JP2007142581A - Image processing method and image processing unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for creating image encoding data enabling image decoding at higher speed in consideration of memory efficiency. <P>SOLUTION: In S503, the encoding data are used to decode an image at a decoding resolution level, and the decoded image is encoded. In S504, a fragment table, which is referred to for using the encoded image, is created instead of a portion required to decode the image at the decoding resolution level in the encoding data. In S505, the encoded image and the fragment table are managed with the encoding data. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image encoding technique.

  In recent years, the resolution of a display used as a computer display device has improved with the evolution of technology. For example, in the past, the resolution of about VGA (640 × 480 pixels) and SVGA (800 × 600 pixels) is now XGA (1024 × 768 pixels) or more even for notebook PCs. In addition, SXGA (1280x1024 pixels) and UXGA (1600x1200 pixels) are prevalent in desktop PCs, and it can be easily guessed that high resolution such as QXGA and QUXGA will be supported in the future. For this reason, the types of display devices that display images have increased, and image display suitable for each device has become necessary. In particular, regarding the resolution, it has become necessary to adjust the size of the thumbnail image data in accordance with the resolution of the display.

  Furthermore, with the advancement of digital cameras, the resolution of images that can be taken has been increasing year by year, and no matter how high the resolution of the display becomes, the entire image cannot be displayed at the maximum resolution at one time. Therefore, a list of thumbnail images of a large number of image data taken with a digital camera or the like is displayed as a list, and the image is switched to single image display by selecting from among the thumbnail images. Applications that display image data have become necessary. In this case, when displaying a large number of image data at once, high speed is regarded as important, and therefore thumbnail image data created in advance different from the main image data is often used. On the other hand, the single image display selected from the thumbnail display uses the main image data to display the entire image until it can be displayed on the window size of the application or the display device, and to display an enlarged image larger than that. Then, a part of the image is displayed.

  When displaying such a single piece of image data, if the application uses a plurality of image sizes different from the image size of the original image data different from the image size of the original image data, as required A decoding process is performed, the image size of the original image data is once reproduced, and then the image data is reduced so as to be the target image size. In this case, the image data is uncompressed or compressed by a sequential encoding method such as JPEG baseline.

  Alternatively, the image data is compressed by a hierarchical encoding method, and the image size required for display is decoded and displayed. As a hierarchical encoding method, there is JPEG2000. This JPEG2000 was standardized by ISO / ITC in 2001 as a hierarchical encoding method that divided one image data into one or more tiles and had one or more resolutions for each divided tile. This is an image encoding method. In addition, as a protocol for accessing only the necessary part of the encoded data file encoded with JPEG2000 on the network, JPEG2000 image coding system-Part 9: Interactivity tools, APIs and protocols (hereinafter, JPIP)). In the case of JPEG2000, the number of resolutions, that is, the image size of the lowest resolution is determined by encoding (during encoding).

  Patent documents 1 and 2 can be cited as technologies related to the JPIP.

  According to Patent Document 1, a compressed image file encoded by JPEG2000 is placed on a server, a code stream is partially requested from a client, and is buffered and stored. Then, a request is newly issued to the server for only the unstored part, and the received code data and the already buffered code data are joined and decoded. The partial request unit at this time is selected from packets, tiles, and code blocks in the JPEG2000 code. Requests in byte units are also possible. On the other hand, the server extracts the partial data of the requested compressed data and returns it to the client.

  On the other hand, according to Patent Document 2, the JPEG2000 code data is requested to the server in subband units, and the coefficients used at the time of decoding are held. When the next upper subband code data is requested, decoding is performed by adding the current data to the coefficient. Then, holding the coefficients used in the decoding is repeated. This provides a system that does not require buffering and storing the code data received on the client side. This also enables progressive display in the resolution direction.

  In addition, there is one minimum resolution data in one hierarchical code data, and it is general to realize hierarchical decoding in the resolution direction by decoding sequentially from the lowest resolution code data. Therefore, in order to extract image data having an image size larger than the minimum resolution image size from the code data, it is necessary to repeat the hierarchical decoding process a plurality of times depending on the required output image size. Specifically, in the case of JPEG2000, the desired image data is obtained by repeating DWT as many times as necessary. Conversely, when requesting image data of a size smaller than the image size of the lowest resolution, it is necessary to first decode the image data of the lowest resolution and then perform a reduction process.

  In order to save intermediate data such as the image data of the intermediate layer processed once or image data smaller than the minimum resolution, it is a file different from the original image data file and independent of the original hierarchical code data. It is necessary to perform encoding processing or the like as the new image data and save it as a separate file. In the case of saving as this separate file, there are a plurality of files for one image in consideration of managing the files, and the process is complicated when moving, deleting, copying, etc. Become. You also need to be careful about how to name the file.

On the other hand, in the case where intermediate data is not stored, it is necessary to perform the above processing every time a user request occurs. That is, even when the same image size is requested a plurality of times, it is necessary to perform the above processing for the number of times requested.
JP2003-023630 JP 2004-040674 A

  For example, it is assumed that there is JPEG2000 code data “A.j2k” obtained by encoding image data with an image size of 2048 × 2048 pixels at decomposition level = 5. In this case, “A.j2k” can reproduce image data of six different resolutions of 64 × 64, 128 × 128, 256 × 256, 512 × 512, 1024 × 1024, and 2048 × 2048. It is also assumed that the image size displayed when an image decoding / display application “APP_1.exe” is selected from thumbnail images is an image size equivalent to 256 × 256 pixels.

  Under these conditions, the processing when displaying one image data selected by "APP_1.exe" decodes the code data of three resolutions of 64x64, 128x128, 256x256 and displays the decoded image Will do. In other words, since this application “APP_1.exe” does not handle image data with a resolution of 256x256 pixels or less, if there is code data with a resolution of 256x256 pixels or less, the decoder must always decode the image data of 256x256 pixels or more. Must be looped. Therefore, there is a problem that decoding takes time. In addition, code data having a resolution that is not used in the application must be stored in a buffer or the like, and there is a problem that an extra storage area is used. Furthermore, when JPIP requests code data from the server, there is a problem that it takes extra time for communication.

  Further, in Patent Document 2, since the decoding process is stopped halfway and the coefficients used in the decoding process are held, more memory for holding is used than when the code data is held. Also, a special decoder is required, such as stopping the decoder halfway or adding subband data.

  In Patent Documents 1 and 2, the compression parameter of the JPEG2000 code data held in the server is not changed, a part of the code data is transmitted to the client in a fragmented manner, and the code data received by the client is not converted. Since the decoding is performed as it is, there is a problem that even the code data of the resolution which is not used in the client side application must be processed. Furthermore, when the number of resolution layers exceeds the number of layers supported on the decoder side, there is a problem that the received code data may not be decoded to the end.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique for creating image encoded data for enabling image decoding in consideration of memory efficiency at a higher speed. And

  In order to achieve the object of the present invention, for example, an image processing method of the present invention comprises the following arrangement.

That is, obtaining encoded data configured to be able to decode an image at a plurality of resolution levels;
Obtaining the decoding resolution level;
A decoding step of decoding an image at the decoding resolution level using the encoded data;
An encoding step of encoding the image decoded in the decoding step;
A creation step of creating reference information to be referred to in order to use the image encoded in the encoding step, instead of a portion necessary for decoding the image at the decoding resolution level in the encoded data;
A management step of managing the image encoded in the encoding step and the reference information together with the encoded data.

