus, and is connected to the general-purpose bus interfaces 1 and 2 (2142) and the LAN controller (2010). The IO bus 2 includes a bus arbiter (not shown).

  The general-purpose bus interface (2142) is a bus bridge that includes two identical bus interfaces and supports a standard IO bus. In this embodiment, an example in which a PCI bus (2143) is employed is shown.

  An HDD (2004) is a hard disk drive that stores system software and image data. It is connected to one PCI bus (2143) via the disk controller (2144).

  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) is connected to the public line (2051) and inputs / outputs information.

  The image ring interface 1 (2147) and the image ring interface 2 (2148) connect the system bus bridge (2007) and the image ring (2008) for transferring image data at high speed, and the compressed data after tiling is stored in RAM ( 2002) and a tile controller (2149).

  The image ring (2008) is configured by a combination of a pair of unidirectional connection paths (image ring 1 and image ring 2). The image ring (2008) includes a tile expansion unit (2103), a command processing unit (2104), a status via the image ring interface 3 (2101) and the tile image interface 4 (2102) in the tile image processing unit (2149). The processing unit (2105) and the tile compression unit (2106) are connected. In this embodiment, an example is shown in which two sets of tile expansion units (2103) and three sets of tile compression units are mounted.

  The tile expansion unit (2103) is connected to the tile bus (2107) in addition to the connection to the image ring interface, and expands the compressed image data input from the image ring and transfers the compressed image data to the tile bus (2107). It is a bridge. In this embodiment, an example in which a decompression algorithm based on the JPEG and Packbits methods is employed will be described.

  The tile compression unit (2106) is connected to the tile bus (2107) in addition to the connection to the image ring interface, compresses the uncompressed image data input from the tile bus, and transfers the compressed image data to the image ring (2008). It is a bridge. In the present embodiment, an example in which a compression algorithm based on JPEG and Packbits is employed as in the tile expansion unit is shown.

  The command processing unit (2104) is connected to the register setting bus (2109) in addition to the connection to the image ring interface, and receives a register setting request issued from the CPU (2001) input via the image ring. Write to the corresponding block connected to (2109). Further, based on a register read request issued by the CPU (2001), information is read from the corresponding register via the register setting bus. The image is transferred to the image ring interface 4 (2102).

  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 it to the image ring interface 4.

  In addition to the above blocks, the following functional blocks are connected to the tile bus (2107).

  Rendering unit interface (2110), image input interface (2112), image output interface (2113), multi-value conversion unit (2119), binarization unit (2118), color space conversion unit (2117), image rotation unit (2030) ), A resolution conversion unit (2116).

  The rendering unit interface (2110) is an interface for inputting a bitmap image generated by a rendering unit described later. The rendering unit and the rendering unit interface are connected by a general video signal (2111). The rendering unit interface has a connection to the memory bus (2108) and the register setting bus (2109) in addition to the tile bus (2107), and the input raster image is set via the register setting bus, and is a predetermined one. The structure is converted into a tile image by the above method, and at the same time, the clock is synchronized and output to the tile bus (2107).

  The image input interface (2112) receives raster image data subjected to correction image processing by a scanner image processing unit (2114) to be described later, and inputs the raster image data to the tile image by a predetermined method set via the register setting bus. The structure conversion and clock synchronization are performed and output to the tile bus (2107).

  The image output interface receives tile image data from the tile bus, converts the structure into a raster image and changes the clock rate, and outputs the raster image to the printer image processing unit (2115).

  An image rotation unit (2030) rotates image data.

  A resolution converter (2116) changes the resolution of the image.

  A color space conversion unit (2117) converts the color space of the color and gray scale image.

  A binarization unit (2118) binarizes a multi-value (color, gray scale) image.

  A multi-value conversion unit (2119) converts a binary image into multi-value data.

  The external bus interface unit (2120) sends write / read requests issued by the CPU (2001) via the image ring interfaces 1, 2, 3, 4, command processing unit, and register setting bus to the external bus 3 (2121). This is a bus bridge that converts and outputs to In this embodiment, the external bus 3 (2121) is connected to the printer image processing unit (2115) and the scanner image processing unit (2114).

  The memory control unit (2122) is connected to the memory bus (2108), and in accordance with the request of each image processing unit, the image data is transferred to the image memory 1 and the image memory 2 (2123) by preset address division. Write, read, and refresh operations are performed as necessary. In this embodiment, an example in which an SDRAM is used as an image memory is shown.

  The scanner image processing unit (2114) performs correction image processing on the image data scanned by the scanner (2070) as an image input device.

  The printer image processing unit performs corrected image processing for printer output, and outputs the result to the Printer (2095).

  The rendering unit (2060) develops the PDL code or the intermediate display list into a bitmap image.

[Whole system]
FIG. 11 shows a configuration diagram of the entire network system of the present invention.

  An apparatus 1001 according to the present invention includes a scanner and a printer, and can output an image read from the scanner to a local area network (1010) (hereinafter referred to as a LAN) and print out an image received from the LAN. Further, the image read from the scanner can be transmitted to the PSTN or ISDN (1030) by a FAX transmission means (not shown), and the image received from the PSTN or ISDN can be printed out by the printer. Reference numeral 1002 denotes a database server which manages binary images and multi-valued images read by the apparatus (1001) of the present invention as a database.

  Reference numeral 1003 denotes a database client of the database server (1002), which can browse / search image data stored in the database (1002).

  An e-mail server 1004 can receive an image read by the apparatus (1001) of the present invention as an e-mail attachment. Reference numeral 1005 denotes an e-mail client that can receive and browse e-mail received by the e-mail server (1004) and send e-mail.

  Reference numeral 1006 denotes a WWW server that provides an HTML document to a LAN, and an HTML document provided by the WWW server can be printed out by the apparatus (1001) of the present invention.

