JP2004236298A - Image data processing apparatus, image data processing method, image data distribution apparatus and image data transmission system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image data processing apparatus and an image data processing method in which a reduced image can be displayed at high speed without deteriorating the picture quality, and to provide an image data distribution apparatus and an image data transmission system . <P>SOLUTION: From encoded image data stored in a buffer 120, a coefficient set extracting part 122 extracts low frequency side coefficients of the number corresponding to a reduction scale of the reduced image as a reduction coefficient set SMBn, and a DC coefficient extracting part 123 extracts only a DC coefficient DCn. A DC coefficient comparison part 125 compares the DC coefficient DCn with the neighboring DC coefficient. When the difference is great as a result of comparison, a worked MB generation part 127 sets the reduction coefficient set SMBn but when the difference is small, the DC coefficient DCn is set and zero is set to all the other coefficients, thereby generating a worked macro block KMBn. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

  The present invention relates to a processing device, a processing method, a distribution device, and a transmission system for coded image data, and more particularly to a technique for generating reduced coded image data for a reduced image from coded image data.

  2. Description of the Related Art In recent years, with the speeding up of networks and the rapid spread of personal computers, digital television receivers, and the like to ordinary households, image information providing services have been widely put into practical use. In particular, such image information is treated as digitized image data, and the amount of data becomes enormous. For this reason, it is necessary to efficiently store and transmit image data, and in the field of handling image data, a technique of compressing and using image data is used. Particularly, in the field of handling still images, an image compression encoding technique called JPEG (Joint Photographic Experts Group) is generally known.

  As an example using the image compression coding technique based on JPEG, there is a field of an electronic still camera generally called a digital camera. That is, in this electronic still camera, a captured image is compression-encoded in a JPEG format to reduce the amount of data and then recorded on a storage medium. As another example, with the development of network technology such as the Internet, an image distribution system that compresses and encodes image data in a JPEG format and distributes the image data has become widespread.

  2. Description of the Related Art In the field of electronic still cameras, generally, image data (hereinafter, referred to as JPEG images) held in a JPEG format in a camera is recorded on a recording medium such as an SD card. At this time, it is also common to create a plurality of JPEG images as reduced images for print reference and record them on a recording medium. Such a reduced image is called a thumbnail image, and an image obtained by collecting a plurality of these thumbnail images and printing them out at the same time is called an index print.

  Index printing requires restoring a plurality of JPEG images and generating a thumbnail image for each. That is, restoration of a plurality of JPEG images requires a considerable amount of processing, and index printing requires a considerable processing time.

  In order to solve this problem, an image processing technique capable of generating a thumbnail image from compression-encoded image data at high speed has been proposed (for example, see Patent Document 1). Patent Document 1 discloses a technique in which an original image is divided into blocks called macroblocks, and a thumbnail image is generated using only DC component coefficients included in encoded data obtained by encoding the spatial frequency components of each macroblock. are doing. In the related art (hereinafter referred to as the related art 1), since a thumbnail image is generated using only the DC component coefficients, the processing time of the AC component coefficients can be omitted, thereby improving the processing speed.

  Further, as described above, as another example using the image compression encoding technology, an image distribution system using a network such as the Internet is known, and as one example, an image distribution system using a thumbnail image is known. Are known. In an image distribution system using thumbnail images, the distribution side transmits thumbnail image data corresponding to the image size requested by the user as an index image for image selection, and the user selects a desired image from the thumbnail images. I do. Further, the distribution request of the image selected by the user is notified from the user to the distribution side. Thereby, the distribution side distributes requested content such as a moving image corresponding to the image selected by the user. In the field of image distribution systems, such image distribution systems have been put to practical use. Further, particularly in this field, there are various requests from users, and reduced images such as thumbnail images are distributed in various sizes. At this time, if the size of the delivered thumbnail image is small, many images can be viewed at the same time, but if the size is too small, it becomes difficult to recognize the image contents.

For this reason, a technique has been proposed in which a thumbnail image can be quickly checked and, if necessary, a thumbnail image is distributed in an image size according to a user's request (for example, see Patent Document 2). Patent Literature 2 divides an original image into macroblocks, selects and transmits low-frequency component coefficients included in encoded data obtained by encoding the spatial frequency components of each macroblock according to the image size to be restored. In addition, a technology for generating a thumbnail image of a desired size on the receiving side is disclosed. More specifically, in this method, first, an original image is divided into macroblocks, and an encoding conversion process is performed on an original image signal sampled with a plurality of sample values of each macroblock. For a plurality of spatial frequency coefficients obtained by this encoding conversion process, a plurality of low frequency component coefficients which are less than the plurality and include only low frequency components are selected as one coefficient set. The distribution side transmits this set of coefficients to the reception side for each macroblock. The receiving side applies a decoding transform to this set of coefficients to fit the size of the sample array for this number of coefficients. Thus, an image having sample values smaller than the sample value of the original image, that is, a reduced image is reproduced. The prior art disclosed in Patent Literature 2 (hereinafter referred to as “prior art 2”) can save transmission time and processing time when the image size can be small. Further, in the prior art 2, since the number of images increases to a higher spatial frequency component coefficient as the image reduction ratio increases and approaches 1, the coding coefficient shortage occurs when the required image size is large. To prevent image quality deterioration.
JP 2000-59612 A JP-A-58-75390

  However, when a relatively large thumbnail image is handled, in the related art 1 using only the DC component coefficients included in the encoded data, the image quality deteriorates conspicuously, and the number of low-frequency component coefficients included in the encoded data is large. In the prior art 2 of selecting the number corresponding to the number according to the image size, there is a problem that the processing time and the transmission speed are not so different from those of the original image.

  That is, in the prior art 1, since only the DC component coefficient corresponding to the pixel average value of each macroblock is used, so-called block noise increases as the image size increases. Further, in the prior art 2, as the image size increases, the number of coefficients closer to the number of the original image is handled, which causes a reduction in processing speed and transmission speed.

  Furthermore, the prior art 2 is a method of encoding and distributing an original image in accordance with a reduction ratio every time the original image is distributed, so it is necessary to perform encoding for each distribution. As a result, when the number of terminals on the receiving side increases, the load of the encoding process increases. That is, since it is necessary to perform encoding for each distribution, the processing load increases as the number of encoding processes increases, and such a processing load may cause a time delay from a distribution request to distribution. . Further, in the prior art 2, since an image is held as original image data that is not compression-encoded, a storage unit having a larger storage capacity is required as compared with the compression-encoded image data.

  In view of the above-described problems, the present invention provides an image data processing apparatus and an image data processing method capable of displaying thumbnail images at high speed without deteriorating image quality even when handling relatively large thumbnail images. It is an object of the present invention to provide a method, and an image data distribution device and an image data transmission system capable of transmitting thumbnail images.

  In order to solve the above problems, the image data processing apparatus of the present invention divides an original image into a plurality of blocks each including a plurality of pixels, and in each block, converts a spatial frequency component of the block into a plurality of spatial frequency An image data processing device that generates reduced coded image data that is coded data of a reduced image obtained by reducing the original image at a given reduction rate, from coded image data coded as a coefficient, Block extraction means for sequentially extracting each of the blocks as a processing target block, DC coefficient extraction means for extracting a DC component coefficient from the plurality of spatial frequency coefficients of the processing target block, DC component of at least one adjacent block adjacent to the target block in any of the horizontal direction, the vertical direction, and the oblique direction An adjacent DC coefficient extracting means for extracting a number as at least one adjacent DC component coefficient; comparing the DC component coefficient of the processing target block with the at least one adjacent DC component coefficient; DC coefficient comparing means for obtaining a value of a function having the DC component coefficient representing the variable and the at least one adjacent DC component coefficient as a variable, and when the value of the function is larger than a predetermined value, the plurality of blocks to be processed For the spatial frequency coefficients of, according to the reduction ratio, from the low frequency spatial frequency coefficients including the DC component coefficients, the number of spatial frequency coefficients less than the plurality is held, and other spatial frequency coefficients are set to zero. Coefficient set setting means to be set, and when the value of the function is less than the predetermined value, for the plurality of spatial frequency coefficients of the processing target block, Holds only serial DC component coefficients, the other spatial frequency coefficients is characterized in further comprising a DC coefficient setting means for setting a zero.

  With such a configuration, the image data processing apparatus according to the present invention provides an image data processing apparatus according to the present invention, in which the brightness and color change of an image are gradual and the value of the function is small in the coded image data of the original image. In the region where the value of the function is large, which is a boundary region such as the outline of an image, only the DC component coefficient is used, and in the block, only the number of low frequency side coefficients corresponding to the image reduction ratio is used. Thus, reduced encoded image data for a reduced image can be generated. That is, based on the visual characteristics of an image, the image data processing apparatus of the present invention first uses only the DC component coefficient for an area in which block noise and the like are not conspicuous and the luminance and color change of the image are gradual. Thus, the number of effective coefficients is reduced, and the amount of coding is reduced. Furthermore, the image data processing apparatus according to the present invention, based on the visual characteristics of the image, reduces the number of low-frequency signals corresponding to the image reduction ratio for a boundary region such as an image contour where image quality deterioration due to coefficient reduction is conspicuous. By selecting the coefficients on the side, the number of effective coefficients is reduced while suppressing the deterioration of the image quality due to the reduction of the coefficients, and the amount of coding is reduced. As described above, the image data processing apparatus of the present invention adaptively changes the number of effective spatial frequency coefficients according to the image reduction ratio and whether or not the image is in the boundary region of the image. Thus, the processing load required to display the reduced image can be reduced, and the reduced image can be displayed while suppressing the deterioration of the image quality. Therefore, by using the reduced coded image data generated by the image data processing apparatus of the present invention, even when a relatively large reduced image is handled, the reduced image can be displayed at high speed without deteriorating the image quality. It is possible to do. Preferably, the at least one adjacent block is one adjacent block, and the value of the function is an absolute value of a difference value between the DC component coefficient and the adjacent DC component coefficient.