  In order to achieve the object of the present invention, for example, an image processing method of the present invention comprises the following arrangement.

That is, obtaining encoded data configured to be able to decode an image at a plurality of resolution levels;
Obtaining the decoding resolution level;
A decoding step of decoding an image having the lowest resolution level in the encoded data;
An encoding step of encoding the image decoded in the decoding step so that an image of an image size between the image size of the decoding resolution level and the image size of the lowest resolution level can be reproduced;
A creation step of creating reference information to be referred to in order to use the image encoded in the encoding step instead of the lowest resolution level image;
A management step of managing the image encoded in the encoding step and the reference information together with the encoded data.

  In order to achieve the object of the present invention, for example, an image processing apparatus of the present invention comprises the following arrangement.

That is, means for acquiring encoded data configured to be able to decode an image at a plurality of resolution levels;
Means for obtaining the decoding resolution level;
Decoding means for decoding an image at the decoding resolution level using the encoded data;
Encoding means for encoding the image decoded by the decoding means;
Creating means for creating reference information to be referred to in order to use an image encoded by the encoding means instead of a portion necessary for decoding an image at the decoding resolution level in the encoded data;
Management means for managing the image encoded by the encoding means and the reference information together with the encoded data.

  In order to achieve the object of the present invention, for example, an image processing apparatus of the present invention comprises the following arrangement.

That is, means for acquiring encoded data configured to be able to decode an image at a plurality of resolution levels;
Means for obtaining the decoding resolution level;
Decoding means for decoding an image having the lowest resolution level in the encoded data;
Encoding means for encoding the image decoded by the decoding means so that an image having an image size between the image size of the decoding resolution level and the image size of the lowest resolution level can be reproduced;
Creating means for creating reference information to be referred to in order to use the image encoded by the encoding means instead of the image of the lowest resolution level;
Management means for managing the image encoded by the encoding means and the reference information together with the encoded data.

  With the configuration of the present invention, it is possible to create image encoded data for enabling image decoding in consideration of memory efficiency at a higher speed.

  Hereinafter, the present invention will be described in detail according to preferred embodiments with reference to the accompanying drawings.

[First Embodiment]
FIG. 1 is a schematic configuration diagram of a system according to the present embodiment. As shown in the figure, the system according to the present embodiment includes two client computers 102 and 103, a server 101, and an image database 104. Further, the two client computers 102 and 103 and the server 103 are connected via the network 100, and are configured to be capable of data communication with each other.

  The network 100 is configured by a LAN, the Internet, or the like, and is configured by one of wireless, wired, or a combination thereof. The image database 104 stores a plurality of encoded image data compressed and encoded by JPEG2000 in a file format. From the client computers 102 and 103, among the encoded image data stored in the image database 104, an acquisition request for desired encoded image data is transmitted to the server 103. Data is acquired from the image database 104 and returned to the client computers 102 and 103.

  Therefore, the server 103 is installed with software necessary for the WWW server function, as well as a JPIP server, which is a JPEG2000 communication protocol for transmitting encoded image data to the client computers 102 and 103.

  Further, the client computer 102, 103 transmits a request for obtaining desired image encoded data to the server 103, decodes the image encoded data transmitted from the server 103 in response to this request, and displays the decoding result. Software necessary for performing a series of processing such as display, for example, client software necessary for displaying each page of the Web, including a Web browser, and decoding / displaying image encoded data Necessary client software and software that implements the JPIP client function are installed.

  The configuration of the system shown in the figure is merely an example, and the number of servers and client computers is not particularly limited to the configuration shown in the figure.

  FIG. 2 is a block diagram showing a hardware configuration of a computer applicable to the client computers 102 and 103 and the server 103.

  A CPU 201 controls the entire computer using programs and data stored in the ROM 204 and the RAM 205, and executes each process described below performed by each device to which the computer is applied.

  A keyboard 202 is operated by an operator of the computer and can input various instructions to the CPU 201. In the figure, a pointing device 202a such as a mouse or a joystick is connected to the keyboard 202. Like the keyboard 202, the pointing device 202a inputs various instructions to the CPU 201 when operated by an operator. Can do.

  A display unit 203 includes a CRT, a liquid crystal screen, and the like, and can display a processing result by the CPU 201 as an image or a character.

  A ROM 204 stores setting data, a boot program, and the like of the computer.

  Reference numeral 205 denotes a RAM, which is an area for temporarily storing programs and data loaded from the hard disk 206 and the floppy (registered trademark) disk 207, and temporarily stores data received from the outside via the network I / F 209. Various areas such as a work area used for the CPU 201 to execute various processes can be provided as appropriate.

  Reference numeral 206 denotes a hard disk, which stores programs and data for causing the CPU 201 to execute processes to be described later performed by the OS and each apparatus to which the computer is applied, and these are loaded into the RAM 205 as appropriate under the control of the CPU 201. Is. As described above, since different software is installed on the server and the client computer, the information stored in the hard disk 206 is applied to the client computer and the server. Some are different.

  A floppy (registered trademark) disk 207 reads a program and data recorded in a floppy (registered trademark) as an example of a storage medium, and outputs it to the RAM 205 and the hard disk 206.

  A printer I / F 208 functions as an interface for connecting a printing device such as a printer to the computer. The computer outputs data to be printed to the printing device via the printer I / F 208. can do.

  Reference numeral 209 denotes a network I / F that functions as an interface for connecting the computer to the network 100 shown in FIG. 1. The computer is connected to the network 100 via the network I / F 209. Data communication with the device can be performed.

  Next, image encoded data stored in the image database 104 will be described. FIG. 3 is a diagram illustrating a configuration example of image encoded data obtained by compressing and encoding an image using general JPEG2000. The configuration shown in the figure is an example of the configuration of JPEG2000 encoded data encoded with the encoding option of Resolution level-Layer-Component-Position progression (hereinafter referred to as Resolution Progression).

  When conforming to Resolution Progression, encoded data is recorded in the order of resolution / layer / component / position. In addition, although the code order such as SNR Progression can be selected, the arrangement order of these encoded data is called a progression order. In the present embodiment, in order to simplify the description, it is assumed that the encoded image data has a configuration according to Resolution Progression (that is, the configuration shown in FIG. 3).

  Next, the relationship between each subband by wavelet transform and resolution will be described. FIG. 4 is a diagram showing a subband configuration obtained when wavelet transform is performed so that images of six resolution levels can be reproduced. Such a subband has decomposition level = 5.

  First, when the subband LL (NL = 0) is decoded, an image of resolution level 0 (resolution level = 0) is obtained. Further, by decoding the three subbands HL (NL = 1), LH (NL = 1), and HH (NL = 1) following LL (NL = 0), resolution level 1 (resolution level = The image of 1) is obtained.

  When the subband of NL = 2 is further decoded, an image of resolution level 2 (resolution level = 2) is obtained. When the subband of NL = 3 is further decoded, resolution level 3 (resolution level = 3) If the NL = 4 sub-band is decoded further, the resolution level 4 (resolution level = 4) image is obtained, and if the NL = 5 sub-band is further decoded, the resolution level 5 An image of (resolution level = 5) is obtained.