  Reference numeral 1007 denotes a router that connects the LAN (1010) to the Internet / intranet (1012). The database server (1002), WWW server (1006), e-mail server (1004), and apparatus similar to the apparatus (1001) of the present invention are connected to the Internet / intranet as 1020, 1021, 1022, and 1023, respectively. ing. On the other hand, the apparatus (1001) of the present invention can transmit and receive with the FAX apparatus (1031) via the PSTN or ISDN (1030).

  A printer (1040) is also connected to the LAN, and is configured to print out an image read by the apparatus (1001) of the present invention.

[Tile image (packet) format]
In the System Controller Unit (2000) according to the present invention, image data, commands from the CPU (2001), and interrupt information issued from each block are transferred in a packetized form. In this embodiment, three different types of packets are used: a data packet shown in FIG. 13, a command packet shown in FIG. 14, and an interrupt packet shown in FIG.

Data packet (Figure 13)
In this embodiment, the image data is handled by being divided into image data (3002) in units of Tile of 32 pixels x 32 pixels. Necessary header information (3001) and image additional information (3003) are added to the image in units of tiles to form a data packet. Padding data for adjusting each data length to a predetermined length is added between the image data and the image additional information and after the image additional information.

  Information included in the header information (3001) will be described below.

The packet is classified into a data packet, a command packet, and an interrupt packet according to the value of the packet type ID (3023) in the packet type (3004) in the header information (3001). In this embodiment, for the Packet Type ID 3 bits,
001b or 101b: Data packet 010b: Command packet 100b: An interrupt packet is allocated.

  Also, the Pckt Type (3004) includes a Repeat Flag (3022), and image data of the data packet, image additional information (3003), and predetermined information in the header information (3001) are output to the system control unit 2150. When the data packet matches, the Repeat Flag (3022) is set to 1, and only the header information 3001 is output to the system control unit 2150.

  When a data packet in which 1 is set in the Repeat Flag (3022) is input to the system control unit 2150, the image data and image additional information of the data packet are the image data and image additional information of the data packet input immediately before. It is determined that it is the same as the process. As a result, the amount of data transferred between the system control unit 2150 and the image processing unit 2149 is reduced.

  Chip ID (3005) indicates the ID of a target chip that transmits a packet. When a plurality of image processing units and system control units are connected, a data packet is sent to a desired block by indicating a target that transmits the packet. Used for identification for output.

Image Type (3006) indicates the type of image data. In this embodiment, the image data type is defined as follows using the upper 2 bits of the 8 bits of Image Type.
00b: 1 pixel is represented by 1 bit 01b: 1 pixel is represented by 8 bits of data 1 component 10b: 1 pixel is represented by 3 components of 8 bits of data 11b: 1 pixel is represented by 4 components of 8 bits of data, 32 in total Represented in bits

  Page ID (3007) indicates a page including a data packet, and Job ID (3008) stores a Job ID for management by software.

  Packet ID Y-coordinate (3009) and X-coordinate (3010) indicate the position of the data packet on the page, and the system control unit 2150 allows each data packet to be output regardless of the order output from the image processing unit 2149. You can know the positional relationship.

  In Process Instruction (3011), an operation mode for performing processing in each unit of the image processing unit is designated. The Process Instruction (3011) is left-justified and set in the processing order of each processing unit every 8 bits, and each processing unit passes the processed processing instruction to the left by 8 BitShift and passes it to the next processing unit. In Process Instruction (3011), eight sets of UnitID (3024) and Mode (3025) are stored, UnitID (3024) specifies the specification of each processing unit, and Mode (3025) specifies the operation mode in each processing unit. . Thereby, it is possible to perform image processing with eight units continuously in one packet.

  Packet Byte Length (3012) indicates the total number of bytes of the packet.

  Image Data Byte Length (3015) and Z Data Byte Length (3016) represent the number of bytes of image data and the number of bytes of image additional information, respectively. Image Data Offset (3013) and Z Data Offset (3014) are from the head of each packet. It represents the Offset value up to the part where data is stored. Note that the byte lengths of the image data and the image additional information shown here do not include the padding data.

  Image Data (3026) and Zdata (3027) in the Compress Flag (3017) each indicate whether the image data and the image additional information are compressed or uncompressed with one bit.

  Information when the image data is compressed is stored in the Q-Table ID (3028) in the Compress Flag (3017). In this embodiment, as will be described later, when the upper 2 bits of the Image Type (3006) are 01b, 10b, 11b, that is, when one pixel is represented by a plurality of bits, the image data is compressed by the JPEG method. , Q-Table ID (3028) stores information indicating the quantization table used for compression.

  The tile compressing unit 2106 and the tile decompressing unit 2103 refer to the value stored in the Q-Table ID (3028), and performs compression / careful processing using the corresponding quantization table.

  Source ID (3018) indicates a source from which image data and image additional information are generated.

  Misc (3019) stores other information. In this embodiment, Char-flag (3029) representing additional information of the entire data packet, which is auxiliary information of the image additional information, and Q-Table Sel (3030) for switching the quantization table for each packet during JPEG compression. Is prepared.

  Ztype (3020) indicates the effective bit width of the image additional information. Image additional information other than the bits indicated by Ztype (3020) is invalid information.

  The Ignor Flag (3022) is used as a trigger for performing re-encoding in a re-encoding process described later. When the system control unit 2150 receives the data packet in which the Ignor Flag (3022) is set, the system control unit 2150 clears the RAM 2002 and starts acquiring image data (data packet) again.

  A value (thumbnail value) representing the image data of the data packet is stored in Thumbnail Data (3021). In this embodiment, a maximum of four values can be stored for each component in the thumbnail data (3021).

Packet Table (Fig. 16)
FIG. 16 shows a state where data packets are stored in the RAM (2002). Each data 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 value of the Table, the packet start address (6002) and the packet Byte Length (6005) are obtained.