  Alternatively, the at least one adjacent block is a plurality of adjacent blocks, and the value of the function is an average value of a plurality of adjacent DC component coefficients included in the plurality of adjacent blocks, and the DC value of the processing target block. It is desirable to be the absolute value of the difference value from the component coefficient.

  Alternatively, the at least one neighboring block is a plurality of neighboring blocks, and the value of the function is obtained by performing a weighted average calculation based on a neighboring direction on a plurality of neighboring DC component coefficients included in the plurality of neighboring blocks. It is desirable to be an absolute value of a difference value between the obtained weighted average value and the DC component coefficient.

  Alternatively, the at least one adjacent block is a plurality of adjacent blocks, and the value of the function is a value of each of the adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficient of the processing target block. It is desirable to set the largest value among the absolute values of the difference values.

  Alternatively, the at least one adjacent block is a plurality of adjacent blocks, and the value of the function is a value of each of the adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficient of the processing target block. Desirably, it is the sum of the absolute values of the difference values.

Also preferably, the ratio of the number of the spatial frequency coefficients less than the plurality retained by the coefficient set setting means to the number of the plurality of spatial frequency coefficients is a coefficient reduction rate,
The coefficient set setting unit divides the reduction rate into a plurality of steps, sets the coefficient reduction rate according to each of the reduction rate steps, and, when the value of the function is larger than the predetermined value, According to the coefficient reduction rate, the number of the spatial frequency coefficients smaller than the plurality is held, and zero is set to the other spatial frequency coefficients.

  More preferably, the image data processing device includes an input processing unit that receives an input of the reduction rate, a decoding unit that decodes the generated reduced encoded image data, and the decoded reduced encoded image that has been decoded. The image processing apparatus further includes an image size reducing unit that reduces an image size of data, and a screen display unit that displays the reduced coded image data on a screen.

  Preferably, the encoded image data is color encoded image data, and the predetermined value is given for each color component.

  Preferably, the coded image data is color coded image data, and the coefficient set setting unit determines the number of the spatial frequency coefficients less than the plurality for each color component.

  Further, it is preferable that the encoded image data is color encoded image data, and the block extracting unit extracts the processing target block only for a predetermined color component.

  Preferably, the coded image data is data obtained by performing variable length coding, and the coefficient set setting means and the DC coefficient setting means indicate the end of a block after all of the spatial frequency coefficients to be held. By assigning a sign and deleting the other spatial frequency coefficient, it represents that the other spatial frequency coefficient is set to zero.

  In the image data processing method of the present invention, the original image is divided into a plurality of blocks each including a plurality of pixels, and in each block, a spatial frequency component of the block is encoded as a plurality of spatial frequency coefficients. An image data processing method for generating reduced coded image data, which is coded data of a reduced image obtained by reducing the original image at a given reduction rate, from coded image data, comprising: A first step of extracting a DC component coefficient, and determining at least one DC component coefficient of at least one adjacent block adjacent to any of the blocks of the extracted DC component coefficient in a horizontal direction, a vertical direction, and an oblique direction. A second step of extracting as two adjacent DC component coefficients, comparing the DC component coefficient with the at least one adjacent DC component coefficient A third step of determining a value of a function having the DC component coefficient representing the degree of difference between the two and the at least one adjacent DC component coefficient as a variable, and when the value of the function is larger than a predetermined value, For the plurality of spatial frequency coefficients, in accordance with the reduction ratio, from the low frequency spatial frequency coefficients including the DC component coefficients, hold a smaller number of spatial frequency coefficients than the plurality, and to the other spatial frequency coefficients Is a fourth step of setting zero, and when the value of the function is less than the predetermined value, for the plurality of spatial frequency coefficients, only the DC component coefficient is held, and the other spatial frequency coefficients are A fifth step of setting zero, wherein each of the steps is performed on each of the blocks to reduce the original image at the reduction rate. It is characterized in that to generate the image data.

  With such a configuration, the image data processing method according to the present invention provides an image data processing method according to the present invention, in which the brightness and color change of the image are gradual with respect to the coded image data of the original image, so that the value of the function is small in the block. In the region where the value of the function is large, which is a boundary region such as the outline of an image, only the DC component coefficient is used, and in the block, only the number of low frequency side coefficients corresponding to the image reduction ratio is used. Thus, reduced encoded image data for a reduced image can be generated. That is, the image data processing method of the present invention, based on the visual characteristics of an image, first uses only the DC component coefficient for an area where the block noise and the like are not conspicuous and the luminance and the color change are gradual. Thus, the number of effective coefficients is reduced, and the amount of coding is reduced. Furthermore, the image data processing method according to the present invention, based on the visual characteristics of the image, reduces the number of low-frequency signals corresponding to the image reduction rate to a boundary region such as an image contour where image quality degradation due to coefficient reduction is conspicuous. By selecting the coefficients on the side, the number of effective coefficients is reduced while suppressing the deterioration of the image quality due to the reduction of the coefficients, and the amount of coding is reduced. As described above, the image data processing method of the present invention adaptively changes the number of effective spatial frequency coefficients according to the image reduction ratio and whether or not the image is in the boundary region of the image. In addition to this, it is possible to generate reduced coded image data for a reduced image in which deterioration in image quality is suppressed. Therefore, by using the reduced coded image data generated by the image data processing method of the present invention, even when a relatively large reduced image is displayed, the processing load until the reduced image is displayed is reduced. At the same time, it is possible to reduce image quality deterioration. Preferably, the at least one adjacent block is one adjacent block, and the value of the function is an absolute value of a difference value between the DC component coefficient and the adjacent DC component coefficient.

  Alternatively, the at least one adjacent block is a plurality of adjacent blocks, and the value of the function is a difference between an average value of a plurality of adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficient. Desirably the absolute value of the value.

  Alternatively, the at least one neighboring block is a plurality of neighboring blocks, and the value of the function is obtained by performing a weighted average calculation based on a neighboring direction on a plurality of neighboring DC component coefficients included in the plurality of neighboring blocks. It is desirable to be an absolute value of a difference value between the obtained weighted average value and the DC component coefficient.

  Alternatively, the at least one adjacent block is a plurality of adjacent blocks, and the value of the function is an absolute value of each difference value between each adjacent DC component coefficient included in the plurality of adjacent blocks and the DC component coefficient. It is desirable to set the largest value among the above.

  Alternatively, the at least one adjacent block is a plurality of adjacent blocks, and the value of the function is an absolute value of each difference value between each adjacent DC component coefficient included in the plurality of adjacent blocks and the DC component coefficient. Is desirable.

Also preferably, the ratio of the number of the spatial frequency coefficients smaller than the plurality of the plurality of spatial frequency coefficients held by the fourth step to the number of the plurality of spatial frequency coefficients is a coefficient reduction rate,
The fourth step divides the reduction rate into a plurality of steps, sets the coefficient reduction rate according to each of the reduction rate steps, and, when the value of the function is larger than the predetermined value, According to the coefficient reduction rate, the number of the spatial frequency coefficients smaller than the plurality is held, and zero is set to the other spatial frequency coefficients.

  More preferably, the image data processing method includes a sixth step of receiving the input of the reduction ratio, a seventh step of decoding the generated reduced coded image data, and a step of decoding the decoded reduced code. An eighth step of reducing the image size of the coded image data and a ninth step of displaying the reduced coded image data on a screen are further provided.

  Preferably, the encoded image data is color encoded image data, and the predetermined value is given for each color component.

  Preferably, the encoded image data is color encoded image data, and in the fourth step, the number of the spatial frequency coefficients smaller than the plurality is determined for each color component.

  Further, it is preferable that the encoded image data is color encoded image data, and in the image data processing method, the first to fifth steps for each block are performed only for a predetermined color component.

  Preferably, the coded image data is data obtained by performing variable length coding, and the fourth step and the fifth step indicate the end of a block after all of the spatial frequency coefficients to be held. By assigning a sign and deleting the other spatial frequency coefficient, it represents that the other spatial frequency coefficient is set to zero.