  As described above, the encoded image data of decomposition level = 5 can reproduce images of six resolution levels by sequentially decoding each NL subband. Each time the resolution level increases, the image size of the reproduced image data doubles in both width and height.

  In the case of the Resolution Progression shown in FIG. 3, the code data of one resolution level are arranged in the order of the smallest layer number, the smallest component number, and the smallest position number. The layer number here corresponds to the S / N ratio for the original image of the image to be restored, and the smaller the layer number, the more the image data with a worse S / N ratio is reproduced.

  Furthermore, the resolution number, layer number, and component number values in the JPEG2000 encoded data file are preset according to the encoding parameters at the time of encoding, and are encoded according to the parameters. Is stored as header information in the encoded data. In addition, a packet that is a logical minimum unit in encoded data includes a packet header part that manages information of all code-blocks stored in each packet, and codes of all code-blocks constituting the packet. It consists of packet data that is data.

  Here, the encoded image data used in the present embodiment has a configuration as shown in FIG. 3 and, as shown in FIG. 4, NL = 0 to 5 so that images of six different resolution levels can be reproduced. It is assumed that it consists of subbands. In this embodiment, since the size of an image obtained as a result of decoding all subbands of NL = 0 to 5 is set to 2048 pixels × 2048 pixels, an image obtained as a result of decoding up to NL = 0 to 5 is decoded. The size is as follows.

Image size obtained as a result of decoding the subband of NL = 0 64 pixels x 64 pixels
Image size obtained when decoding subbands with NL = 0, 1 128 pixels x 128 pixels
Image size obtained when decoding subbands with NL = 0 to 2 256 pixels x 256 pixels
Image size obtained when decoding subbands with NL = 0 to 3 512 pixels x 512 pixels
Image size obtained when subbands with NL = 0 to 4 are decoded 1024 pixels x 1024 pixels
The size of the image obtained when the subband of NL = 0 to 5 is decoded 2048 pixels × 2048 pixels That is, the image encoded data according to this embodiment is 2048 pixels from the size of 64 pixels × 64 pixels according to NL. Images of six types of sizes (resolution levels) up to × 2048 pixels can be reproduced.

  Furthermore, in this embodiment, for the sake of simplicity, it is assumed that such image encoded data is stored in the hard disk 206 of the client computer, and the client computer stores the image encoded data in the hard disk 206. The case of reading from and handling will be described. The client computer may be any of the client computers (client computers 102 and 103) shown in FIG. 1, but in the following description, the client computer 102 is used as an example.

  FIG. 6 is a diagram illustrating a configuration example of image encoded data according to the present embodiment. In the figure, reference numeral 600 denotes the entire encoded image data. Reference numeral 600-1 denotes "Start Of Codestream", which is a marker code indicating the start of encoded image data. Reference numeral 601 denotes a main header in which encoding parameters and the like are stored. For example, the maximum image size is 2048 pixels × 2048 pixels, and “decomposition level = 5” and “Resolution Progression” indicate that encoded image data is configured.

  Reference numerals 602-0 to 602-5 are subband data at each resolution level (NL) stored in “Resolution Progression”. That is, Resolution 0 (602-0) is encoded data of a subband of NL = 0 (that is, the LL subband shown in FIG. 4). If decoded, an image of 64 × 64 pixels (resolution level) 0 image) is obtained.

  Also, Resolution 1 (602-1) is a subband of NL = 1 (three subbands of HL (NL = 1), LH (NL = 1), and HH (NL = 1) shown in FIG. 4). ) And is decoded together with the previous Resolution 0 encoded data, whereby an image of 128 pixels × 128 pixels (image of resolution level 1) can be obtained.

  Also, Resolution 2 (602-2) is a subband of NL = 2 (three subbands of HL (NL = 2), LH (NL = 2), and HH (NL = 2) shown in FIG. 4). ) And is decoded together with the previous Resolution 0, 1 encoded data, an image of 256 pixels × 256 pixels (an image of resolution level 2) can be obtained.

  Also, Resolution 3 (602-3) is a subband of NL = 3 (three subbands of HL (NL = 3), LH (NL = 3), and HH (NL = 3) shown in FIG. 4). ) And is decoded together with the encoded data of Resolution 0 to 2 above, an image of 512 pixels × 512 pixels (image of resolution level 3) can be obtained.

  The resolution 4 (602-4) is a subband of NL = 4 (three subbands of HL (NL = 4), LH (NL = 4), and HH (NL = 4) shown in FIG. 4). ) And is decoded together with the encoded data of Resolution 0 to 3 above, an image of 1024 pixels × 1024 pixels (image of resolution level 4) can be obtained.

  Also, Resolution 5 (602-5) is a subband of NL = 5 (three subbands of HL (NL = 5), LH (NL = 5), and HH (NL = 5) shown in FIG. 4). ) And is decoded together with the encoded data of Resolutions 0 to 4 above, an image of 2048 pixels × 2048 pixels (image of resolution level 5) can be obtained.

  In this way, the encoded data of each subband is included in the encoded image data in order of increasing resolution level. In the following description, a subband of NL = X may be referred to as a “subband of resolution level X”.

  Reference numeral 600-2 denotes "End of codestream", which is a marker code indicating the end of encoded image data.

  Here, when an application that handles only an image having a size of, for example, 256 pixels × 256 pixels or more handles such image encoded data, a request for decoding the image encoded data such as image display is generated by this application. In such a case, it is necessary to always decode at least sub-band encoded data of NL = 0-2. It is necessary to decode an image with a size of at least 256 pixels × 256 pixels, and when an image with a size larger than this is required, it is necessary to further decode the encoded data of a subband of NL = 3 or more. Because there is. Further, it is necessary to perform the process of decoding the encoded data of the subband of NL = 0-2 every time such a request occurs.

  That is, in such an application, every time processing that requires decoding of an image occurs, decoding processing of encoded data of subbands NL = 0 to 2 is always performed, and the processing becomes redundant every time. . In the present embodiment, in order to cope with such a problem, the encoded data that is decoded every time a decoding request is received (in this example, the encoded data of the subband of NL = 0 to 2) is first decoded. Processing that is decoded in the processing is newly set as a new LL subband.

  As a result, the encoded data portion that has been decoded each time has already been decoded in the form of the new LL subband in the first decoding process, and this may be used. As a result, in the second and subsequent decoding processes, this redundant decoding process need not be performed. Accordingly, the second and subsequent decoding processes can be performed at a higher speed, and the redundant portion decoded at the first time does not have to be held in a separate file form, which is preferable from the viewpoint of memory efficiency. .

  FIG. 5 is a flowchart of processing performed by the client computer 102 when a request for decoding the encoded image data stored in the hard disk 206 is generated. Note that programs and data for causing the CPU 201 to execute the processing according to the flowchart of FIG. 5 are stored in the hard disk 206 and are appropriately loaded into the RAM 205 under the control of the CPU 201. Then, when the CPU 201 executes processing using this, the client computer 102 executes each processing described below.

  First, in step S500, it is checked whether or not the decoding request for the image encoded data to be processed is the first time. In general, an application that accesses a file holds a history of which file has been accessed in the past. Therefore, also in this step, it is checked whether or not the same decoding request for the image encoded data has been made in the past by referring to the past history. If it is determined and not made, it is determined that this request is the first time.