Packet Address Pointer (27 bits) + 5b00000 = Packet head Address
Packet Length (11 bits) + 5b00000 = Packet Byte Length

  When one data packet is input to Packet Address Pointer, only the value of Byte Length of the packet is added, and indicates the first address of the next packet.

  Note that the packet table (6001) and the chain table (6010) are not divided.

The Packet Table (6001) is always arranged in the scanning direction, and Yn / Xn = 000 / 000,000 / 001,000 / 002,. . . . It is lined up in this order. The entry of the packet table (6001) uniquely indicates one tile. In addition, the following Entry of Yn / Xmax becomes Yn + 1 / _{X 0.}

  Here, Yn and Xn are the same values as Packet ID Y-coordinate (3009) and X-coordinate (3010) of the header information (3001).

  When a packet in which the Repeat Flag (3002) in the header information (3001) is set is input, the packet is not written on the Memory, and is the same Packet Address Pointer as the previous Entry in the Entry of the Packet Table. , Packet Length is stored. A single packet data is pointed to by two table entries. In this case, Repeat Flag (6003) of the second Table Entry is set.

  When the packet is divided into a plurality by Chain DMA, the Divide Flag (6004) is set, and the Chain Table number (6006) of the Chain Block containing the head portion of the packet is set.

  The entry of the chain table (6010) is composed of a chain block address (6011) and a chain block length (6012), and 0 is stored in both the address and the length of the last entry of the table.

Command Packet Format (Figure 14)
This Packet Format is used to access the register setting bus (2109). By using this packet, the COU (2001) can also access the image memory (2123).

  The Chip ID (4004) stores an ID representing the image processing unit (2149) that is the transmission destination of the command packet.

  Page ID (4007) and Job ID (4008) store Page ID and Job ID for management by software.

  Packet ID (4009) is expressed in one dimension. Only X-coordinate of Data Packet is used.

  The packet byte length (4010) is fixed to 128 bytes.

  The packet data part (4002) can store a maximum of 12 commands, with a set of address (4011) and data (4012) as one command. The type of command for writing or reading is indicated by CmdType (4005), and the number of commands is indicated by Cmdnum (4006).

Interrupt Packet Format (FIG. 15)
This PacketFormat is for notifying an interruption from the image processing unit (2149) to the CPU (2001). When the status processing unit (2105) transmits the interrupt packet, the status processing unit (2105) must not transmit the interrupt packet until the next transmission is permitted.

  The packet byte length (5006) is fixed to 128 bytes.

  The packet data part (5002) stores status information (5007) of each internal module of the image processing part (2149). The status processing unit (2105) can collect the status information of each module in the image processing unit (2149) and collectively send it to the system control unit (2150).

  The Chip ID (5004) is an ID representing the system control unit (2150) that is the destination of the Interrupt Packet, and the IntChip ID (5005) is an ID representing the image processing unit (2149) that is the source of the Interrupt Packet. Is stored.

[Tile compression section]
FIG. 1 is a block diagram of the tile compression unit 2106 in FIG.

  In FIG. 1, reference numeral 101 denotes a tile bus interface unit, which performs handshaking with the tile bus 2107, acquires a data packet input from the tile bus 2107, and outputs each data to each processing block connected to the subsequent stage. .

  Further, the tile bus interface unit 101 analyzes the header information (3001) sent from the tile bus. If there is a contradiction in the header information, the tile bus interface unit 101 outputs an interrupt signal to a register setting unit 109 described later, and then outputs a reset signal ( The acquisition of the data packet from the tile bus is stopped until (not shown) is input.

  If there is no contradiction in the header information (3001), the header information is output to the header information holding unit 102 connected in the subsequent stage, and then image data and image additional information are acquired from the tile bus. For the acquired image data and image additional information, the first compression processing unit 103 (in this embodiment performs compression processing by the JPEG method) or the second compression processing unit 104 (in this embodiment) refers to the header information Image Type (3006). In the example, the compression process is performed by the Packbits method), and image data and image additional information are output to each.

  Specifically, when the upper 2 bits of the Image Type in the header information are 00b (image data is 1 bit), the image data is output to the second compression processing unit 204. At this time, the image additional information is discarded by the tile bus interface unit 101.

  If the upper 2 bits of ImageType are other than 00b, the image data is output to the first compression processing unit 103 and the image additional information is output to the second compression processing unit 104. However, when Ztype (3020) indicates that there is no image additional information, the image additional information is not output to the second compression processing unit, and the compression processing by the second compression processing unit is not performed.

  Reference numeral 102 denotes a header information holding unit that holds header information while the first compression processing unit 103 and the second compression processing unit 104 perform compression processing of image data and image additional information. In addition, the header information holding unit 102 outputs information necessary for compression processing from the stored header information to the first compression processing unit and the second compression processing unit.

  Reference numeral 103 denotes a first compression processing unit, which performs compression using the JPEG method in this embodiment.

  Note that it is also possible to output image data to the packet generation unit without performing compression processing in the first compression processing unit, depending on the setting.

In this embodiment, the first compression processing unit generates first image data to be output to the system control unit 2150 and second image data for performing re-encoding processing by a re-encoding unit to be described later. Two encoded image data are output to the packet generator 105. Further, when there is a request for raw data output from the packet generation unit 105, the first compression processing unit outputs image data before compression processing to the packet generation unit 105.
The first compression unit 103 will be described in detail later.

  Reference numeral 104 denotes a second compression processing unit, which performs compression according to a compression method with no information loss, specifically, a pack bits method in this embodiment.

  In the second compression processing unit, when the upper 2 bits of the Image Type (3006) of the header information is 00b (image data is 1 bit), the image data is compressed, and the upper 2 bits of the Image Type (3006) is 00b. Otherwise, the image additional information is compressed. The first compression processing unit does not operate when image data compression processing is performed in the second compression processing unit.