  Further, the image data distribution device of the present invention divides the original image into a plurality of blocks each including a plurality of pixels, and in each block, encodes a spatial frequency component of the block as a plurality of spatial frequency coefficients. Generating, from the encoded image data, reduced encoded image data that is encoded data of a reduced image obtained by reducing the original image at a given reduction rate, and transmitting the reduced encoded image data via a network; An image data distribution device, comprising: encoded image data storage means for storing a plurality of the encoded image data; and the plurality of encoded image data stored in the encoded image data storage means, into one image. The image data processing apparatus according to the present invention further includes a coded image data reading unit that reads corresponding coded image data, and an image data processing device according to the present invention. The reduced encoded image data is generated from the encoded image data read by the encoded image data reading unit, and the image data distribution device transmits the reduced encoded image data generated by the image data processing device to the network. And transmitting means for transmitting the information.

  Furthermore, an image data transmission system according to the present invention is an image data transmission system comprising: the image data distribution device according to the present invention; and a terminal device that can connect the image data distribution device and itself via the network. A terminal for receiving the reduced coded image data transmitted by the transmitting unit; a decoding unit for decoding the received reduced coded image data; The image processing apparatus further includes: an image size reducing unit configured to reduce an image size of the reduced reduced encoded image data; and a screen display unit configured to display the reduced reduced encoded image data on a screen.

  With such a configuration, the image data distribution device and the image data transmission system according to the present invention provide an area in which the value of the above function is small due to a gradual change in luminance and color of the image with respect to the encoded image data of the original image Then, only the DC component coefficients of the block are used, and in a region where the value of the function is large, which is a boundary region such as the outline of an image, only the number of low-frequency coefficients corresponding to the image reduction ratio in the block are used. In this manner, reduced encoded image data for a reduced image can be generated and transmitted. That is, the image data distribution device and the image data transmission system according to the present invention are based on the visual characteristics of an image. By using only the number, the number of effective coefficients is reduced, and the amount of coding is reduced. Furthermore, the image data transmission system according to the present invention, based on the visual characteristics of the image, reduces the number of low-frequency signals corresponding to the image reduction ratio to a boundary region such as an image contour where image quality degradation due to coefficient reduction is conspicuous. By selecting the coefficients on the side, the number of effective coefficients is reduced while suppressing the deterioration of the image quality due to the reduction of the coefficients, and the amount of coding is reduced. As described above, the image data distribution device and the image data transmission system of the present invention adaptively change the number of effective spatial frequency coefficients depending on whether the image is a reduction ratio and an image boundary region, By reducing the amount of coding, the transmission load when transmitting the reduced image is reduced, and the transmission of the reduced image in which the deterioration of the image quality is suppressed is enabled. Therefore, by using the reduced coded image data generated by the image data distribution device and the image data transmission system of the present invention, even when a relatively large reduced image is handled, high-speed operation can be performed without deteriorating the image quality. Thus, the terminal device can quickly display a reduced image of a desired reduced size in which deterioration in image quality is suppressed, in the terminal device. Preferably, in the image data transmission system, the terminal device sends a request for designating encoded image data corresponding to one image from the plurality of encoded image data stored in the encoded image data storage unit. Further comprising: an image request inputting means for accepting; and an image request transmitting means for transmitting the accepted request to the image data distribution device via the network, wherein the image data distribution device receives the transmitted request. Receiving means is further provided, wherein the coded image data reading means reads coded image data corresponding to the request from the plurality of coded image data stored in the coded image data storage means.

  More preferably, the terminal device further includes a reduction ratio input unit that receives an input of the reduction ratio, and a reduction ratio transmission unit that transmits the received reduction ratio to the image data distribution device through the network. The image data distribution device further includes a reduction ratio receiving unit that receives the transmitted reduction ratio, and the image data processing device sets the received reduction ratio as the given reduction ratio, and sets the received reduced image data as the given reduction ratio. Generate

  As described above, when generating a reduced image such as a thumbnail image, the present image data processing apparatus uses a block in an area where the luminance or color change of the image is gradual with respect to the encoded image data of the original image. In the boundary region such as the outline of the image, only the low frequency side coefficients corresponding to the image reduction ratio are used to generate reduced encoded image data for the reduced image. I do. That is, the present image data processing apparatus adaptively changes the reduction ratio of the image and the number of effective spatial frequency coefficients according to whether or not the image is in the boundary region based on the visual characteristics of the image. Even in the case of displaying a very small image, it is possible to reduce the processing load until the reduced image is displayed by reducing the coding amount, and also to reduce the image quality deterioration. Accordingly, the present image data processing apparatus can display thumbnail images at high speed without deteriorating image quality even when handling relatively large thumbnail images.

  In addition, when generating a reduced image such as a thumbnail image, the present image data processing method has a DC component coefficient of the block in an area where the luminance and color change of the image are gradual with respect to the encoded image data of the original image. Only in the boundary area such as the outline of the image, the reduced coded image data for the reduced image is generated by using only the number of low-frequency coefficients corresponding to the reduction ratio of the image. That is, the present image data processing method adaptively changes the number of effective spatial frequency coefficients according to the image reduction ratio and whether or not the image is in the boundary region based on the visual characteristics of the image. It is possible to generate reduced coded image data for a reduced image in which the amount of conversion is reduced and the deterioration of the image quality is suppressed. Furthermore, by using such reduced coded image data, even when a relatively large reduced image is displayed, the processing load required to display the reduced image and the image quality deterioration are reduced. Is also possible.

  Further, when transmitting the reduced image such as the thumbnail image, the present image data distribution device and the present image data transmission system, in the area where the luminance and color change of the image are gradual with respect to the encoded image data of the original image, Only the DC component coefficients of the block are used, and in the boundary area such as the outline of the image, only the low frequency side coefficients corresponding to the image reduction rate are used to reduce the reduced encoded image data for the reduced image. Is generated and transmitted. That is, the present image data transmission system adaptively changes the reduction ratio of the image and the number of effective spatial frequency coefficients depending on whether or not the image is in the boundary region based on the visual characteristics of the image. Even in the case of transmitting a very large reduced image, the transmission load is reduced by reducing the amount of coding, and the deterioration of image quality can be reduced. Thus, the present image data distribution apparatus and the present image data transmission system can transmit thumbnail images at high speed without deteriorating image quality even when handling relatively large thumbnail images.

  Therefore, according to the present invention, even when a relatively large thumbnail image is handled, an image data processing device, an image data processing method, and an image data processing method capable of displaying a thumbnail image at high speed without deteriorating image quality are provided. An image data distribution device and an image data transmission system capable of transmitting thumbnail images can be provided.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(Embodiment 1)
FIG. 1 is a block diagram showing an overall configuration of an image data processing device according to Embodiment 1 of the present invention. The image data processing apparatus 1 extracts image data specified by a user from a storage unit that stores image data compressed and encoded by the JPEG method or the like, and further extracts the image data at an image size of a reduction rate specified by the user. Is displayed.

  In FIG. 1, an input unit 18 is a keyboard, a remote controller, or the like, and various instructions from the user are input by operating the input unit 18 by the user. The input processing unit 17 notifies the user instruction determination unit 16 of the instruction input by the user as instruction information. A user instruction determination unit (hereinafter, referred to as determination unit) 16 analyzes the content of the instruction from the user based on the instruction information. Further, the determination unit 16 instructs each unit to execute a process according to the analysis result, and notifies information for that. For example, when a user instructs to display a reduced image, the determination unit 16 outputs information such as the image file name and image size designated by the user at that time. In FIG. 1, in order to perform a process of reducing an image which is a feature of the image data processing apparatus 1, the determination unit 16 displays the image with the image identification information for searching for the specified image file and the specified image size. For outputting image size information for the purpose.

  The coded image file storage means (hereinafter, referred to as storage means) 10 functions as coded image data storage means, has a recording medium such as a hard disk drive, and is compression-coded by the JPEG method or the like. Image data (hereinafter, referred to as encoded image data) is stored in the recording medium as an image file. The stored image file is obtained from the Internet, a CD-ROM (Compact Disc-ROM), a DVD (Digital Video Disc), an electronic still camera, or the like by a user operation. Thus, the storage means 10 stores a plurality of image files as encoded image files of the original image. An encoded image data reading unit (hereinafter, referred to as a reading unit) 11 reads an image file corresponding to image identification information from a plurality of image files stored in the storage unit 10. The read image file is subjected to various processes, and is displayed by the screen display unit 15 on a monitor 19 functioning as a display unit.

  In this way, the user can specify and display a desired image file from the image files stored in the storage unit 10 in a plurality of files. Further, when searching for a desired image from a large number of image files stored in the storage means 10, the user can easily search for the desired image by using an index display for displaying a plurality of thumbnail images. . Further, the image data processing apparatus 1 is characterized in that the display size of the thumbnail image can be specified. That is, the image data processing apparatus 1 is characterized in that it has means for generating a thumbnail image with an image size of a reduction ratio designated by the user.

  The encoded image data processing means (hereinafter referred to as processing means) 12 converts the image file read by the reading means 11, that is, the encoded image data corresponding to one image, into a thumbnail image, that is, image data for a reduced image. It is a means for processing so as to be converted. The processing unit 12 converts the encoded image data into reduced encoded image data that is encoded image data corresponding to the reduced size based on the image size information from the determination unit 16.