  Accordingly, as a result of the process in step S500, if the current decoding process is not the first, the process proceeds to step S506, and if it is the first, the process proceeds to step S501. In step S501, the resolution level (decoding resolution level) of the image to be decoded is acquired. The acquisition form is not particularly limited, and various forms are conceivable. For example, data describing a predetermined decoding resolution level may be acquired from the hard disk 206. In addition, a GUI (graphical user interface) that prompts the input of the decoding resolution level is displayed on the display unit 203, and the decoding resolution input by the user via the GUI is obtained using the keyboard 202 or the pointing device 202a. Also good.

  In this embodiment, the decoding resolution level is “2” (resolution level = 2, ie, NL = 2). That is, the size of the image to be decoded is 256 pixels × 256 pixels. In step S501, any information may be input as long as it is information that can specify a portion to be decoded in order to obtain an image having a resolution desired by the user in each subband.

  Next, in step S502, the file containing the encoded image data is referred to, and it is checked whether or not the input encoded data of the subband of the decoding resolution level exists. That is, if the subband of the decoding resolution level is an LL subband in the encoded image data, or if a new LL subband described later is already managed in this file, the new LL subband is used. In some cases, it is determined that the subband encoded data of the input decoding resolution level exists in this file. Further, even if the decoding resolution level subband is not the LL subband in the encoded image data and a new LL subband described later is already managed in this file, this new LL subband is also managed. If not, it is determined that the subband encoded data of the input decoding resolution level does not exist in this file.

  If it exists as a result of this check process, the process advances to step S506. On the other hand, if it does not exist, the process proceeds to step S503.

  In step S503, all subbands to be decoded are decoded in order to obtain an image having a designated decoding resolution level (ie, an image having a size of 256 pixels × 256 pixels), and the decoded image is converted into a new LL. Subband. Details of the processing in step S503 will be described later.

  In step S504, a fragment table, which will be described later, is referred to in order to use the new LL subband created in step S503 instead of all the subbands to be decoded in order to obtain an image of the specified decoding resolution level. Processing for creating new image encoded data including the new LL subband created in S503 is performed. The items created in step S504 are collectively referred to as “new code”. Details of the processing in step S504 will be described later.

  In step S505, the encoded image data and the new code created in step S504 are written into one same file. Details of the processing in step S505 will be described later.

  On the other hand, in step S506, processing for specifying a reading method when reading desired encoded data from a file is performed. Details of the processing in step S506 will be described later.

  Next, in step S507, desired encoded data is read and decoded (decoded) in accordance with the reading method specified in step S506.

  In step S590, the new LL subband created in step S503 or the result decoded in step S507 is output as output image data. Although the output destination varies depending on the request for image decoding, for example, if the request for image display is input by the keyboard 202 or the pointing device 202a, the output destination is the display unit 203.

  Next, new LL subband creation processing in step S503 will be described with reference to FIG. 7A. FIG. 7A is a flowchart showing details of the process in step S503.

  First, in step S700, the number of tiles in the encoded image data is obtained based on the encoding parameter stored in the main header. As a calculation method, the maximum image size and the size of one tile are used to calculate the maximum image size / the size of one tile.

  In step S701, the decoding resolution level acquired in step S501 is set in the decoder as the number of decoding layers. In this embodiment, the decoder is realized by the CPU 201 executing the decoding program. In this case, the decoding resolution level acquired in step S501 is set as a variable used when the decoding program is executed. The encoding conditions can be changed for each tile.

  When the decoder is realized by hardware, the hardware may be set to a memory used for decoding. Here, the variable Dlevel for controlling the decoding resolution is set to zero.

  Next, in step S702, one tile is selected, and processing for decoding the encoded data of the subband of the resolution level indicated by Dlevel of the selected tile is performed. When the decoding process is completed, in step S703, it is checked whether the resolution level of the subband decoded in step S702 has reached the decoding resolution level. In this embodiment, since the decoding resolution level is “2”, it is checked whether or not the resolution level of the subband decoded in step S702 is “2”.

  As a result of this check, if the resolution level indicated by Dlevel of the subband decoded in step S702 has not reached the decoding resolution level, that is, the decoding process of this tile at the decoding resolution level acquired in step S501 is performed in step S702. If not done yet, Dlevel is incremented by one, and the process returns to step S702 to perform decoding processing on the encoded data of the subband at the next resolution level of this tile.

  On the other hand, if the resolution level of the subband decoded in step S702 reaches the decoding resolution level, that is, if the decoding process of this tile at the decoding resolution level acquired in step S501 is performed in step S702, the process In step S704, it is checked whether the decoding process has been performed for all tiles.

  If the result of this check is that decoding has not been performed for all tiles, the process returns to step S701, the next unprocessed tile is selected, and the processing from step S702 onward is performed for this selected tile.

  By performing the processing according to the flowchart of FIG. 7A described above, subbands at all resolution levels from resolution level 0 to the decoding resolution level (“2”) can be decoded for all tiles. As a result, an image of 256 pixels × 256 pixels can be obtained. In this embodiment, this image is used as a new LL subband. Here, it should be noted that no change processing is performed on the original image encoded data by the above processing.

  Next, the new code creation process in step S504 will be described with reference to FIG. 7B. FIG. 7B is a flowchart showing details of the processing in step S504.

  First, in step S705, a main header (old main header) included in the original image encoded data is extracted. In step S706, a process of creating a new main header is performed using the extracted main header. As one of the processes, decomposition level = 5 described in the main header is described as decomposition level = 3 in the new main header. This is because the subband of NL = 0 to 2 is set as one subband (new LL subband) in the process in step S503. That is, if this new LL subband is NL = 0, the subband of NL = 1 is the subband of original NL = 3, the subband of NL = 2 is the subband of original NL = 4, and the subband of NL = 3 This is because the band is an original NL = 5 subband, and as a result, the maximum value of the NL is “3”. The new main header constitutes new encoded data described later together with the new LL subband.

  Next, in step S707, the image created in step S503 is encoded with EBCOT so as to be encoded data of the new LL subband. In step S708, instead of subbands of NL = 0-2, a process for creating each information to be described later to be used for using the new LL subband is performed. In step S709, the information created in step S708 is performed. Is written to the fragment table described later.

  A specific configuration of the new code created as a result of the above processing will be described with reference to FIG. In the drawing, a portion indicated by reference numeral 603 is new image encoded data including the new LL subband created in step S503. Also, the parts indicated by reference numerals 603-1 and 604 are the new main header created in step S706. Also, the portion indicated by 605-1 is the encoded data of the new LL subband created in step S707.

  In the processing in step S708, the number of bytes of the portion indicated by 603-1, 604, 605-1 is stored in the Length field indicated by 607-2. In a field indicated by 607-1 (Offset field), Offset of the new encoded data 603 from the head of the file in which the encoded image data 600 and the new encoded image data 603 are written is stored.

  Then, in order to indicate the encoded data after the resolution 3 subband encoded data (602-3) in the encoded image data 600, the field indicated by 608-1 includes the encoded image data 600, the new image code, and the like. Stores the Offset of Resolution 3 of 602-3 from the top of the file in which the digitized data 603 is written. The field indicated by 608-2 stores encoded data 602-4 and 602-5 after Resolution 3 of 602-3 and the total number of bytes of EOC indicated by 600-2.