  Depending on the setting, it is also possible to output image additional information or image data to the packet generation unit without performing compression processing in the second compression processing unit.

  In this embodiment, the second compression processing unit also generates first image additional information output to the system control unit 2150 and second image additional information for performing re-encoding processing by a re-encoding unit described later. Then, the two encoded image additional information is output to the packet generator 105. Further, when there is a request for raw data output from the packet generation unit 105, the second compression processing unit outputs image additional information or image data before the compression processing to the packet generation unit 105.

  The second compression unit 104 will also be described in detail later.

  Reference numeral 105 denotes a packet generation unit that acquires header information from the header information holding unit 102, and encoded image data and image additional information from the first compression processing unit 103 and the second compression processing unit 104. After the predetermined value is set, the data packet shown in FIG. 13 is generated and output to the image ring output unit 107 and the re-encoding unit 106.

  Depending on the encoded image data and the amount of encoded image additional information acquired from the first compression processing unit 103 and the second compression processing unit 104, the packet generation unit 105 may encode the image data and the image encoded for the subsequent stage. Instead of the additional information, image data and image additional information (hereinafter referred to as raw image data and raw image additional information) that are not encoded are output.

  For this purpose, the packet generation unit 105 requests the first compression processing unit 103 and the second compression processing unit 104 to output raw image data and / or raw image additional information.

  At this time, the following three conditions can be considered as conditions for requesting the first compression processing unit 103 and the second compression processing unit 104 to output raw image data or raw image additional information.

  1. When the data amount of the encoded image data acquired from the first compression processing unit 103 is larger than the raw image data amount (compression ratio is 1 or less), a raw data output request is made to the first compression processing unit 103, and the raw image data To get.

  2. When the data amount of the encoded image additional information acquired from the second compression processing unit 104 is larger than the data amount of the raw image additional information (compression ratio is 1 or less), a raw data output request is made to the second compression processing unit 104. The raw image additional information is acquired.

  3. The added value of the data amount of the encoded image data acquired from the first compression processing unit 103 and the data amount of the encoded image additional information acquired from the second compression processing unit 104 is the data amount of the raw image data and the raw image additional information. Is greater than the added value, a raw data output request is issued to the first compression processing unit 103 and the second compression processing unit 104, and raw image data and raw image additional information are acquired.

  Under the above conditions 1 and 2, the image data and the image encoding information output encoded data or raw data independently. That is, the image data is selected with reference to only the data amount of the encoded image data to output raw data or encoded data.

  Similarly, the image additional information selects whether to output raw data or encoded data by referring only to the data amount of the encoded image additional information.

  When data to be output is selected under the conditions 1 and 2, the data amount of the packet can be minimized.

  The condition 3 is that the total amount of the encoded image data and the image additional information is equal to that of the image data or the image additional information, even if one of the encoded image data or the image additional information is larger than the raw data. If it is smaller than the total amount of raw data, encoded data can be output. In order to perform re-encoding in a re-encoding unit, which will be described later, in the configuration of the present embodiment, encoded image data is necessary. Therefore, a packet may be generated using the three conditions.

  In this embodiment, in order to minimize the amount of encoded data, it is selected whether to output encoded data or raw data under the conditions 1 and 2.

  A re-encoding unit 106 stores the data packet sent from the packet generation unit 105 in a local memory in the re-encoding unit. Further, in accordance with a request from the image ring output unit 107, the data packet stored in the local memory is sent to the image ring output unit, and the image data included in the data packet is recompressed and stored in the local memory again.

  The re-encoding unit 106 will be described in detail later using a block diagram.

  An image ring output unit 107 outputs a data packet sent from the packet generation unit 105 and the re-encoding unit 106 to the image ring interface 4 (2102). Further, the image ring output unit adds the byte amount of the data packet sent to the RAM 2002 with reference to the header information of the data packet, and when the added value exceeds a predetermined value, the Ignore of the header information (3001) of the data packet. Flag (3022) is set and output to the system control unit 2150, and a code amount control signal is output to the re-encoding unit 106 and the first compression processing unit 103.

  Reference numeral 109 denotes a register setting unit for performing setting related to processing in the tile compression unit 2106. In order to cause the tile compression unit 2106 to perform predetermined compression processing, it is necessary to set a predetermined value in the register setting unit 109 There is.

  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. 14, and are sent from the command processing unit to the tile compression unit 2106 via the register setting bus 2109.

  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 processing determined by referring to these setting values.

  In addition to setting a value to the register setting unit using a command packet, it is also possible to output a setting value held by the register setting unit to the system control unit 2150 using a command packet.

  Further, the register setting unit has a register for storing header information. When an interrupt signal is input from the tile bus interface unit 101, the header information is fetched from the tile bus interface 101 and set in the register, and the status processing unit An interrupt signal notifying that an interrupt has occurred and a status signal indicating an error state are output to 2105.

  A register setting bus interface unit 108 is a block for performing format conversion between the register setting bus 2109 and the register setting unit 109.

[First compression processing unit]
FIG. 2 shows a block diagram of the first compression processing unit 103.

  In FIG. 2, reference numeral 201 denotes a first data buffer for storing image data sent from the tile bus interface unit 101. When a predetermined amount of data is sent, the DCT conversion unit 202 connected to the subsequent stage is sent to the first data buffer. The data is output according to a predetermined order.

  The first data buffer receives the header information ImageType (3006) from the header information holding unit 102, and the order of the data output from the first data buffer to the DCT conversion unit 202 is controlled by the image type. The

  When there is a request from the packet generation unit 105, the stored image data is output to the packet generation unit 105 in a predetermined order.

  When a data is input from the first data buffer 201, a DCT conversion unit 202 performs discrete cosine transform and converts it into frequency component data. At this time, the DC component value generated by the discrete cosine transform is output to the thumbnail generation unit 212 (to be described later) together with the latch signal.