  The decoding unit 13 decodes the encoded image data with respect to the reduced encoded image data, and restores decoded image data having the same size as the original image. An image size reducing unit (hereinafter referred to as a reducing unit) 14 generates reduced image data by thinning out pixels of the decoded image data restored by the decoding unit 13 according to an image size specified by a user. I do. The screen display means 15 outputs the reduced image data to the monitor 19 so that the reduced image data is displayed on the screen of the monitor 19.

  As described above, the image data processing apparatus 1 according to the first embodiment reads out the encoded image data of the original image stored in the storage unit 10, and uses the processing unit 12 to output the reduced code for the reduced image corresponding to the image size information. The thumbnail image of the original image is displayed on the monitor 19 by generating encoded image data, decoding and reducing the reduced encoded image data.

  FIG. 2 is a block diagram showing a detailed configuration of the processing means 12 which is a feature of the image data processing apparatus in the present embodiment. In FIG. 2, a coded image data buffer (hereinafter, referred to as a buffer) 120 temporarily stores coded image data designated by a user and read by the reading unit 11.

  A macroblock extraction unit (hereinafter, referred to as an MB extraction unit) 121 sequentially extracts encoded image data corresponding to each macroblock from the buffer 120. FIG. 2 illustrates a case where the MB extracting unit 121 extracts the n-th macroblock MBn. Since the following processing is performed on the extracted macro block MBn, the extracted macro block MBn is referred to as a processing target block. For example, when the encoding method is the JPEG method, the unencoded original image data is divided into blocks each having 8 × 8 pixels, and discrete cosine transform is performed on each block. As a result, A spatial frequency coefficient corresponding to each spatial frequency component of each block is obtained. For one block, the spatial frequency coefficient ranges from a DC component coefficient (hereinafter, referred to as a DC coefficient), which is a DC component coefficient, to an AC coefficient, which is a AC component coefficient, from a component having a high spatial frequency. × 8 coefficients are obtained. These 8 × 8 coefficients are configured as a macroblock which is one block. The encoded image data is configured by combining a plurality of such macroblocks. Thus, when the encoded image data is in the JPEG format, the MB extracting unit 121 extracts 8 × 8 coefficients of the macroblock MBn when processing the nth macroblock.

  The coefficient set extraction unit 122 extracts, from a plurality of coefficients included in the macroblock MBn, a number of coefficients corresponding to a reduction ratio for reducing the original image as a reduced coefficient set. This reduced coefficient set is smaller than the number of coefficients included in the macroblock MBn, and is extracted from low-frequency coefficients including DC coefficients. The coefficient set extraction unit 122 outputs the reduced coefficient set thus extracted as a reduced coefficient set SMBn.

  The DC coefficient extraction unit 123 extracts only DC coefficients from a plurality of coefficients included in the macroblock MBn, and outputs DC coefficients DCn as DC component coefficients.

  The adjacent DC coefficient extraction unit 124 extracts an adjacent DC component coefficient (hereinafter, referred to as an adjacent DC coefficient) of a macroblock adjacent to the macroblock MBn to be processed. Here, an example is shown in which adjacent DC coefficients DCn−1 and DCn + 1 are extracted as adjacent DC coefficients of macroblocks adjacent to the left and right in the horizontal direction of an image. Note that the present invention is not limited to macroblocks adjacent in the horizontal direction, and it is also possible to use adjacent DC coefficients of macroblocks adjacent in the vertical or oblique directions.

  The DC coefficient comparison unit 125 is supplied with the DC coefficient DCn from the DC coefficient extraction unit 123 and the adjacent DC coefficients DCn−1 and DCn + 1 from the adjacent DC coefficient extraction unit 124, and compares and determines these DC coefficients. Do. First, as a DC coefficient comparison, the DC coefficient comparison unit 125 compares the magnitude of the DC coefficient DCn of the processing target block with the adjacent DC coefficients DCn−1 and DCn + 1 to be compared. When a plurality of adjacent macroblocks are used, for example, an average value of each adjacent DC coefficient is obtained, and the average value is compared with the corresponding DC coefficient DCn as an average DC coefficient. Further, for example, the weight value may be changed depending on the adjacent direction, and the weighted average value may be used as the average DC coefficient. As still another example, the magnitudes of the corresponding DC coefficient DCn and each of the adjacent DC coefficients may be compared, and the difference having the largest absolute value of the difference value may be used as the comparison result. Alternatively, the comparison result may be the sum of the absolute values of the difference values, which are the differences between the DC coefficient DCn and the adjacent DC coefficients. It is also possible to use only one adjacent macroblock. In this case, the absolute value of the difference value that is the difference between the DC coefficient DCn and the adjacent DC coefficient may be used as the comparison result. As described above, the DC coefficient comparison unit 125 compares the magnitude of the target DC coefficient DCn with the magnitude of the adjacent DC coefficient, and detects a difference value that is the result of the comparison.

  As an example of the weighted average value, 24 blocks arranged around a macroblock MBn in a 5 × 5 block centered on the macroblock MBn to be processed are set as adjacent blocks, and weighted according to the distance from the macroblock MBn. You may. In this case, the weight should be reduced as the distance from the macroblock MBn increases. Alternatively, the weight may be set to 1 for macroblocks vertically and horizontally adjacent to the macroblock MBn, and set to 1 / √2 for macroblocks adjacent diagonally. In addition, the weight in the specific direction can be increased according to the characteristics of each image. For example, weighting such as lowering the upper right direction and lowering the upper left direction can be performed.

  The DC coefficient comparison unit 125 generally includes a function (for example, DCn, DCn-1, and DCn + 1) indicating the degree of difference between the DC coefficient DCn and at least one adjacent DC coefficient (for example, DCn−1 and DCn + 1). Function) is calculated. Absolute values of the above-described various difference values correspond to specific examples of the value of the function.

  Next, the DC coefficient comparing unit 125 compares the magnitude of the detected absolute value of the difference value with a predetermined determination value. At this time, if the difference value is larger than the determination value, a boundary detection result indicating that a boundary has been detected is notified to the selection unit 126 described below. When the absolute value of the difference value is smaller than the determination value, the selection unit 126 is notified of a boundary detection result indicating that no boundary has been detected. As described above, the DC coefficient comparison unit 125 determines whether the target macroblock MBn is a boundary portion of an image such as a contour line by comparing the target DC coefficient DCn with an adjacent DC component. I have. That is, since the DC coefficient indicates the average value of each pixel value of one macroblock, it is determined that a macroblock having a large difference in pixel average value between macroblocks is a boundary region of an image. ing. Such a determination result of the boundary region of the image is notified to the selection unit 126 as a boundary detection result.

  The selection unit 126 extracts the reduced coefficient set SMBn from the coefficient set extraction unit 122 or the DC coefficient DCn from the DC coefficient extraction unit 123 based on the boundary detection result. When notified of a boundary detection result indicating that a boundary has been detected, the selection unit 126 selects the reduction coefficient set SMBn. When notified of a boundary detection result indicating that no boundary has been detected, the selection unit 126 selects the DC coefficient DCn. The selection unit 126 supplies the extracted and extracted coefficients to a processed macro block generation unit (hereinafter, referred to as a processed MB generation unit) 127.

  The processed MB generation unit 127 sets the extracted reduced coefficient set SMBn or DC coefficient DCn among the same number of coefficients as the number of pixels in the macro block MBn as a coefficient corresponding to the frequency, and sets the other frequencies. Are all set to zero and output as the processed macroblock KMBn.

  As shown in FIG. 2, the coefficient set extraction unit 122, the selection unit 126, and the processed MB generation unit 127 function as the coefficient set setting unit 132 of the present invention, and the selection unit 126 and the processed MB 127 It functions as coefficient setting means 133.

  The detailed operation of the image data processing device 1 according to the first embodiment configured as described above will be sequentially described below. Here, the operation of the encoded image data processing means 12 which is a feature of the present image data processing apparatus will be mainly described.

  FIG. 3 is a diagram showing the state of macroblocks and coefficients processed by the processing means 12 described in FIG. 3, the encoded image data 300 is an example of the encoded image data temporarily stored in the buffer 120 in FIG. Here, an example of an image of 32 × 32 pixels is shown as an original image. This image is composed of 4 × 4 macroblocks by being divided into macroblocks of 8 × 8 pixels. Further, each macroblock is converted into 8 × 8 spatial frequency coefficients by, for example, a discrete cosine transform, and a macroblock MB is configured as a set of 64 coefficients. In each macro block MB, these coefficients are composed of a DC coefficient DC and a plurality of AC coefficients [AC] corresponding to each frequency (hereinafter, a plurality of coefficients are shown using parentheses []). Have been. FIG. 3 shows a case where the macroblock MB to be processed is a macroblock MBn that is the n-th macroblock. Hereinafter, a description will be given of the macro block MBn as a representative, but a similar process is performed on each macro block.

  As described above, even when a reduced image such as a thumbnail image is displayed, the image data processing device 1 reads the encoded image data 300 corresponding to the original image from the storage unit 10 and temporarily stores the encoded image data 300 in the buffer 120.