  Thereby, the fragment table indicated by 607 can be completed. Therefore, referring to this fragment table from the top (upper part), first, the encoded data of the new LL subband to be used instead of the encoded data (602-0 to 602-2) of the original Resolution 0-2 ( 605-1) can be specified, and then encoded data after Resolution 3 can be specified.

  Next, details of the file writing process in step S505 will be described with reference to FIG. 8A. FIG. 8A is a flowchart showing details of the processing in step S505.

  First, in step S801, the new main header (604 in FIG. 6) created in step S706 is acquired. Next, in step S802, the acquired new main header is referred to, the encoded data of the new LL subband is specified, and this is read out. Next, in step S803, the fragment table created in step S709 is acquired. In step S804, in order to indicate that the new LL subband is a subband of the lowest resolution level (minimum image size), information 606 indicating “256 pixels × 256 pixels is the minimum image size”, and a fragment table 706 are displayed. Are linked using a mechanism such as Association BOX in the case of JPEG2000.

  Finally, in step S805, the fragment table associated with the information 606 and the new encoded data (603 in FIG. 6) are stored in BOX (new BOX), respectively, and the image encoded data 600, new image encoded A process of writing (appending) the new BOX to the file storing the data 603 is performed.

  As a result, an image encoded data file having two LL subbands (encoded data of Resolution 0 of 602-0 and encoded data of Resolution 0 of 605-1) can be created.

  Next, details of the process for specifying the reading method in step S506 will be described with reference to FIG. 8B. FIG. 8B is a flowchart showing details of the processing in step S506.

  First, in step S811, the same processing as that in step S501 is performed. As a result of this check process, if the current decoding process is not the first, the process proceeds to step S810. If it is the first, the process proceeds to step S806.

  In step S806, the inside of the image encoded data file is referred to and it is checked whether or not there is sub-band encoded data of the decoding resolution level acquired in step S501. That is, whether the decoding resolution level is the resolution level 0 (that is, 602-0 in FIG. 6) in the original encoded image data (600 in FIG. 6), or in the newly encoded data (603 in FIG. 6). It is checked whether the resolution level is 0 (that is, 605-1 in FIG. 6).

  In the present embodiment, the decoding resolution level input in step S501 is an image having a size of 256 pixels × 256 pixels. In this case, the image size of resolution level 0 in the original image encoded data is 64 pixels × 64 pixels, and the image size of resolution level 0 in the new encoded data is 256 pixels × 256 pixels. In step S501, the input decoding resolution level means the resolution level 0 in the new encoded image data.

  As a result of the above check processing, if the decoding resolution level input in step S501 means the resolution level 0 in the original encoded image data, the process proceeds to step S809, and the original encoded image data is processed as usual. Decide to decode the LL subband.

  On the other hand, if it means the resolution level 0 in the new encoded data, the process proceeds to step S807. In step S807, information 606 indicating “256 pixels × 256 pixels is the minimum image size” is searched. In step S808, the fragment table associated with the information 606 is read out, and is decoded with reference to the fragment table. decide.

  Therefore, in step S507 performed after the processing according to the flowchart shown in FIG. 8B, the encoded data of the subbands can be specified in the decoding order by referring to the read fragment table in order from the top. Therefore, a desired image can be obtained by sequentially decoding the identified encoded data.

  On the other hand, in step S810, since the reading method is already determined in either step S808 or S809, it is determined that this determined method is used.

  FIG. 17 is a diagram showing the configuration of an image encoded data file that manages each of the BOXes. Of course, this file conforms to JPEG2000. In the figure, reference numeral 1700 denotes a BOX for holding a fragment table. If there are a plurality of fragment tables, the number of fragment list boxes indicated by 1700-1 may be as many as the number of fragment tables.

  Reference numeral 1701 denotes a BOX that holds the original encoded image data. Reference numeral 1702 denotes a BOX for holding new encoded data. Reference numeral 1703 denotes a BOX for storing Association information. By storing and adding each data described so far in this format, a general JPEG2000 decoder can perform decoding processing.

  As described above, according to the present embodiment, by having data of arbitrary resolution in the original image encoded data as the encoded data of the new LL subband, it is in the original image encoded data. , It can save the decoding of the low-resolution part that is not used. Thereby, the decoding process can be performed at high speed.

  In particular, if the processing can be performed by decoding only the LL subband, the repetition of decoding necessary for decoding the hierarchical code is eliminated. Also, since the original code data is not changed and new information is added, the original code data can be used as it is.

  In the present embodiment, the case where one new code is added to the encoded image data file has been described. However, a plurality of new codes may be added as required. That is, in the case of image encoded data that can reproduce images of six resolution levels, new codes are created and added for five types of subbands (NL = 1 to 5) other than the LL subband. Can do.

  In addition, in the encoding of the new LL subband in step S707, there is no need for discrete wavelet transform processing, and since it can be performed only by EBCOT encoding, the encoding processing can be performed at high speed.

  Furthermore, the fragment table used in the present embodiment is a table and management method defined in Part 2 of JPEG2000, and reads code data while using the fragment table during decoding processing from a newly added LL subband. No special processing system is required for the processing system.

[Second Embodiment]
In the first embodiment, a new LL subband to be used instead of the LL subband in the original image encoded data is created with a larger image size than the LL subband. In the present embodiment, a new LL subband having a smaller image size than the LL subband in the original image encoded data is created.

  FIG. 14 is a diagram illustrating a configuration example of encoded image data according to the present embodiment. In the figure, the same parts as those in FIG. Reference numeral 1400 denotes the entire encoded image data, and the configuration thereof is substantially the same as the configuration shown in FIG. The difference is that the encoded data 600 shown in FIG. 6 has “decomposition level = 5”, whereas the encoded image data 1400 shown in FIG. 14 has “decomposition level = 3”. It is.

  Also in this embodiment, since the size of the image obtained as a result of decoding all subbands of NL = 0 to 3 is 2048 pixels × 2048 pixels, the image obtained as a result of decoding up to each of NL = 0 to 3 is also used. The size is as follows.

Image size obtained as a result of decoding subbands with NL = 0 256 pixels x 256 pixels
Size of image obtained when subband of NL = 0, 1 is decoded 512 pixels x 512 pixels
Image size obtained when subbands with NL = 0 to 2 are decoded 1024 pixels x 1024 pixels
The size of the image obtained when the subband of NL = 0 to 3 is decoded 2048 pixels × 2048 pixels That is, the encoded image data according to the present embodiment is 2048 pixels from the size of 256 pixels × 256 pixels according to NL. Images of four types of sizes (resolution levels) up to × 2048 pixels can be reproduced.

  Here, in the present embodiment, a case will be described in which an image having a size of 128 pixels × 128 pixels is decoded and displayed using the encoded image data 1400 having the configuration shown in FIG. Here, conventionally, in order to achieve the same purpose, the minimum image (image of resolution level 0) is decoded and reduced to half in length and width. In the present embodiment, hierarchical coding processing is performed on the original LL subband to create a new LL subband. The processing after the creation of the new LL subband is the same as in the first embodiment.

