JP2008283327A - Image data restoration method, and image processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image data restoration method for reducing the storage capacity of a storage device necessary for an operation to restore an original image, and for reducing an labor to be imposed on an arithmetic processing unit for restoration processing. <P>SOLUTION: This image data restoration method for restoring image data compressed by a JPEG system into an original image includes a process A for acquiring quantized data by decoding Huffman encoded data corresponding to blocks from the leading of the first block line of the original image to the last of the block line which is the m-th block line or more and the (n-1)th block line or less when the original image is formed of n (n is a natural number) lines of block lines and blocks ranging from the first block line of the original image to the m(m is a natural number and m<n) block line of the original image are defined as a display range at a display device; and a process B for restoring the original image based on the quantized data acquired by the process A. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image data restoration method and an image processing apparatus, and more particularly, to an image data restoration method for restoring image data compressed by the JPEG method to an original image and an image data compressed by the JPEG method. The present invention relates to an image processing apparatus that restores an image and displays it on a display device.

Conventionally, as described in Patent Document 1, a copy that can be displayed on a display device in order to confirm beforehand whether an image to be transmitted by facsimile or an image to be copied is appropriate before facsimile transmission or copying A facsimile multifunction machine has been invented. The copy facsimile multifunction peripheral described in Patent Document 1 displays an image to be transmitted by facsimile or an image to be copied on an external display device, but displays an image to be copied or an image to be transmitted by Internet facsimile communication on a display device included in the own device. Internet facsimile machines have also been invented. In such an Internet facsimile apparatus, an image to be displayed on the display device can be enlarged or reduced according to a user instruction.
JP 2006-246284 A

  In the Internet facsimile apparatus described above, image data of an image displayed on the display device is often stored in a storage device in JPEG format, and the image is displayed on the display device, or the displayed image is enlarged or reduced. In each case, it is necessary to restore the original image by decoding the image data in JPEG format, and a large amount of storage capacity of the storage device is necessary for the restoration process. For this reason, there is a problem that the burden on the arithmetic processing unit is large.

  Therefore, the present invention can reduce the storage capacity of the storage device necessary for the restoration process of the original image, and can reduce the burden on the arithmetic processing unit for the restoration process. An object is to provide a data restoration method and an image processing apparatus.

  In order to achieve the above object, an image data restoration method according to claim 1 is an image data restoration method for restoring image data compressed by the JPEG method to an original image, wherein the original image is n (n is a natural number). ) Block lines, and when the display range from the first block line to the m-th block (m is a natural number and m <n) of the original image is the display range of the original image, Step A for obtaining quantized data by decoding Huffman encoded data corresponding to the block from the beginning of the first block line to the last block of the block line that is not less than the m-th block line and not more than the (n-1) -th block line. And a step B of restoring the original image based on the quantized data obtained by decoding in the step A.

  Here, the “display range in the display device” is a range in which the original image can be displayed on the display device without scrolling. A “block” is a unit of processing in JPEG encoding, and is a collection of 8 × 8 pixels that divide the entire original image. Further, the “block line” is a line formed by blocks that are arranged in a row in the horizontal direction (x-axis direction) from the left end to the right end of the drawing of the image as shown in FIG. The first block line is the uppermost block line in the drawing where the original image starts to be drawn on the display device. Further, “more than the mth block line” means the direction of increasing y including the mth block line, and “below the (n−1) th block line” includes the n−1th block line. Therefore, this means a direction of decreasing y. These are the same in the following claims.

  The image data restoration method according to claim 2 is the image data restoration method according to claim 1, wherein the step B is performed from the left end of the original image to the right end or more of the display range of the original image on the display device. A step B-1 for obtaining a DC component and a frequency component by dequantizing the quantized data of blocks up to the right end of the original image, and a DC component and a frequency obtained by the inverse quantization in the step B-1. And a step B-2 of obtaining an original image by performing an inverse discrete cosine transform based on the components.