  Next, in order to display a reduced image at a reduction ratio in accordance with an instruction from the user, the processing means 12 performs processing of each macroblock of the encoded image data 300. The MB extracting unit 121 sequentially extracts encoded image data included in each macroblock from the buffer 120. The macroblock MBn301 to be processed in FIG. 3 is the current macroblock to be processed as the nth macroblock. Further, in the coefficient set extracting unit 122, a reduced area is set in order to generate a reduced coefficient set SMBn corresponding to the reduced image. This reduction area is set according to the reduction ratio of the display image. Here, when the area of the image is, for example, a reduced image of 1 / K, the case where the reduced area is also set to 1 / K is shown. More specifically, when it is desired to make the image half the height and width, the reduction area is also set to half the height and width. Here, the vertical means the vertical direction of the image, and the horizontal means the horizontal direction. In FIG. 3, 8 × 8 coefficients are included in the macroblock, and this reduced area covers 4 × 4 coefficients which are half the length and width, and the spatial frequency coefficients which are sequentially lower from the DC coefficient are used. . The coefficient set extraction unit 122 extracts a target coefficient in the reduced area, and outputs it as a reduced coefficient set SMBn302 as shown in FIG.

  The ratio of the size of the reduced image to the size of the original image is defined as the reduction ratio, and the number of spatial frequency coefficients for selecting a smaller number of spatial frequency coefficients than the number of spatial frequency coefficients included in the macroblock is selected. Assuming that the ratio of the numbers is the coefficient reduction rate, as illustrated in FIG. 4, the reduction rate may be divided into a plurality of stages, and the coefficient reduction ratio may be set according to each reduction ratio stage. Further, nonlinear conversion may be performed as illustrated in FIG. 5. For example, when the reduction ratio is set to １, the coefficient reduction ratio may be set to 25/64. In this case, the size of the image is half the length and width, and if the macroblock includes 8 × 8 coefficients, the reduction area includes 5 × 5 coefficients. Such conversion is effective when it is desired to prevent the image quality from being deteriorated due to the reduction. Conversely, for example, when the size of an image is set to half the length and width, for a macro block including 8 × 8 coefficients, the reduction area may be set to include 3 × 3 coefficients. . Such conversion is effective when emphasizing processing speed or the like rather than image quality.

  Returning to FIG. 3, the DC coefficient DCn extracted by the DC coefficient extraction unit 123 is subjected to the size comparison with the adjacent DC coefficient by the DC coefficient comparison unit 125 as described above, and the boundary is detected. The selection unit 126 is notified of the boundary detection result. Further, the selection unit 126 selects between the reduced coefficient set SMBn from the coefficient set extraction unit 122 and the DC coefficient DCn from the DC coefficient extraction unit 123 based on the boundary detection result. Further, the processed MB generating unit 127 generates a processed macro block KMBn that has the same number of coefficients as the macroblock MBn, but has zero set for all coefficients other than the reduced coefficient set SMBn or the DC coefficient DCn. You. The processed macro block KMBn 303 shown in FIG. 3 is an example of the processed macro block KMBn generated in this manner.

  As described above, the coded image data processing unit 12 processes the coded image data as it is, without decoding the coded image data, and generates reduced coded image data. Accordingly, when the coded image data of the macroblock MBn extracted by the MB extraction unit 121 and the coded image data of the processed macroblock KMBn output by the processed MB generation unit 127 are exemplified, a bit string as shown in FIG. 6 is obtained. . That is, in the encoded image data of the macro block MBn, the bit sequence of the AC coefficient is arranged in order from the low frequency side following the bit sequence of the DC coefficient. The coded image data of the processed macro block KMBn output from the processed MB generation unit 127 is such that, from the bit string of the coded image data of the macro block KMBn, a portion from the beginning to a certain portion is valid, and the rest is discarded. . If the bit string of the macroblock MBn has been subjected to variable-length coding (Huffman coding in the example of JPEG), the processed MB generating unit 127 adds the end of the bit string of one macroblock to the end of the remaining bit string. An EOB (End Of Block; for example, “1010” in binary), which is the code shown, is inserted, and the remaining bit string is deleted. The EOB indicates that the remaining bit sequence that has been deleted is zero, that is, all the coefficients other than the reduced coefficient set SMBn or the coefficients other than the DC coefficient DCn are set to zero. If the bit sequence of the macroblock MBn is before the variable-length coding, the processed MB generation unit 127 expresses that all the remaining bit sequences are 0 in run length. In any case, the encoded image data of the processed macroblock KMBn is compressed into data having a smaller number of bits than the encoded image data of the macroblock MBn. As a result, the reduced coded image data can reduce the load on the transmission medium accompanying the transmission, and can also reduce the time required for transmission.

  By the processing as described above, the processing unit 12 processes the coded image data 300 into reduced coded image data for a reduced image in which the data amount is reduced in accordance with the visual characteristics of the image. are doing. From the visual characteristics of humans, when an image is reduced, as the reduction ratio decreases and approaches zero, it becomes impossible to identify fine parts of the image. In other words, as the reduction ratio of the image becomes smaller, the high frequency component becomes unnecessary in the AC coefficient of the macroblock. Utilizing such visual characteristics, the coefficient set extracting unit 122 extracts the number of low frequency side coefficients to be extracted as the reduction ratio becomes smaller. Furthermore, in an image, an image boundary portion such as an outline contains many high-frequency components as spatial frequency components. Conversely, in regions where the luminance and color change of the image are gradual, the spatial frequency components concentrate on the low-frequency side. I do. In other words, in an area where the luminance and color change of the image are gradual, even if only the DC component is included in the macroblock, the deterioration of the image quality is small.

  FIG. 7 is an explanatory view exemplifying the encoded image data input to the processing means 12, and FIG. 8 is a processing result of the encoded image data of FIG. FIG. 3 is an explanatory diagram showing an example of encoded image data, that is, an example of reduced encoded image data. FIG. 7 shows a case where a boundary region 400 such as an outline exists in a corresponding macroblock position in the figure in the original image. The macroblocks other than the boundary area 400 are areas where the luminance and color change are gradual. Further, in each macroblock, each AC coefficient (indicated by [AC]) corresponding to a high-frequency component indicated by a hatched area including a DC coefficient is set. As shown in FIG. 8, in an area where the luminance and color change of the image are gradual, zero (indicated by [0]) is set for each coefficient other than the DC coefficient. For an image boundary portion such as an outline, a reduction coefficient set corresponding to a reduction ratio including a DC coefficient is set, and zero is set for other than the reduction coefficient set.

  As described above, when generating the reduced image, the image data processing apparatus 1 uses the above-described visual characteristics of the image, and uses only the DC coefficient of the macroblock in an area where the luminance and color change of the image are gradual. Further, in the boundary region such as the outline of the image, the number of low-frequency coefficients corresponding to the reduction ratio of the image is used. Further, the image data processing apparatus 1 sets all unused coefficients on the high frequency side to zero as ineffective coefficients (for example, inserts EOB as shown in FIG. 6 and deletes the subsequent bit string). ), And output as a processed macroblock, and such a processed macroblock is decoded by the decoding means 13. Therefore, as shown in FIG. 8, since the unnecessary coefficient of each processed macroblock is set to zero, the decoding process is reduced, and the speed of displaying the reduced image can be increased. Further, the image data processing device 1 adaptively changes the number of effective coefficients according to the reduction ratio of the image and whether or not the image is in the boundary region of the image. A small reduced image can be displayed. Further, since the image data processing device 1 generates image data for a reduced image by processing encoded image data of an original image, the image data for the reduced image is stored and held in advance. No means are required.

  The image data processing apparatus 1 may be configured with only hardware that does not require a program, but may be configured to include a microcomputer and a memory that stores, as a program, each procedure of processing realized by the image data processing apparatus 1. You may. In the former case, FIG. 1 and FIG. 2 mean a hardware configuration diagram of the image data processing device 1, and in the latter case, at least a part of the blocks depicted in FIG. 1 and FIG. 1 means a configuration diagram based on the function, that is, a functional diagram. When the image data processing device 1 includes a microcomputer, the microcomputer reads out the program from the memory and executes the program to execute each procedure of the image data processing method defined by the program. The program can be supplied through a recording medium such as a ROM (Read Only Memory) or a CD-ROM, or can be supplied through a transmission medium such as a network.

  In FIG. 2, regardless of the comparison result by the DC coefficient comparing unit 125, the coefficient set extracting unit 122 extracts the reduced coefficient set SMBn, the DC coefficient extracting unit 123 extracts the DC coefficient DCn, The example in which the selection unit 126 selects any coefficient based on the comparison result by the node 125 has been illustrated. On the other hand, only when the DC coefficient comparing unit 125 determines that the difference value is larger than the determination value, the coded image data processing unit 12 extracts the reduced coefficient set SMBn so that the coefficient set extracting unit 122 extracts the reduced coefficient set SMBn. May be configured.

  FIG. 9 and FIG. 10 are flowcharts schematically showing the flow of the processing procedure realized by the image data processing device 1. Since the details of the contents of each process have already been described, the description will not be repeated. As shown in FIG. 9, when the process is started, first, the user inputs a request for designating encoded image data to be reduced-coded and information such as a reduction ratio to the input processing means 17 (S1). The input processing unit 17 receives various instructions from the user. Next, the coded image data processing means 12 generates reduced coded image data based on the reduction ratio specified by the user from the coded image data specified by the user (S2). Next, the decoding unit 13 decodes the generated reduced encoded image data (S3). Subsequently, the image size reducing unit 14 reduces the image size of the decoded reduced encoded image data (S4). Thereafter, the image display means 15 displays the reduced coded image data on the screen of the display means 19 (S5).