  Here, when a request to decode the encoded image data 1400 stored in the hard disk 206 is generated, the flowchart of the processing according to the present embodiment performed by the client computer 102 is basically the one shown in FIG. However, the processing in steps S503 to S505 is slightly different. Below, the process in step S503-S505 performed by this embodiment is demonstrated.

  FIG. 15A is a flowchart showing details of the processing in step S503 according to the present embodiment. In the figure, the same steps as those in FIG.

  First, in step S1500, the LL subband of the encoded image data 1400 is decoded. Thereby, an image having a size of 256 pixels × 256 pixels is obtained. Next, in step S700, the number of tiles is obtained as described above. Next, in step S1501, when an image having a size of 256 pixels × 256 pixels is divided into tiles by the number of tiles, the number of hierarchies at the time of performing hierarchical coding is set for each tile. In the present embodiment, since hierarchical encoding that can represent two layers (two resolution levels) of an image having a size of 256 pixels × 256 pixels and an image having a size of 128 pixels × 128 pixels is performed, the number of encoding layers is “2”. Therefore, the number of hierarchies set in step S1501 is “2”.

  In step S1502, one tile is selected, and the selected tile is subjected to hierarchical encoding processing having two layers. In step S703, it is determined whether or not all layers have been encoded. If not, the process returns to step S1502. If so, the process proceeds to step S704, and all tiles are processed. Check if the encoding process has been performed. If all tiles have been encoded, this process ends. If all tiles have not been encoded, the process returns to step S1501, and the encoding process is not yet performed. Subsequent processing is performed for tiles that are not.

  Based on the above processing, based on an image having a size of 256 pixels × 256 pixels, encoded data capable of reproducing two resolution levels of an image having a size of 128 pixels × 128 pixels and an image having a size of 256 pixels × 256 pixels. Can be created. In the following, this created image of 128 × 128 pixels is used as the new LL subband.

  FIG. 15B is a flowchart showing details of the process in step S504 according to the present embodiment. The flowchart shown in FIG. 7 is obtained by omitting step S707 in the flowchart shown in FIG. 7B, and the processing that is basically performed is the same. That is, in the present embodiment, only the new LL subband used is different, and the rest is the same as in the first embodiment.

  Therefore, in the process in step S504, the number of bytes of the portion indicated by 603-1, 1404, 1405-1, and 1405-2 is stored in the Length field indicated by 1407-2. In a field indicated by 1407-1 (Offset field), Offset of the new encoded data 1403 from the head of the file storing the new encoded image data 1403 is stored.

  Thereafter, in order to indicate encoded data after Resolution Level = 1 in the encoded image data 1400, the field indicated by 1408-1 includes 1402-1 from the head of the file storing the encoded image data 1400. Stores the Resolution 1 offset. The field indicated by 1408-2 stores the encoded data 1402-2 and 1402-3 after Resolution 1 of 1402-1 and the total number of bytes of EOC indicated by 600-2.

  FIG. 16 is a flowchart showing details of the processing in step S505 according to the present embodiment. The flowchart shown in the figure is obtained by changing step S802 to step S1600 in the flowchart shown in FIG. 8A. Therefore, hereinafter, the processing in step S1600 will be described. In step S1600, the new main header acquired in step S801 is referred to, the encoded data of the new LL subband is specified, and this is read out.

  Through the above processing, an image encoded data file including each unit shown in FIG. 14 can be created. The new encoded data 1403 is different from the first embodiment in that Resolution 0 (1402-0) in the image encoded data 1400 is divided into two resolution levels (1405-1 and 1405-2).

  Further, as described above, in the fragment table 1407, the number of bytes of the portion indicated by 603-1, 1404, 1405-1, 1405-2 is stored in the Length field indicated by 1407-2. In a field indicated by 1407-1 (Offset field), Offset of the new encoded data 1403 from the head of the file storing the new encoded image data 1403 is stored.

  Thereafter, in order to indicate encoded data after Resolution Level = 1 in the encoded image data 1400, the field indicated by 1408-1 includes 1402-1 from the head of the file storing the encoded image data 1400. Stores the Resolution 1 offset. The field indicated by 1408-2 stores the encoded data 1402-2 and 1402-3 after Resolution 1 of 1402-1 and the total number of bytes of EOC indicated by 600-2.

  In order to indicate that the minimum resolution level (minimum image size) at which the new LL subband can be reproduced is “128 pixels × 128 pixels”, information indicating “128 pixels × 128 pixels is the minimum image size”. 1406 and the fragment table 1407 are associated using a mechanism such as an Association BOX in the case of JPEG2000.

  Finally, in step S805, the fragment table 1407 associated with the information 1406 and the new encoded data (1403 in FIG. 14) are stored in a BOX (new BOX), and the encoded image data 1400 is stored. The new BOX is written (appended) to the existing file.

  As described above, according to the present embodiment, the following points can be solved. That is, depending on the application, image data smaller than the lowest resolution image data stored in the hierarchical code data may be required. At this time, after decoding the image data of the lowest resolution, the image data of the desired image size is not obtained by simple reduction processing, but reduction processing and encoding are performed by hierarchical encoding, and the new LL subband image is obtained. By setting the size to the desired image size, it is possible to perform high-speed decoding even when image data of the size of 128 pixels × 128 pixels is required again in subsequent application processing. It becomes.

  Furthermore, since the encoding process by this hierarchical encoding targets image data with a small image size, it can be performed at high speed.

  In the present embodiment, the case where one new code is added to the encoded image data file has been described. However, a plurality of new codes may be added as required.

[Third Embodiment]
In the first and second embodiments, it is assumed that the encoded image data is stored in the hard disk 206 of the client computer 102, and the client computer 102 reads the encoded image data from the hard disk 206 and handles it. . In the present embodiment, it is assumed that the encoded image data is held in the image database 104 shown in FIG. In this case, as described in the first embodiment, the client computer 102 transmits an image acquisition request to the server 101 that manages the image database 104. At that time, the server 101 performs each process described in the first and second embodiments in accordance with the requested resolution level. In the present embodiment, the communication protocol between the client computer 102 and the server 101 is described as using JPIP.

  If JPIP / JPEG2000 encoded data is used, the user can receive only necessary data from the server 101 without acquiring all the image data in the server 101. As a unit of received data of the user, a JPEG2000 packet or a tile unit which is an encoding unit larger than the packet can be considered. For example, a packet unit is assumed as a data unit received by the user from the server 101. FIG. 9 shows a conceptual diagram of requests and responses in packet units.

  FIG. 9 is a diagram for explaining an exchange performed between the client computers 102 in order to obtain desired encoded data from the server 101. On the client computer 102 side, the tile number, resolution level, layer, component, and position number of the desired image are specified, and then the specified encoded data request 900 is transmitted to the server 101.

  Upon receiving this request 900, the server 101 analyzes the code stream of the encoded image data 903, extracts the packet data 911 corresponding to the tile number, resolution level, layer, component, and position number specified by the request 900, and sends the client Send to computer 102. The packet data 911 corresponds to the tile number, resolution level, layer, component, and position number specified by the request 900, as indicated by 910.

  Next, the structure of response data when data is transmitted and received using packet data as a transmission unit using JPIP will be described with reference to FIGS.