  Here, “more than the right end” in “more than the right end of the display range” means the drawing direction of the original image, that is, the increasing direction of x in FIG. The term “less than” does not include the right end, and means the reverse direction of the original image, that is, the decreasing direction of x. This also applies to the following claims.

  The image data restoration method according to claim 3 is the image data restoration method according to claim 2, wherein in the step B-1, quantized data corresponding to a direct current component is displayed on the display device. It is characterized in that the quantized data corresponding to the frequency component of the frequency up to the height corresponding to the reduction ratio of the original image is inversely quantized.

  In order to achieve the above object, an image processing apparatus according to claim 4 is an image processing apparatus that restores image data compressed by the JPEG method to an original image and displays the original image on a display device. The first block of the original image is formed by (n is a natural number) block lines, and the range from the first block line of the original image to the m-th (m is a natural number) block line is the display range in the display device. Decoding means for obtaining quantized data by decoding Huffman encoded data corresponding to from the head of the line to the last block of the block line that is not less than the m-th block line and not more than the (n-1) -th block line; Of the quantized data obtained by decoding by the decoding means, from the left end of the original image to the right end of the original image that is not less than the right end of the display range of the original image on the display device Inverse quantization means that obtains a DC component and a frequency component by inverse quantization of quantized data of blocks up to less than the block, and an inverse based on the DC component and frequency component obtained by inverse quantization by the inverse quantization means And an inverse discrete cosine transform unit that obtains an original image by performing a discrete cosine transform, wherein the inverse quantizing unit is configured to reduce quantization data corresponding to a DC component and a reduction ratio of the original image displayed on the display device. It is characterized in that the quantized data corresponding to the frequency component of the frequency up to the corresponding height is inversely quantized.

  According to the image data restoration method of claim 1, the Huffman necessary for restoring a part of the original image including the display range in the display device is not decoded, but all the Huffman encoded data constituting the image data is decoded. Since only the encoded data is decoded, the amount of data to be handled can be reduced, the storage capacity of the storage device necessary for the restoration process of the original image can be reduced, and the arithmetic processing unit for the restoration process The effect that the burden concerning it can be reduced is acquired.

  According to the image data restoration method according to claim 2, instead of performing inverse quantization on all of the quantized data obtained by decoding the Huffman encoded data, some elements including a display range in the display device are included. Since only the quantized data necessary for image restoration is inversely quantized, the amount of data handled is reduced, the storage capacity of the storage device required for the restoration of the original image can be reduced, and restoration is also possible. As a result, it is possible to reduce the burden on the arithmetic processing unit for the above processing.

  According to the image data restoration method according to claim 3, instead of performing inverse quantization on all quantized data included in a block obtained by decoding Huffman encoded data, quantization corresponding to a DC component is performed. Only the data and the quantized data corresponding to the frequency component of the frequency up to the height corresponding to the reduction ratio of the original image displayed on the display device are inversely quantized, so the amount of data handled is reduced and the original image is restored. Thus, it is possible to reduce the storage capacity of the storage device necessary for this processing and to reduce the burden on the arithmetic processing device for the restoration processing.

  According to the image processing device of claim 4, the Huffman code necessary for restoring a part of the original image including the display range in the display device is not decoded, but all the Huffman encoded data constituting the image data is decoded. Necessary to restore only part of the original image including the display range on the display device, rather than decoding only the quantized data and not performing inverse quantization on all the quantized data obtained by decoding the Huffman encoded data Rather than performing inverse quantization on all quantized data included in the block obtained by decoding the Huffman encoded data, the quantized data corresponding to the DC component Only the quantized data corresponding to the frequency component of the frequency up to the height corresponding to the reduction ratio of the original image displayed on the display device is inversely quantized. Accordingly, since the amount of data to be handled is reduced, the storage capacity of the storage device required for the restoration process of the original image can be reduced, and the burden on the arithmetic processing unit for the restoration process can be reduced. The effect that it can do is acquired.