  The DCT transform unit outputs a latch signal and a DC component value each time discrete cosine transform is performed to generate a thumbnail value to be described later.

  A quantization unit 203 quantizes the frequency component data output from the DCT conversion unit 202 using predetermined quantization values. A value used for quantization is input from a quantization table described later. The quantization table used for quantization is determined by referring to the header information.

  Reference numerals 204 and 205 denote code amount control units that control data so that the code amount is reduced when the frequency component data quantized by the quantization unit 203 is encoded by a Huffman encoding unit described later. .

  In this embodiment, the code amount is controlled by dividing the data input to the code amount control unit by the power of 2 (n = 0, 1,...). Note that the value of n can be set independently for each of the code amount control units 204 and 205 based on the register value of the register setting unit 109 and the code amount control signal input from the image ring output unit 107.

  Reference numerals 206 and 207 denote Huffman encoding units that perform predetermined encoding processing on the data output from the code amount control units 204 and 205 to generate encoded data.

  Reference numerals 208 and 209 denote a second data buffer and a third data buffer, which are buffers for storing the encoded data encoded by the Huffman encoding units 206 and 207, respectively.

  When the second data buffer 208 finishes inputting the encoded data from the Huffman encoding unit 206, the second data buffer 208 outputs the encoded data amount as Data Byte Length 1 and stores it in response to the request of the packet generation unit 105 Output data.

  Similarly, when the third data buffer 209 also finishes inputting the encoded data from the Huffman encoding unit 207, the third data buffer 209 outputs the amount of data encoded as Data Byte Length 2, and responds to the request from the packet generation unit 105. Output the stored data.

  A thumbnail generation unit 212 generates a thumbnail value for each tile using a DC component value output in synchronization with the latch signal from the DCT conversion unit 202 and outputs the thumbnail value to the packet generation unit 105.

  The generated thumbnail value is output as Thumbnail Data to the packet generation unit 105, and stored in Thumbnail Data (3021) of the header information in a predetermined format.

  Reference numeral 210 denotes a quantization table, which stores quantization values for the quantization unit 203 to perform quantization. A plurality of quantization tables are stored in the quantization table of the present embodiment, and a predetermined quantization table is selected by a selection signal input from a quantization table selection unit described later, and the quantization unit 203 performs quantization. Output the value.

  Reference numeral 213 denotes a quantization table selection unit, which outputs a quantization table selection signal to the quantization table 210 to select a predetermined table from a plurality of tables stored in the quantization table.

  The quantization table selection unit 213 receives ImageType (3006), Mode (3025), Char-flag (3029), and Q-Table Sel (3030) from the header information holding unit 202, and the quantization table selection unit The quantization table to be used is determined from these header information.

  When the quantization table to be used is determined, the quantization table selection unit outputs a quantization table selection signal to the quantization table so as to select a predetermined quantization table. Further, the Q-Table ID representing the selected quantization table is output to the packet generation unit 105.

  Data Byte Length 1, Data Byte Length 2, Q-Table ID, and Thumbnail Data described above are referred to when the packet generation unit 105 generates header information.

[Second compression processing unit]
FIG. 9 shows a block diagram of the second compression processing unit 104.

  In FIG. 9, reference numeral 901 denotes a fourth data buffer for storing image data or image additional information sent from the tile bus interface unit 101. For the data mask unit and the data amount detection unit connected in the subsequent stage, Therefore, data is output.

  Further, when there is a request from the packet generation unit 105, the stored image additional information or image data is output to the packet generation unit 105 in a predetermined order.

  Reference numerals 902 and 903 denote data mask units, which perform data mask processing for fixing all predetermined bits to 1 or 0 for the image additional information output from the data buffer 4.

  The input image additional information stores image information of each pixel such as whether the pixel is a pixel included in the character area, whether it is a chromatic color pixel or an achromatic color pixel. In the data mask section, the predetermined information of the image additional information is fixed to 1 or 0. As a result, the data is input to the packbits encoding unit at the subsequent stage.

FIG. 10 shows data mask processing by the data mask unit in this embodiment. In this embodiment, the image additional information is composed of 8 bits, and each bit represents the following information.
bit 7-5: Fixed value bit 4-3: Character internal signal (represents character thickness)
bit 2: Character signal (indicates whether or not the pixel is included in the character)
bit 1: Achromatic signal (represents whether the pixel is achromatic or connected)
bit 0: halftone character (indicates whether the pixel is a character inside a halftone dot)

The data mask unit performs the following five stages of data mask processing on these pieces of information of the image additional information.
Mask level 0 No mask (outputs additional image information input to the data mask part)
Mask level 1 When the character signal (bit2) is 0, the achromatic signal (bit1) is forcibly set to 0. Mask level 2 When the character signal is 0, the character internal flag (bits 4 and 3) is forcibly set to 0. Mask level 3 When the character signal is 0, the character flag (bit 0) in the network is forced to 0. Mask level 4 Bits other than the character signal are set to 0.

  By performing the above-described data mask processing on the image additional information, the same value is frequently output continuously from the data mask information output from the data mask unit. When encoding is performed by the encoding unit, the amount of data after encoding can be reduced as compared to the case where image additional information before data mask processing is encoded.

  In this embodiment, the mask level setting performed by the data mask unit is set independently for each of the data mask units 902 and 903 based on the register value of the register setting unit 109 and the code amount control signal input from the image ring output unit 107. can do.

  Note that the data mask processing described above in the present embodiment is performed only when image additional information is input to the second compression processing unit 104. When 1-bit image data is input to the second compression processing unit 104, mask processing is not performed.

  Numerals 904 and 905 denote pack bits encoding units, which generate encoded data by performing pack bit encoding processing on the image additional information output from the data mask units 902 and 903.