  The processing procedure of step S2 is represented by a flowchart of FIG. When the process of step S2 starts, for example, the MB extracting unit 121 resets the control variable n to 0 (S11). Next, the MB extracting unit 121 extracts the encoded image data of the macro block MBn from the encoded image data stored in the encoded image data buffer 120 (S12). Next, the DC coefficient extraction unit 123 extracts the DC coefficient of the macro block MBn (S13). Before or after that, or at the same time, the adjacent DC coefficient extraction unit 124 extracts the DC coefficient of the adjacent block (S14). The adjacent DC coefficient extracting unit 124 can identify the processing target macroblock MBn by acquiring the control variable n from, for example, the MB extracting unit 121 (see FIG. 2), and can specify the adjacent block corresponding thereto. it can.

  Next, the DC coefficient comparing unit 125 calculates a difference value of the DC coefficient (S15), and compares the absolute value of the calculated difference value with a determination value (S16). If the absolute value of the difference value is equal to or greater than the determination value (No in S16), the coefficient set setting unit 132 holds the spatial frequency coefficient on the low frequency side selected from the spatial frequency coefficients of the macroblock MBn according to the reduction ratio, and Is set to zero (S17). On the other hand, if the absolute value of the difference value is smaller than the determination value (Yes in S16), the DC coefficient setting unit 133 holds only the DC coefficient of the macro block MBn and sets other spatial frequency coefficients to zero (S18). . That is, the coefficient set setting unit 132 and the DC coefficient setting unit 133 generate the processed macro block KMBn. Thereafter, the coefficient set setting unit 132 and the DC coefficient setting unit 133 output the processed macro block KMBn (S19).

  Next, the MB extracting unit 121 determines whether the macro block MBn is the last macro block (S20). If the macroblock MBn is the last macroblock (Yes in S20), the coded image data processing unit 12 ends the process. If the macroblock MBn is not the last macroblock (No in S20), the MB extraction unit 121 n is incremented (S21). After step S21, the coded image data processing unit 12 returns the processing to step S12.

  If the coded image data to be processed by the coded image data processing means 12 is color coded image data, the coded image data processing means 12 For MBn, for example, the same processing as in steps S12 to S19 in FIG. 10 may be performed for each of three types of color components. In FIG. 11, the same processes as those in FIG. 10 are denoted by the same reference numerals. The processing procedure of FIG. 11 repeatedly executes the processing of steps S32 to S39 for control variables k = 0, 1, and 2 corresponding to three types of color components for each macroblock MBn. That is, in the loop of steps S32 to S41 that make a round for each control variable k, the processing of steps S32 to S39 is executed. The processes in steps S32 to S39 are similar to the processes in steps S12 to S19 in FIG. 10 for the k-th color component MBnk of the macro block MBn. For example, the process of step S32 is executed by the MB extracting unit 121, as in step S12. Further, the control variable k may be set by the MB extracting unit 121 similarly to the control variable n.

  Here, the determination value in step S36 may be given individually for each color component. Further, the coefficient reduction rate in step S37 may be individually given for each color component. For example, when the encoded image data to be processed is expressed in a Y (luminance) U (hue) V (color difference) color system, the Y component is expressed by the U component and the V component. It is also preferable to lower the determination value and increase the reduction rate (weak the degree of reduction).

  Further, as described as step S50 in FIG. 11, the spatial frequency coefficient may not be reduced (No in S50) or may be reduced (Yes in S50) depending on the control variable k corresponding to the color component. . Only when the spatial frequency coefficient is reduced, the processes of S33 to S38 are performed. When the spatial frequency coefficient is not reduced, the k-th color component MBnk of the macro block MBn is output as it is as the k-th color component of the processed macro block KMBn. (S39). In order to output the k-th color component MBnk of the macro block MBn as it is as the k-th color component of the processed macro block KMBn, the determination value is set to a negative value in step S36 for the control variable k, and step S37 is performed. , The coefficient reduction rate may be set to 1 and the processing of steps S33 to S37 may be executed. On the other hand, in the block diagram of FIG. 2, for the control variable k, the MB extraction unit 121 directly inputs the processed macro block 127 to the processed MB generation unit 127 so as to output the macro block MBn as the processed macro block KMBn. May be.

(Embodiment 2)
FIG. 12 is a block diagram showing an overall configuration of an image data transmission system according to Embodiment 2 of the present invention. The image data transmission system 600 converts image data compressed and encoded by the JPEG method or the like (hereinafter, referred to as encoded image data) from a distribution device 501 as an image data distribution device to a terminal device 502 owned by a user. This is a system for transmission via a network 503 such as the Internet. Further, the user can specify the reduced size of the image to be transmitted to the distribution device 501, and the distribution device 501 distributes encoded image data corresponding to the reduced size specified by the user.

  12, components denoted by the same reference numerals as in FIG. 1 have the same functions, and description thereof will be omitted.

  As shown in FIG. 12, an image data transmission system 600 according to the second embodiment has a configuration in which a distribution device 501 and a plurality of terminal devices installed for each user are connected via a network 503 such as the Internet. It is. In FIG. 12, only one terminal device 502 of the plurality of terminal devices is shown as a representative.

  In FIG. 12, a distribution device 501 owned by a service provider that provides an image or the like stores encoded image data corresponding to each image in the encoded image file storage unit 10 in order to provide an image according to a user's request. Has accumulated. The distribution device 501 further distributes the encoded image data according to the user's request to the terminal device 502 via a network 503 such as the Internet.

  The terminal device 502 is a device for transmitting information such as a user request to the distribution device 501 by a user operation, and receiving various information including image data. For example, when an image distribution request is issued from the terminal device 502 to the distribution device 501, the image data is distributed from the distribution device 501, and the terminal device 502 receives the distributed image data, and sequentially restores and reproduces the image data.

  As described above, the image data transmission system 600 is a system in which image data is distributed from the distribution device 501 to the terminal device 502 via the network 503, and a user of the terminal device 502 views a desired image.

  Further, in the present image data transmission system 600, when the user requests a reduced image of a desired image, for example, from the terminal device 502, the distribution device 501 distributes the reduced image data of the corresponding image as reduced encoded image data. It is characterized by the following. For example, when the user wants to select a desired image from a plurality of images, the user first requests the distribution device 501 to transmit a reduced image such as a thumbnail image. Further, the user displays, for example, an index image including a plurality of thumbnail images, and selects a desired image from the index images. As described above, the user can select a desired image in advance by the thumbnail image, and the convenience for the user is improved.

  The procedure of the process performed by the image data transmission system 600 will be described below. In FIG. 12, various instructions from the user are input to the input unit 18. The input processing means 17 receives these instructions. To this end, the input processing unit 17 requests a request for specifying encoded image data corresponding to one image from a plurality of encoded image data stored in the encoded image file storage unit 10 (for example, the image file name is The request includes an image request input unit 62 for receiving a request (designation is made) and a reduction ratio input unit 64 for receiving an input of a reduction ratio (for example, performed by inputting an image size).

  When the user inputs a request for distributing a reduced image from the input unit 18, information such as an image file name and an image size designated by the user is notified to the communication unit 54 via the input processing unit 17. The communication unit 54 of the terminal device 502 transmits request information indicating a request from the user to the distribution device 501 via the network 503, and receives image data and the like from the distribution device 501, and The communication with the device 501 is performed. For this purpose, the communication unit 54 includes an image request transmission unit 63 that transmits a request for encoded image data, a reduction ratio transmission unit 65 that transmits a reduction ratio, and reception of reduced encoded image data that receives reduced encoded image data. Means 61 are included.

  The communication unit 53 of the distribution device 501 communicates with the terminal device 502 via the network 503. That is, the information of the image file name and the size of the image designated by the user is transmitted via the communication means 54, the network 503, and the communication means 53 to the user instruction determination means (hereinafter, referred to as determination means) 16 of the distribution apparatus 501. Will be notified. To this end, the communication unit 53 includes an image request receiving unit 71 that receives a request for encoded image data, a reduction ratio receiving unit 72 that receives a reduction ratio, and a transmission unit 73 that transmits reduced encoded image data. I have.

  The determination unit 16 outputs image identification information for searching for an image file based on the image file name or the like designated by the user, and image size information for displaying at the image size designated by the user. The coded image file storage means (hereinafter referred to as storage means) 10 functioning as coded image data storage means stores coded image data compressed and coded according to the JPEG method or the like as a plurality of image files. An encoded image data reading unit (hereinafter, referred to as a reading unit) 11 reads an image file corresponding to image identification information from a plurality of image files stored in the storage unit 10. The coded image data of the read image file is processed by coded image data processing means (hereinafter referred to as processing means) 12 as a thumbnail image having a reduction rate corresponding to the image size information, ie, reduced coded image data for the reduced image. It is processed to convert to The processed image data for the reduced image is supplied to the communication unit 53 functioning as the transmission unit 73, and transmitted to the terminal device 502 via the network 503. That is, the processed image data for the reduced image is distributed as reduced encoded image data processed as shown in FIG.