  Hereinafter, an image data restoration method and an image processing apparatus according to the present invention will be described with reference to the drawings. The image data restoration method according to the present invention is a method for restoring image data compressed by the JPEG method to an original image, and is realized in, for example, the Internet facsimile machine 1 as an image processing apparatus. The Internet facsimile machine 1 is a so-called facsimile multifunction machine having a facsimile communication function, an electronic mail communication function, a scan function (including a scan-to-folder function), a copy function, and a print function. The Internet facsimile machine 1 is installed in an office or the like, and is connected to a public switched telephone network (PSTN) 2 and a LAN 3 (Local Area Network). In addition to the Internet facsimile machine 1, a client PC (Personal Computer), a mail server, and the like are connected to the LAN 3. A router is also connected to the LAN 3, and the LAN 3 is connected to the Internet via the router.

  FIG. 1 is a block diagram showing a configuration example of the Internet facsimile machine 1. As shown, the Internet facsimile machine 1 includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, an HDD (Hard Disk Drive) 104, an operation unit 105, a display unit. 106, a document reading unit 107, an image processing unit 108, a codec (CODEC: Coder and Decoder) 109, an image memory 110, a printer 111, a FAX communication unit 112, and a LAN-I / F (Local Area Network Interface) 113. The units 101 to 113 are communicably connected via a bus 114.

  The CPU 101 controls the operation of each part of the Internet facsimile machine 1 according to a control program stored in the ROM 102. The ROM 102 stores the control program for controlling the operation of each unit of the Internet facsimile machine 1 by the CPU 101. The RAM 103 stores various data such as setting information used for the operation of the Internet facsimile machine 1 in a readable and writable state, and also functions as a work area for the CPU 101. The HDD 104 is for storing image data transmitted and received by the facsimile function.

  Although not shown, the operation unit 105 includes a start key for instructing the start of a document reading (scanning) operation, a numeric keypad for inputting a facsimile number and the number of copies, a character input key for inputting characters, Cursor keys for specifying the operation target are provided. The display unit 106 is a touch panel liquid crystal display (LCD) (hereinafter, referred to as “display device”) 106 a that displays various setting states, operation states of the device 1, and the like with characters and graphics. In addition, an LED (Light Emitting Diode) lamp that is turned on or off is provided. The display of the display device 106a changes according to the operation of each key provided in the operation unit 105 or the tap of the panel of the display device 106a provided in the display unit 106, and each key of the operation unit 105 corresponding to the display of the display device 106a is changed. Various inputs can be performed by operating or tapping the panel of the display device 106a. Designation of the enlargement / reduction ratio of the JPEG image data displayed on the display device 106a, scrolling instructions, and the like can also be performed by operating each key of the operation unit 105 or tapping the panel of the display device 106a.

  An original reading unit 107 reads an image of an original and generates image data. Although not shown, for example, a flat bed scanner (FBS: Flat) that reads an original placed on a transparent original placing plate is used. Bed Scanner) and an automatic document feeder (ADF: Automatic Document Feeder) that conveys the document to read the document placed on the document tray.

  The image processing unit 108 performs processing such as color adjustment, color space conversion, and binarization on the image data output from the document reading unit 107. The codec 109 encodes (encodes) the image data processed by the image processing unit 108 and the like, and decodes (decodes) the encoded image data. The image data input to the codec 109 is encoded based on the JPEG, MH, MR, MMR, JBIG method, etc., and stored in the image memory 110. The image data restoration method according to the present invention is executed by the codec 109. Further, the image data encoded by the codec 109 is further formed into an image file such as a TIFF (Tagged Image File Format) -S file by the image processing unit 108 and stored in the HDD 104.

  The printer 111 prints out the image data read from the image memory 110 and decoded by the codec 109.