  Reference numerals 906 and 907 denote a fifth data buffer and a sixth data buffer, which are buffers for storing the encoded data encoded by the Packbits encoding units 904 and 905, respectively.

  When the fifth data buffer 906 finishes inputting the encoded data from the Packbits encoding unit 904, the fifth data buffer 906 outputs the stored data in response to the request from the packet generation unit 105.

  Similarly, the sixth data buffer 907 also outputs the stored data in response to the request from the packet generation unit 105 when the input of the encoded data from the Packbits encoding unit 905 is completed.

  Reference numeral 908 denotes a data amount detection unit. When data mask processing is performed on the input image additional information at each mask level shown in FIG. 10, the image additional information subjected to the mask processing at each mask level is displayed. On the other hand, this is a block for calculating the amount of data when pack bits processing is performed by the pack bits encoding unit 904 or 905.

  Note that the data amount detection unit 908 does not perform pack bit encoding but only calculates the amount of encoded data when the pack bit encoding process is performed. From the data amount detection unit, five data amount values of mask levels 0 to 4 are output as Zdata Byte Length to the packet generation unit 105.

[Recode part]
FIG. 8 shows a block diagram of the re-encoding unit 106.

  In FIG. 8, reference numeral 801 denotes a data buffer 1 for temporarily storing a data packet sent from the packet generator 105, and has a capacity for one data packet. The data buffer 1 stores packet data generated using the encoded data stored in the data buffer 3 (209) in FIG.

  The data packet stored in the data buffer 1 is output to the memory controller 802 to be stored in the memory in the re-encoding unit.

  Reference numeral 802 denotes a memory controller that controls read / write operations to the memory. The above-mentioned data buffer 1 (801), data buffer 2 (804) and data buffer 3 (808) described later are connected to the memory controller 802, and a predetermined address of the memory is set by a read / write request of each data buffer. Read / write is performed.

  Reference numeral 803 denotes a page memory in the re-encoding unit, and a data packet input from the packet generation unit 105 is stored in the memory 803 via the data buffer 1 (801) and the memory controller 802.

  The memory 803 outputs data stored in the memory to the data buffer 2 (804) via the memory controller in response to a request from the memory controller 802, and performs re-encoding processing described later on the data. The obtained data is stored again in the memory via the memory controller.

  Reference numeral 804 denotes a data buffer 2 (804) for outputting a data packet stored in the memory 803 to the image ring output unit in response to a request from the image ring output unit 207, and has a capacity for one data packet.

  The data packet stored in 804 is sent to the image ring output unit, header information is sent to data packet 3, and image additional information and image data are sent to data mask unit 809 and Huffman decoding unit 805 for re-encoding processing. Sent.

  A Huffman decoding unit 805 performs decoding processing on the encoded image data sent from the data buffer 2, generates quantized frequency component data, and sends the quantized frequency component data to the code amount control unit 806.

  Reference numeral 806 denotes a code amount control unit which controls data so that the code amount is reduced with respect to the quantized frequency component data generated by the Huffman decoding unit.

  In this embodiment, the data input to the code amount control unit is divided by 2 to the nth power (n = 0, 1,...) As in the code amount control units 204 and 205 in FIG. Thus, the code amount is controlled. Note that the value of n can be set by a code amount control signal sent from a register (not shown) or the image ring output unit 107.

  Reference numeral 807 denotes a Huffman encoding unit that performs predetermined encoding processing on the data output from the code amount control unit 806 to generate encoded data. In this embodiment, an operation equivalent to that of the Huffman encoders 206 and 207 in FIG. 2 is performed.

  A data partial mask processing unit 809 performs the same processing as the data mask units 902 and 903 only on the portion representing the data value for the pack-bits-encoded image additional information output from the data buffer 2. That is, the Packbits-encoded data input from the data buffer 2 includes a “data value portion” that represents a data value and a “numerical value portion” that represents the number of consecutive data values or the number of input data values. It is configured. The data portion mask processing unit 809 analyzes the input Packbits encoded data and performs data mask processing only on the “data value portion”.

  If unencoded data is input from the data buffer 2, data mask processing is performed on all input data.

  The mask level of the data mask processing performed by the data partial mask processing unit 809 is output from the image ring output unit as a code amount control signal.

  810 is a redundant part pack bits encoding unit, and as a result of performing mask processing in the data part mask processing unit 809, for the redundant part generated in the encoded data due to occurrence of many parts where the same value continues, The Packbits encoding process is performed again.

  When unencoded data is output from the data partial mask processing unit 809, the redundant part pack bits encoding unit 810 performs normal pack bits encoding processing.

  Pack bit encoded data compressed again by the redundant portion pack bit encoding section is stored in the data buffer 3.

  In addition, when the redundant part generated by the redundant part pack bits encoding unit 810 is pack bit compressed by the redundant part pack bits encoding unit 810, the data after compression is rarely increased even though there are many consecutive data values. The amount may be larger than the data amount before the data mask process is performed. In that case, the redundant part pack bits encoding unit outputs the data to the data buffer 3 without including the redundant part without performing the compression process.

  Here, it is determined whether or not the data amount is increased when the pack-bits compression is performed by the redundant part pack-bits encoding unit 810. The pack-bits encoded data amount at each mask level output from the data amount detection unit 908 in FIG. This is done by referring to Zdata Byte Length.

  Reference numeral 808 denotes a data buffer 3, which has a capacity for one data packet.

  The data buffer 3 acquires header information from the data buffer 2, image data encoded from the Huffman encoding unit 807, and image additional information from the redundant portion Packbits encoding unit 810, and stores predetermined information in the header information within the buffer. Then, a new data packet is generated from the encoded image data and the image additional information.

  The generated data packet is stored in the memory via the memory controller.