  FIG. 10 is a diagram illustrating a configuration example of a precinct data-bin configured by a packet data group. In the figure, reference numeral 1001 denotes this precinct data-bin. As shown in the figure, the precinct data-bin is composed of a plurality of packet data 1003.

  In JPIP, response data is created based on a JPEG2000 packet data block called “precinct data-bin”. The precinct data-bin is a lump of data in which packets of all layers constituting the resolution level rn of the precinct pn and the component number cn in Tile tn are arranged in an ascending order.

  FIG. 11 shows an example of JPIP response data created using Packet (tn, rn, cn, pn, 1) extracted from the precinct data-bin shown in FIG. JPIP response data includes a message header 1101 and a message body 1103.

  In message body 1103, JPEG2000 encoded data cut out from precinct data-bin 1001 is entered. That is, when the packet (tn, rn, cn, pn, 1) 1003 in FIG. 10 is extracted and JPIP response data is created, the packet (tn, rn, cn, pn, 1) 1003 is stored in the message body 1103. The message header 1101 is composed of four parameters.

  The first parameter contains an identifier indicating that the data contained in the message body 1103 is a precinct data-bin. The second parameter contains a precinct data-bin ID (PrID). This is uniquely determined from the tile number tn, resolution level number rn, component number cn, and precinct number pn, and can be obtained by the following equation.

PrID (tn, rn, cn, pn) = tn + (cn + s × (number of components)) × tile number However,
s = pn + tn x (number of precincts per tile at resolution level rn)
+ (total number of precincts for tile tr from resolution level 0 to (rn-1))
The third parameter is an offset value PrOffset 1002 in the response data-precinct data-bin. That is, the data contained in the message body 1103 is a value indicating the number of bytes starting from the data data bin 1001 of PrID (tn, rn, cn, pn), which is indicated by 1002 in FIG. It is the same as the value of PrOffst. PrOffst indicates an offset value from the head of the precinct data-bin. The last parameter of the message header 1101 is response data, that is, the byte length of the message body 1103. That is, ResLen 1102 contains the value of ResLen [byte] 1104.

  Here, it is assumed that the following request is issued from the client computer 102 to the server 101.

http://www.image.com/jpip.cgi?target=d.jp2&fsiz=256,256&type=jpp-stream
A process performed by the server 101 when such a request is received will be described with reference to FIG. 12 showing a flowchart of the process. It should be noted that a program and data for the CPU 201 of the server 101 to perform processing according to the flowchart of FIG. 10 are stored in the hard disk 206 of the server 101 and are used after the CPU 201 of the server 101 loads them into the RAM 205. Thus, the server 101 executes each process described below.

  First, in step S1201, the request is received. This request character string specifies the image encoded data file name stored in the server 101 by “target”, the size of the entire image by “fsiz”, and the transmission data format by “type” is doing.

  In step S1202, the request character string received from the client computer 102 in step S1201 is analyzed. This makes it possible to acquire conditions such as the requested file name, return data format, and the size of the image data to be returned. That is, in the case of the present embodiment, the image file having the file name “d.jp2” is requested to have a size that fits the size of 256 pixels × 256 pixels, and the return format is as follows. JPIP's precinct data-bin.

  In step S1203, response data corresponding to this request is generated. Details of the processing in step S1203 will be described later. In step S1204, the response data created in step S1203 is transmitted to the client computer 102.

  FIG. 13 is a flowchart showing details of response data creation processing in step S1203. In the figure, the same steps as those in FIG. 5 are denoted by the same step numbers, and the description thereof is omitted.

  First, in step S500, it is checked whether or not the same request was previously requested from the client computer 102 (whether or not this time is the first access). When JPIP communication is performed in a stateful manner, whether or not this is the first time can be easily determined in the request flow. On the other hand, in the case of stateless, it can be determined by using log information of the server.

  As a result of such check processing, if the same request was previously requested from the client computer 102, the processing proceeds to step S506. On the other hand, if the same request has not been requested from the client computer 102 last time, the process proceeds to step S501. In step S501, the image size described in the request received from the client computer 102 is acquired as the decoding resolution level.

  In step S502, the image encoded data file specified by the file name described in the request is referred to, and it is checked whether or not the acquired encoded data of the subband of the decoding resolution level exists. This check process is performed in substantially the same manner as in the first embodiment.

  As a result of this check process, if it is determined that it exists, the process proceeds to step S506. If it is determined that it does not exist, the process proceeds to step S1301. In step S1301, the processing from step S503 to step S505 is performed. As described in the second embodiment, the processing from step S503 to step S505 differs between the first embodiment and the second embodiment. Accordingly, the image size of the LL subband existing in the image encoded data file and, if present, the smaller one of the image sizes of the existing new LL subband are selected, and the selected image size is the image size of the decoding resolution level. If it is smaller, the processing in steps S503 to S505 according to the first embodiment is performed. If the magnitude relationship is reversed, the processing in steps S503 to S505 according to the second embodiment is performed. In the case of the present embodiment, since an image having a size of 256 pixels × 256 pixels is requested from the client computer 102, the processes in steps S503 to S505 according to the first embodiment are performed.

  Next, in step S1302, processing for specifying the method for reading the new LL subband created in step S1301 is performed. In this step, processing according to the flowchart of FIG. 8B is performed.

  In step S1303, the encoded data read by the reading method determined in any of steps S1302 and S506 is created as return data. Thereafter, the server 101 transmits this return data to the client computer 102.

  Here, in step S500, it is determined whether or not the current request is the first request in order to determine which read-out method is used for the future request in the first request. It is. Specifically, in the case of this embodiment in which image data having a size of 256 pixels × 256 pixels is requested by the first request, an enlargement request is generated due to a user request or the like, and then image data of 512 pixels × 512 pixels. When the server 101 is requested, the client computer 102 recognizes that the LL subband is encoded data of an image having a size of 256 pixels × 256 pixels, and therefore encodes an image having a size of 256 pixels × 256 pixels. It is necessary to return the code data of the difference from the data. Therefore, in the case of the second and subsequent requests, it is necessary to perform processing according to the reading method specified in the first time.

  As described above, even when the server 101 originally stores encoded data of an image having an LL subband size of 64 pixels × 64 pixels, the request from the client computer 102 is a size of 256 pixels × 256 pixels. In the case of the request starting from the encoded data of the image, since the return data is returned as if the LL subband is the encoded data of the image of 256 pixels × 256 pixels, the application on the client computer 102 side Particularly, in the decoder, decoding of an image size that is not used by the application of the client computer 102, specifically, 256 pixels × 256 without decoding image data of a size of 64 × 64 pixels and 128 × 128 pixels. Direct decoding of pixel-size image data Therefore, decoding can be performed at high speed.

  Furthermore, also in the amount of encoded data sent from the server 101, encoded data of three layers (size of 64 pixels × 64 pixels, size of 128 pixels × 128 pixels, size of 256 pixels × 256 pixels) is sent. Rather, it is possible to reduce the amount of data and return traffic on the network by returning the encoded data of one LL subband. Therefore, the throughput in the entire system can be improved.

  Here, according to each embodiment described above, the following effects can be expected.

  That is, the minimum resolution of the code data can be matched with the minimum resolution image size used by the application, and the time required for decoding can be shortened. Also, the application can easily determine which hierarchical code data in the code data should be decoded, such as when enlarging and displaying the image size from which the image is displayed first.