  The terminal device 502 receives the reduced encoded image data processed as described above, and the reduced encoded image data is transmitted through the communication unit 54 functioning as the reduced encoded image data receiving unit 61. The data is acquired by a data acquisition unit (hereinafter, referred to as an acquisition unit) 51 and is temporarily stored. The decoding unit 13 decodes the encoded image data with respect to the reduced encoded image data, and restores the decoded image data having the same size as the original image. The decoded image data storage unit 52 stores the decoded image data having the same size as the restored original image. An image size reducing unit (hereinafter referred to as a reducing unit) 14 thins out pixels of the decoded image data restored by the decoding unit 13 in accordance with a reduction ratio of a reduced size specified by a user, thereby reducing a reduced image. Generate data. The screen display means 15 outputs the reduced image data to the monitor 19 so as to be displayed on the screen of the monitor 19 functioning as the display means.

  FIG. 13 and FIG. 14 are diagrams illustrating examples of reduced images displayed on the screen of the monitor 19. FIG. 13 shows an example in which six reduced images are index-displayed as thumbnail images on the screen of the monitor 19, and FIG. 14 shows an example in which four reduced images are index-displayed as thumbnail images on the screen of the monitor 19. As described above, in the image data transmission system 600, the user can specify the size of the reduced image as the distribution request. As a result, as shown in FIGS. 13 and 14, reduced images of different sizes can be displayed on the screen of the monitor, and the user selects a desired image using a thumbnail image of a desired size. It becomes possible.

  As described above, in response to the request for distributing the reduced image transmitted from the terminal device 502, the distribution device 501 distributes the reduced encoded image data corresponding to the request to the terminal device 502, and the terminal device 502 The encoded image data is decoded and reproduced as a thumbnail image. At this time, the image data transmission system 600 is characterized in that the coded image data is processed into a reduced image by the processing means 12 configured as shown in FIG.

  In other words, as described in the first embodiment, the processing unit 12 sets unnecessary coefficients to zero for each processed macroblock in accordance with the visual characteristics of the image as shown in FIG. ing. For this reason, the amount of coding can be reduced, and the transmission load on the network 503 can be reduced. As a result, the communication speed can be increased. Further, the image data transmission system 600 adaptively changes the number of effective coefficients according to the image reduction ratio and whether or not the image is in the boundary region of the image. The terminal device 502 can display a thumbnail image with little image quality degradation. Therefore, the user can quickly search for a desired image from a large number of image files existing on the Internet or the like using the thumbnail images. If the thumbnail image is too small to be easily identified, the user can request a larger image size and select an image he or she desires while checking in more detail. Also, since the image data transmission system 600 adaptively changes the encoding amount according to the image size, even if the size of the thumbnail image is increased, the image quality is prevented from deteriorating and the encoding amount is reduced. The transmission speed is prevented from dropping.

  Furthermore, when the distribution device 501 holds the original image as encoded image data and distributes the reduced image, it is only necessary to process and transmit the encoded image data. It is possible to reduce the processing load. For example, in the case of a method of encoding and transmitting according to the reduction rate every time the original image is transmitted, it is necessary to perform encoding for each transmission, and when the number of terminals increases, this encoding process is performed. The load becomes heavy. Also, in the case of a method in which encoded image data is decoded, the image is reduced, re-encoded and transmitted, the processing load of decoding and re-encoding also becomes heavy. On the other hand, the distribution device 501 of the image data distribution system 600 first stores image data to be normally distributed as encoded image data, so that the image data is compared with non-coded image data. The storage capacity to be held may be small. Further, when distributing a reduced image, the distribution device 501 only needs to perform simple processing by the processing unit 12 on the encoded image data, so that the processing load for generating the reduced image is reduced. Thus, even if the number of terminals increases, it is possible to sufficiently cope with the case.

(Other application examples)
(1) The present invention is also applicable to a technology for displaying an image on a device having a narrow screen such as a mobile phone and a mobile terminal. That is, even when an image is displayed not on a thumbnail image but on a screen smaller than a normal monitor screen, the image data processing device, the image data processing method, the image data distribution device, or the image data transmission system of the present invention is used. The generated reduced encoded image data can be decoded and reduced, and displayed on a screen.

  (2) The present invention is also applicable to a technique for displaying a still image on a web browser. Also, for the convenience of the user when selecting a moving image, the reduced coded image data generated by the technique of the present invention is also used when displaying an I frame (intra-coded frame) of the moving image as a still image. Can be used.

  (3) The present invention is applicable to image editing. For example, when performing a process of cutting out a part of an image, a process of changing the size of an image, a process of mosaicing an image, and the like, temporarily displaying an image during editing before displaying the final image after processing. To do so, it is possible to use reduced coded image data generated using the technique of the present invention. Similarly, when editing multimedia elements (audio, video, text, etc.) including a video using an authoring tool, the technology of the present invention can be used to display an image being edited.

  INDUSTRIAL APPLICABILITY The present invention is industrially useful because even when a relatively large thumbnail image is handled, the thumbnail image can be displayed at high speed without deteriorating the image quality.

FIG. 1 is a block diagram illustrating an overall configuration of an image data processing device according to a first embodiment of the present invention. FIG. 2 is a block diagram illustrating a detailed configuration of an encoded image data processing unit of the image data processing device according to the first embodiment of the present invention. FIG. 3 is a diagram illustrating macroblocks processed by encoded image data processing means and the state of each coefficient. 9 is a graph illustrating an example of a relationship between a reduction rate and a coefficient reduction rate. It is a graph which shows another example of the relationship between a reduction rate and a coefficient reduction rate. FIG. 4 is an explanatory diagram illustrating a relationship between a macro block MBn and a processed macro block KMBn. FIG. 3 is a diagram illustrating an example of input encoded image data. FIG. 5 is a diagram illustrating an example of encoded image data processed by an encoded image data processing unit. 3 is a flowchart illustrating an operation procedure of the image data processing device in FIG. 1. 10 is a flowchart illustrating an example of a processing procedure of step S2 in FIG. 10 is a flowchart illustrating another example of the processing procedure of step S2 in FIG. 9. FIG. 7 is a block diagram illustrating an overall configuration of an image data transmission system according to a second embodiment of the present invention. FIG. 4 is a diagram illustrating an example of a reduced image displayed on a monitor. FIG. 11 is a diagram illustrating another example of the reduced image displayed on the monitor.

Explanation of reference numerals

DESCRIPTION OF SYMBOLS 1 Image data processing apparatus 10 Encoded image file storage means (encoded image data storage means; storage means)
11 Encoded image data reading means (encoded image data reading means; reading means)
12 Encoded image data processing means (processing means)
13 decoding means 14 image size reducing means (reducing means)
15 screen display means 16 user instruction determination means (determination means)
Reference Signs List 17 input processing means 18 input means 19 monitor 51 encoded image data acquisition means 52 decoded image data storage means 53, 54 communication means 61 reduced encoded image data reception means 62 image request input means 63 image request transmission means 64 reduction rate input Means 65 Reduction rate transmission means 71 Image request reception means 72 Reduction rate reception means 73 Transmission means 120 Encoded image data buffer (buffer)
121 Macro block extraction unit (MB extraction unit)
122 coefficient set extraction unit 123 DC coefficient extraction unit 124 adjacent DC coefficient extraction unit 125 DC coefficient comparison unit 126 selection unit 127 processing macro block generation unit (processing MB generation unit)
132 Coefficient set setting means 133 DC coefficient setting means 501 Image data distribution device (distribution device) 502 Terminal device 503 Network 600 Image data transmission system

Claims (28)