  FIG. 7 is a flowchart of the present embodiment focusing on the operation of the image ring output unit 107. FIG. 7 shows the operation when a plurality of bits of scan data is input from the scanner 2070 for convenience, but the same processing is performed if the image data input to the tile compression unit is a data packet having a plurality of bits. Is possible.

  In the following, only image data will be described in FIGS. 7 and 3 to 6, and re-encoding of image additional information will be omitted.

  The operation will be described below with reference to FIG.

  First, before receiving a data packet in the tile compression unit, n = 0 (outputs the input data) to the code amount control unit 204 as an initial setting, and n = 1 (inputs to the code amount control unit 205). And n = 1 (similar to the code amount control unit 205) is set for the code amount control unit 806.

  Further, the value of the page counter in the image ring output unit 107 is set to 0 (step 701).

  When a data packet of image data obtained by scanning a document from the scanner 2070 is input to the tile compression unit, the image data is sent to the first image compression processing unit 103, and is subjected to DCT conversion, quantization, and code amount control units 204 and 205. The JPEG conversion is performed by the code amount control by, and the Huffman coding.

  Here, since the code amount control unit 204 is set to “through” (1/1) in step 701, the image data subjected to JPEG conversion through the code amount control unit 204 is quantized by the quantization unit 203. First image code data obtained by performing Huffman coding on the data quantized by the value is generated.

  On the other hand, since the code amount control unit 205 is set to output half the input data in step 701, the image data JPEG converted through the code amount control unit 205 is set in the quantization unit. Second image code data obtained by performing Huffman coding on the data that has been quantized twice the quantized value is generated.

  Here, since the second image code data is quantized with a larger value, the capacity of the code data is smaller than that of the first image code data.

  When predetermined data is input from the header information unit 102, the first compression processing unit 103, and the second compression processing unit 104, the packet generation unit 105 uses the first encoded image data from the data. The second data packet using the second data packet and the second image encoded data is generated, and a data output request is made to the image ring output unit.

  When there is a data output request from the packet generation unit 105 (step 702), the image ring output unit receives the first data packet from the packet generation unit (step 703) and stores the second data packet in the local memory 803. (Step 704).

  The image ring output unit obtains the capacity of the data packet by referring to the header information of the first data packet, adds the capacity to the page counter (step 705), and then the value of the page counter and the main memory capacity (RAM 2002). Are compared (step 706).

  In step 706, if the value of the page counter is less than or equal to the main memory capacity, the main memory has a capacity to store the data packet acquired by the image ring output unit. The data packet is output to the unit 2150, the data packet is stored in the RAM 2002 (step 707), and the data transfer for one packet is completed.

  When the data transfer for one packet is completed, the image ring output unit determines whether the data packet for one page has been output (step 708), and the operation is performed when the transfer of the data packet for one page has been completed. finish. On the other hand, if the transfer of the data packet has not ended, the process returns to step 702 and the operations from step 702 to step 707 are repeated.

  On the other hand, if the value of the page counter exceeds the capacity of the main memory in step 706, the main memory cannot store a data packet for one page, so that data with a high compression ratio and a small data capacity is re-started from the beginning of the page. There is a need to store it in main memory.

  For this purpose, the image ring output unit first sets the Ignore Flag (3022) in the header information 3001 of the data packet acquired from the packet generation unit, and outputs the data packet to the system control unit to clear the main memory ( At the same time as step 709), the page counter value is cleared to 0 (step 710).

  When the system controller obtains the data packet in which the Ignor Flag is set, the RAM controller 2124 clears the RAM (specifically, the Packet Address Pointer in FIG. 6 is set to 0), so that the data stored in the RAM until then. Invalidate the packet.

  When the main memory and the page counter are cleared, the image ring output unit next instructs the local memory 803 to output the stored data packet (step 711).

  As a result, the local memory issues an output request to the image ring output unit until all the data packets stored in the local memory are output to the image ring output unit.

  The image ring output unit outputs a code amount control signal so that the code amount control units 204 and 205 control the code amount. When the code amount control signal is input from the image ring unit, the code amount control unit newly sets a value obtained by halving the division value set in the code amount control unit to the previously set value. (Step 712).

  As a result, the frequency component data output from the DCT transform unit 202 is quantized with a value twice that before the Huffman coding.

  Further, since the value newly set in the code amount control unit 204 in step 711 is the value set in the code amount control unit 205 until then, the Huffman encoding unit 206 to the packet generation unit 105 after step 711. The data packet output to the main memory via the image ring output unit 107 and the data packet output from the Huffman encoding unit 207 to the local memory via the packet generation unit 105 before step 711 have the same quantized value. The image data is quantized by.

  When the setting value of the code amount control unit is changed in step 711, the image ring output unit returns to step 702 and determines whether there is a scan data output request from the packet generation unit. If there is a scan data output request, the scan data is acquired again in steps 703 to 705, and the data packet is stored in the main memory and the local memory.

  If there is no scan data output request from the packet generation unit in step 702, the process proceeds to step 713. If an output request is issued from the local memory, the data packet is acquired from the local memory (step 714). The data packet is re-encoded with respect to the image data by the Huffman decoding unit 805, the code amount control unit 806, and the Huffman encoding unit 807, and the data packet is generated again and stored in the local memory (step 715).

  If no output request is issued from the local memory in step 713, the process returns to step 702 to detect the scan data output request again.

  3 to 6 are diagrams showing the flow of image data of the tile compression unit in the present embodiment. Hereinafter, the operation of the present embodiment will be described with reference to FIGS.

  3 to 6 show only main parts.

  FIG. 3 shows a state before the scan operation is started and the page counter value exceeds the main memory capacity. The code amount control units 204 and 205 operate with the values set in step 701. As a result, the encoded image data quantized by the quantizing unit 203 in the main memory is quantized by the quantizing unit 203 in the local memory and further reduced to a half value (quantizing unit). Encoded image data (quantized with a value twice the value set in) is stored.