  Since this code conversion process is performed on image data having a small image size, it can be converted at high speed. Alternatively, in this code conversion process, the process of converting the code data in the middle layer already existing in the hierarchical code data to the LL subband does not perform the DWT process, and simply uses the EBCOT code with the image data as the LL subband. The calculation cost is low because only the conversion processing is performed.

  Furthermore, even when performing re-encoding processing to generate an LL subband having a size smaller than the image size of the LL subband of the code data, hierarchical encoding is performed for a small image size, so that high-speed processing is performed. Can be done.

  Further, depending on the usage, it is possible to decode the code data of all resolutions with the code data of the LL subband. In this case, the code data of each layer can be handled as independent code data.

  Furthermore, when the code data is requested from the server, the server side can automatically determine the request character string by using the JPIP mechanism. For this reason, the client can easily obtain the lowest resolution image data set by the application without being aware of the encoding options of the code data managed by the server, and by reducing the number of layers, It is possible to reduce the communication amount of codes returned from the server.

  Furthermore, since the encoded data managed by the server can be reused in response to a request from another client, the encoded data can be used efficiently.

  Alternatively, it is possible to negotiate before communication of image data between the server and the client. By doing this, it becomes possible to more accurately notify both states. Furthermore, this negotiation can be implemented as JPIP's “vendor capability” and can be implemented within the scope of the JPIP mechanism.

  The above embodiments may be used in combination as appropriate.

[Other Embodiments]
Needless to say, the object of the present invention can be achieved as follows. That is, a recording medium (or storage medium) that records a program code of software that implements the functions of the above-described embodiments is supplied to a system or apparatus. Then, the computer (or CPU or MPU) of the system or apparatus reads and executes the program code stored in the recording medium. In this case, the program code itself read from the recording medium realizes the functions of the above-described embodiment, and the recording medium on which the program code is recorded constitutes the present invention.

  Further, by executing the program code read by the computer, an operating system (OS) or the like running on the computer performs part or all of the actual processing based on the instruction of the program code. Needless to say, the process includes the case where the functions of the above-described embodiments are realized.

  Furthermore, it is assumed that the program code read from the recording medium is written in a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer. After that, based on the instruction of the program code, the CPU included in the function expansion card or function expansion unit performs part or all of the actual processing, and the function of the above-described embodiment is realized by the processing. Needless to say.

  When the present invention is applied to the recording medium, program code corresponding to the flowchart described above is stored in the recording medium.

1 is a schematic configuration diagram of a system according to a first embodiment of the present invention. 2 is a block diagram illustrating a hardware configuration of a computer applicable to client computers 102 and 103 and a server 103. FIG. It is a figure which shows the structural example of the image coding data obtained by carrying out the compression coding of the image using general JPEG2000. It is a figure which shows the structure of the subband obtained when a wavelet transformation is performed so that the image of six resolution levels can be reproduced. 4 is a flowchart of processing performed by a client computer when a request for decoding encoded image data stored in a hard disk is generated. It is a figure which shows the structural example of the image coding data which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the detail of the process in step S503. It is a flowchart which shows the detail of the process in step S504. It is a flowchart which shows the detail of the process in step S505. It is a flowchart which shows the detail of the process in step S506. FIG. 3 is a diagram for explaining an exchange performed between client computers 102 in order to obtain desired encoded data from a server 101. It is a figure which shows the structural example of precinct data-bin comprised by the packet data group. It is a figure which shows the example of JPIP response data produced using Packet (tn, rn, cn, pn, 1) extracted from the precinct data-bin shown in FIG. It is a flowchart of the process which the server 101 performs when a request is received. It is a flowchart which shows the detail of the production | generation process of the response data in step S1203. It is a figure which shows the structural example of the image coding data which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the detail of the process in step S503 which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the detail of the process in step S504 which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the detail of the process in step S505 which concerns on the 2nd Embodiment of this invention. It is a figure which shows the structure of the image coding data file which manages each BOX.

Claims (11)

Obtaining encoded data configured such that an image can be decoded at a plurality of resolution levels;
Obtaining the decoding resolution level;
A decoding step of decoding an image at the decoding resolution level using the encoded data;
An encoding step of encoding the image decoded in the decoding step;
A creation step of creating reference information to be referred to in order to use the image encoded in the encoding step, instead of a portion necessary for decoding the image at the decoding resolution level in the encoded data;
An image processing method comprising: a management step of managing the image encoded in the encoding step and the reference information together with the encoded data.   The image processing method according to claim 1, wherein in the management step, the image encoded in the encoding step, the encoded data, and the reference information are managed in one file.   In the creating step, the location information in the file of the image encoded in the encoding step, the location in the file of the encoded data other than the portion necessary for decoding the image at the decoding resolution level The image processing method according to claim 2, wherein a table in which information is registered in this order is created as the reference information.   The processing by the decoding step, the encoding step, and the creation step is processing performed when an image having an image size of the decoding resolution level is not managed together with the encoded data. 2. The image processing method according to 1.   2. The method according to claim 1, further comprising a step of decoding and outputting the image when the image having the image size of the decoding resolution level is already managed together with the encoded data. Image processing method. Obtaining encoded data configured such that an image can be decoded at a plurality of resolution levels;
Obtaining the decoding resolution level;
A decoding step of decoding an image having the lowest resolution level in the encoded data;
An encoding step of encoding the image decoded in the decoding step so that an image of an image size between the image size of the decoding resolution level and the image size of the lowest resolution level can be reproduced;
A creation step of creating reference information to be referred to in order to use the image encoded in the encoding step instead of the lowest resolution level image;
An image processing method comprising: a management step of managing the image encoded in the encoding step and the reference information together with the encoded data.   The minimum image size that the encoded data can be reproduced, and the size relationship between the smallest size of the images already managed in the management process together with the encoded data, and the image size of the decoding resolution level An image processing method according to any one of claims 1 to 5 and an image processing method according to claim 6. Means for obtaining encoded data configured to be capable of decoding an image at a plurality of resolution levels;
Means for obtaining the decoding resolution level;
Decoding means for decoding an image at the decoding resolution level using the encoded data;
Encoding means for encoding the image decoded by the decoding means;
Creating means for creating reference information to be referred to in order to use an image encoded by the encoding means instead of a portion necessary for decoding an image at the decoding resolution level in the encoded data;
An image processing apparatus comprising: a management unit that manages the image encoded by the encoding unit and the reference information together with the encoded data. Means for obtaining encoded data configured to be capable of decoding an image at a plurality of resolution levels;
Means for obtaining the decoding resolution level;
Decoding means for decoding an image having the lowest resolution level in the encoded data;
Encoding means for encoding the image decoded by the decoding means so that an image having an image size between the image size of the decoding resolution level and the image size of the lowest resolution level can be reproduced;
Creating means for creating reference information to be referred to in order to use the image encoded by the encoding means instead of the image of the lowest resolution level;
An image processing apparatus comprising: a management unit that manages the image encoded by the encoding unit and the reference information together with the encoded data.   A program causing a computer to execute the image processing method according to any one of claims 1 to 7.   A computer-readable storage medium storing the program according to claim 10.

Images