The original image is divided into a plurality of blocks composed of a plurality of pixels, and in each block, the original image is given from encoded image data obtained by encoding the spatial frequency components of the block as a plurality of spatial frequency coefficients. An image data processing device that generates reduced encoded image data that is encoded data of a reduced image reduced at a reduced reduction rate,
Block extraction means for sequentially extracting each of the plurality of blocks as a processing target block,
DC coefficient extracting means for extracting a DC component coefficient from the plurality of spatial frequency coefficients of the processing target block,
An adjacent DC coefficient for extracting, as at least one adjacent DC component coefficient, a DC component coefficient of at least one adjacent block adjacent to the processing target block in any of the horizontal direction, the vertical direction, and the oblique direction of the plurality of blocks. Extraction means;
The DC component coefficient of the processing target block is compared with the at least one adjacent DC component coefficient, and the DC component coefficient and the at least one adjacent DC component coefficient representing a degree of difference between the two are defined as variables. DC coefficient comparing means for determining a value of a function to be performed, wherein, when the value of the function is larger than a predetermined value, the DC component coefficients are included in accordance with the reduction ratio for the plurality of spatial frequency coefficients of the processing target block. From a low frequency spatial frequency coefficient, a coefficient set setting unit that holds a smaller number of spatial frequency coefficients than the plurality, and sets other spatial frequency coefficients to zero.
When the value of the function is less than the predetermined value, for the plurality of spatial frequency coefficients of the processing target block, only the DC component coefficient is held, and the DC coefficient for setting the other spatial frequency coefficients to zero is set. An image data processing apparatus comprising: a setting unit.   The image according to claim 1, wherein the at least one adjacent block is one adjacent block, and the value of the function is an absolute value of a difference value between the DC component coefficient and the adjacent DC component coefficient. Data processing device. The at least one neighboring block is a plurality of neighboring blocks,
The function value is an absolute value of a difference value between an average value of a plurality of adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficient of the processing target block. 2. The image data processing device according to 1. The at least one neighboring block is a plurality of neighboring blocks,
The value of the function is the absolute value of the difference between the weighted average value obtained by the weighted average calculation based on the adjacent direction and the DC component coefficient for a plurality of adjacent DC component coefficients included in the plurality of adjacent blocks. The image data processing device according to claim 1, wherein The at least one neighboring block is a plurality of neighboring blocks,
The value of the function is the largest value among the absolute values of the difference values between the adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficients of the processing target block. The image data processing device according to claim 1. The at least one neighboring block is a plurality of neighboring blocks,
2. The function value according to claim 1, wherein the value of the function is a sum of absolute values of respective difference values between each adjacent DC component coefficient included in the plurality of adjacent blocks and the DC component coefficient of the processing target block. Image data processing device. The ratio of the number of the spatial frequency coefficients less than the plurality held by the coefficient set setting means to the number of the plurality of spatial frequency coefficients is a coefficient reduction rate,
The coefficient set setting unit divides the reduction rate into a plurality of steps, sets the coefficient reduction rate according to each of the reduction rate steps, and, when the value of the function is larger than the predetermined value, The method according to any one of claims 1 to 6, wherein according to the coefficient reduction rate, the number of the spatial frequency coefficients smaller than the plurality is held, and zero is set to the other spatial frequency coefficients. Image data processing device. Input processing means for receiving an input of the reduction ratio,
Decoding means for decoding the generated reduced encoded image data,
Image size reducing means for reducing the image size of the decoded reduced encoded image data,
The image data processing apparatus according to claim 1, further comprising: a screen display unit configured to display the reduced coded image data on a screen. The encoded image data is color encoded image data,
9. The image data processing apparatus according to claim 1, wherein the predetermined value is provided for each color component. The encoded image data is color encoded image data,
The image data processing apparatus according to claim 1, wherein the coefficient set setting unit determines the number of the spatial frequency coefficients less than the plurality for each color component. The encoded image data is color encoded image data,
The image data processing device according to claim 1, wherein the block extracting unit extracts the processing target block only for a predetermined color component. The encoded image data is variable-length encoded data,
The coefficient set setting means and the DC coefficient setting means attach a code indicating the end of a block after all of the spatial frequency coefficients to be retained, and delete the other spatial frequency coefficients, so that the other spatial frequency coefficients are deleted. The image data processing device according to claim 1, wherein the image data represents that the frequency coefficient is set to zero. The original image is divided into a plurality of blocks composed of a plurality of pixels, and in each block, the original image is given from encoded image data obtained by encoding the spatial frequency components of the block as a plurality of spatial frequency coefficients. An image data processing method for generating reduced coded image data that is coded data of a reduced image reduced at a given reduction rate,
A first step of extracting a DC component coefficient from the plurality of spatial frequency coefficients;
A second step of extracting, as at least one adjacent DC component coefficient, a DC component coefficient of at least one adjacent block adjacent to the extracted DC component coefficient block in any of the horizontal direction, the vertical direction, and the oblique direction; ,
The DC component coefficient is compared with the at least one adjacent DC component coefficient, and a value of a function having the DC component coefficient and the at least one adjacent DC component coefficient representing a degree of a difference between the two as variables is defined as: A third step to seek;
When the value of the function is larger than a predetermined value, for the plurality of spatial frequency coefficients, according to the reduction ratio, from the low frequency spatial frequency coefficients including the DC component coefficients, A fourth step of retaining the coefficients and setting other spatial frequency coefficients to zero;
A fifth step of, when the value of the function is less than the predetermined value, holding only the DC component coefficient for the plurality of spatial frequency coefficients and setting zero to other spatial frequency coefficients. And
An image data processing method comprising: performing the above steps on each of the blocks to generate the reduced coded image data which is coded data of a reduced image obtained by reducing the original image at the reduction ratio. .   The image according to claim 13, wherein the at least one adjacent block is one adjacent block, and the value of the function is an absolute value of a difference value between the DC component coefficient and the adjacent DC component coefficient. Data processing method. The at least one neighboring block is a plurality of neighboring blocks,
The image data according to claim 13, wherein the value of the function is an absolute value of a difference value between an average value of a plurality of adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficient. Processing method. The at least one neighboring block is a plurality of neighboring blocks,
The value of the function is the absolute value of the difference between the weighted average value obtained by the weighted average calculation based on the adjacent direction and the DC component coefficient for a plurality of adjacent DC component coefficients included in the plurality of adjacent blocks. 14. The image data processing method according to claim 13, wherein The at least one neighboring block is a plurality of neighboring blocks,
14. The value of the function according to claim 13, wherein the value of the function is the largest value among the absolute values of the difference values between the adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficients. Image data processing method. The at least one neighboring block is a plurality of neighboring blocks,
14. The image data processing method according to claim 13, wherein the value of the function is a sum of absolute values of respective difference values between adjacent DC component coefficients included in the plurality of adjacent blocks and the DC component coefficient. . The ratio of the number of the spatial frequency coefficients less than the plurality retained by the fourth step to the number of the plurality of spatial frequency coefficients is a coefficient reduction rate,
The fourth step divides the reduction rate into a plurality of steps, sets the coefficient reduction rate according to each of the reduction rate steps, and, when the value of the function is larger than the predetermined value, The method according to any one of claims 13 to 18, wherein the number of the spatial frequency coefficients less than the plurality is held according to the coefficient reduction rate, and zero is set for the other spatial frequency coefficients. Image data processing method. A sixth step of receiving an input of the reduction ratio;
A seventh step of decoding the generated reduced encoded image data;
An eighth step of reducing an image size of the decoded reduced encoded image data,
A ninth step of displaying the reduced reduced coded image data on a screen, further comprising the steps of: The encoded image data is color encoded image data,
21. The image data processing method according to claim 13, wherein the predetermined value is provided for each color component. The encoded image data is color encoded image data,
22. The image data processing method according to claim 13, wherein in the fourth step, the number of the spatial frequency coefficients smaller than the plurality is determined for each color component. The encoded image data is color encoded image data,
23. The image data processing method according to claim 13, wherein the image data processing method performs the first to fifth steps for each of the blocks only for a predetermined color component. The encoded image data is variable-length encoded data,
The fourth step and the fifth step are performed by attaching a sign indicating the end of a block after all of the spatial frequency coefficients to be retained and deleting the other spatial frequency coefficients to thereby obtain the other spatial frequency coefficients. The image data processing method according to any one of claims 13 to 23, wherein a representation that zero is set in the frequency coefficient is expressed. Divide the original image into multiple blocks consisting of multiple pixels,
In each block, a reduced code that is encoded data of a reduced image obtained by reducing the original image at a given reduction rate from encoded image data obtained by encoding the spatial frequency components of the block as a plurality of spatial frequency coefficients. An image data distribution apparatus for generating encoded image data and transmitting the reduced encoded image data via a network,
Coded image data storage means for storing a plurality of the coded image data,
From the plurality of encoded image data stored in the encoded image data storage unit, encoded image data reading unit that reads encoded image data corresponding to one image,
An image data processing device according to any one of claims 1 to 12,
The image data processing device generates the reduced encoded image data from the encoded image data read by the encoded image data reading unit,
The image data distribution device further comprises a transmission unit that transmits the reduced coded image data generated by the image data processing device to the network. An image data distribution device according to claim 23,
An image data transmission system, comprising: a terminal device capable of connecting the image data distribution device and itself via the network.
The terminal device,
Reduced encoded image data receiving means for receiving the reduced encoded image data transmitted by the transmission means,
Decoding means for decoding the received reduced encoded image data,
Image size reducing means for reducing the image size of the decoded reduced encoded image data,
An image data transmission system, comprising: screen display means for displaying the reduced coded image data on a screen. The terminal device,
From the plurality of encoded image data stored in the encoded image data storage unit, an image request input unit that receives a request to specify encoded image data corresponding to one image,
Image request transmitting means for transmitting the received request to the image data distribution device through the network,
The image data distribution device,
Further comprising an image request receiving means for receiving the transmitted request,
27. The encoded image data reading unit according to claim 26, wherein the encoded image data reading unit reads encoded image data corresponding to the request from the plurality of encoded image data stored in the encoded image data storage unit. Image data transmission system. The terminal device,
Reduction rate input means for receiving an input of the reduction rate;
A reduction rate transmitting unit that transmits the received reduction rate to the image data distribution device through the network,
The image data distribution device,
Further comprising a reduction rate receiving means for receiving the transmitted reduction rate,
28. The image data transmission system according to claim 26, wherein the image data processing device generates the reduced encoded image data using the received reduction ratio as the given reduction ratio.

Images