  In FIG. 3, reference numeral 301 denotes the internal state of the local memory, and 302 denotes the internal state of the RAM 2002, which is the main memory. Encoded image data quantized by the quantizing unit 203 (displayed by vertical lines) is stored in the main memory, while the local memory is quantized with a value twice the value set in the quantizing unit. The encoded image data (displayed with diagonal lines) is stored.

  FIG. 4 shows a state after the value of the page counter exceeds the main memory capacity in step 706. In step 712, the code amount control units 204 and 205 are set to 1/2 in 204 and 1/4 in 205, respectively, and the main memory is quantized with a value twice the value set in the quantization unit. The encoded image data is quantized in the local memory after being quantized by the quantizing unit, and is reduced to a quarter value (quantized by four times the value set in the quantizing unit). Data is stored.

  Further, the local memory outputs the image data (quantized with a value twice the value set in the quantization unit) stored so far to the main memory, and the Huffman complexing unit 805. Code amount control unit 806 (set to 1/2 by step 701) and Huffman encoding unit 807 generate encoded image data quantized with a value four times the value set in the quantization unit And store it in local memory.

  As a result of the above operation, the encoded image data (displayed with diagonal lines) quantized with a value twice the value set in the quantization unit for output to the main memory as shown in 301 in the local memory, Stored is encoded image data (displayed by a halftone line) quantized with a value four times the value set in the quantization unit generated by the Huffman encoding units 207 and 807.

  The main memory is set to the quantized unit generated by the Huffman encoding 206 and the encoded image data quantized with a value twice the value set in the quantizing unit stored in the local memory. Encoded image data (displayed with diagonal lines) quantized with twice the value is stored.

  In FIG. 4, the image data stored in the main memory has the image data output from the local memory and the scan data input from the packet generation unit stored in the main memory in no particular order. Also, the encoded image data quantized with a value four times the value set in the quantization unit stored in the local memory is the image data output from the Huffman encoding unit 207 and the image output from 807. Data is stored out of order.

  These images manage the position on the document and the printing order based on the packet IDs (3009) and (3010) of the header information 3001.

  FIG. 5 shows a state after all the encoded image data quantized with a value twice the value set in the quantization unit from the local memory in FIG. 4 is stored in the main memory. In FIG. 4, the image data is not transferred from the local memory to the main memory, and the operation of the Huffman decoding unit 805, the code amount control unit 806, and the Huffman encoding unit 807 is stopped.

  Then, the encoded image data quantized with the value 2 set in the quantization unit in the main memory is quantized with a value four times the value set in the quantization unit in the local memory. Each of the encoded image data is stored.

  FIG. 6 shows a state after the value of the page counter exceeds the main memory capacity again in step 706 from the state of FIG. In step 712, the code amount control units 204 and 204 are set to 1/4 in 204 and 1/8 in 205, respectively, and the main memory is quantized with a value four times the value set in the quantization unit. The encoded image data obtained by quantizing the encoded image data with a value eight times the value set in the quantization unit is stored in the local memory.

  At the same time, the image data stored so far is output from the local memory to the main memory, and set in the quantization unit by the Huffman combination unit 805, the code amount control unit 806, and the Huffman encoding unit 807. Coded image data quantized with 8 times the value is generated and stored in the local memory (indicated by a horizontal line).

  By repeating the above operation until the data input from the scanner is completed, the main memory can store an image having an optimum compression rate that can be stored in the memory.

  In the above description, the code amount control is performed by dividing the data input by the code amount control unit by a predetermined value. However, the data capacity is reduced by the Huffman coding connected to the subsequent stage, that is, JPEG compression. If the compression rate is improved when performing the above, the operation of the code amount control unit is not limited.

  In the above description, the JPEG method is used as a method for compressing image data. However, the present embodiment is not limited to the JPEG compression method. If the compression method can change the compression rate according to the setting, the Packbits method or It is also possible to compress the image data by other compression methods.

  Further, in the above description, the compression ratio stored in the main memory is changed by comparing the page counter value with the main memory capacity. However, the value to be compared with the page counter value may be another value and should be read. It is also possible to change the ratio according to the ratio of read image data to all image data.

Block diagram of the tile compression unit 2106 in the present embodiment Block diagram of the first compression processing unit 103 First operation explanatory diagram in this embodiment Second operation explanatory diagram in the present embodiment 3rd operation explanatory drawing in a present Example 4th operation explanatory drawing in a present Example. Flow chart showing the operation in this embodiment Block diagram of re-encoding unit 106 Block diagram of the second compression processing unit 103 The figure for demonstrating operation | movement of the data mask parts 902 and 903 Diagram showing actual usage environment of this system Overall block diagram of this system controller A diagram representing an image packet Diagram showing command packet Diagram showing interrupt packet Diagram showing packet table

Explanation of symbols

101 tile bus interface unit 102 header information holding unit 103 first compression processing unit 104 second compression processing unit 105 packet generation unit 106 recoding unit 107 image ring output unit 108 register setting bus interface unit 109 register setting unit

Claims (4)

Image input means for inputting image information;
A dividing unit for dividing the image information input from the image input unit into predetermined units;
The image information divided by the dividing means has a plurality of data.
A plurality of encoding means for encoding each of a plurality of data in a predetermined unit divided by the dividing means;
Comparison means for comparing the amount of data before and after being encoded by the encoding means;
Data output means for selecting and outputting either the encoded data or the data before encoding according to the comparison result of the comparison means;
An image processing apparatus comprising:   2. The image processing apparatus according to claim 1, wherein the image input means is a document reading apparatus such as a scanner.   2. The image processing apparatus according to claim 1, wherein one of the plurality of encoding means is an encoding process by a JPEG method.   The image processing apparatus according to claim 1, wherein one of the plurality of encoding units is an encoding process by a Packbits method.

Images