  The FAX communication unit 112 performs facsimile (FAX) communication of document image data, and includes a modem (MODEM-DEModulator) and an NCU (Network Control Unit) (not shown). The modem is, for example, a recommendation V.16 of ITU-T (International Telecommunication Union Telecommunication Standardization Sector). Modulation and demodulation of transmission / reception data according to the 34 standard or the like. The NCU is a line network control device that controls a telephone line to make or cut a call, and is connected to the PSTN 2. In addition to the Internet facsimile machine 1, a G3 facsimile machine or the like is communicably connected to the PSTN 2, and the Internet facsimile machine 1 can transmit or receive image data by facsimile communication with the G3 facsimile machine or the like. is there.

  The LAN-I / F 113 enables the Internet facsimile machine 1 to be connected to the LAN 3. A client PC, a mail server, and the like are also connected to the LAN 3, and the Internet facsimile apparatus 1 can communicate data with the client PC, the mail server, and the like. The Internet facsimile machine 1 is connected to other Internet facsimile machines provided on the Internet via the LAN 3 and the router, and can perform Internet facsimile communication with other Internet facsimile machines. . Here, the Internet facsimile communication is, for example, communication in which an electronic mail attached with a file (image data) in TIFF-S format is transmitted / received by SMTP (Simple Mail Transfer Protocol).

  In such an Internet facsimile machine 1, color image data generated by reading an image of an original by the original reading unit 107 when executing a copy function or a scan-to-folder function is output by a codec 109. , Encoded based on the JPEG method, and stored in the image memory 110. The JPEG color image data stored in the image memory 110 is used to confirm whether the image of the color image data is appropriate, that is, whether the image is a desired image intended by the user. Thus, the data is restored by the codec 109 by the image data restoration method of the present invention and displayed on the display device 106a. In the Internet facsimile machine 1, not only the JPEG color image data generated by the document reading unit but also an external storage device such as a USB (Universal Serial Bus) memory is taken into the device 1 or on the Internet. Even when JPEG color image data downloaded from a predetermined site to the device 1 is displayed on the display device 106a, the image data is restored by the image data restoration method of the present invention. Note that the image data restored by the image data restoration method of the present invention is not limited to color image data but may be monochrome image data as long as it is JPEG image data. Will be described without distinction between color image data and monochrome image data.

First, in general JPEG encoding performed in the codec 109, image data is divided into blocks of 8 pixels × 8 pixels as shown in FIG. Specifically, first, a two-dimensional discrete cosine transform is performed on each block of 8 pixels × 8 pixels, and a spatial region composed of the values of each pixel of the block is converted into a frequency region. The two-dimensional discrete cosine transform is defined by Equation 1. The value of N (64) pixels included in the block of FIG. 2 is expressed as f (x, y) and (x, y = 0,1,2,3,4,5,6,7). By applying, two-dimensional discrete cosine transform is performed on the block. The codec 109 performs this two-dimensional discrete cosine transform by performing a known matrix calculation.

  The space domain of the block is transformed into the frequency domain by the binary discrete cosine transform according to Equation 1, and one DC component (DC component) value (F (0,0)) and N−1 (63) frequencies. Component (AC component) values (F (0,1), F (0,2), ..., F (1,0), F (1,1), ..., F (7,6), F (7 , 7)). Note that the 63 frequency components become higher frequency components from the upper left to the lower right in FIG. 3, and a large value of the frequency components tends to appear in the low frequency components.

  The DC component and the frequency component obtained by the two-dimensional discrete cosine transform are quantized by the codec 109 using a quantization table as shown in FIG. That is, the DC component value (F (0,0)) shown in FIG. 3 and the frequency component values (F (0,1), F (0,2),..., F (1,0), F (1,1),..., F (7,6), F (7,7)) are divided by the respective values in the table shown in FIG. Becomes a quantized value. As shown in FIG. 4, the value of the quantization table increases from the upper left to the lower right. As described above, since a large value of the frequency component tends to appear in the low-frequency component, that is, the value of the high-frequency component is small, most of the high-frequency component values are “ 0 ".

  Next, the quantized data is sequentially converted into a Huffman code by Huffman code table conversion and converted into Huffman encoded data (code stream) in block units in the codec 109, as shown in FIG. At this time, the quantized data is sequentially Huffman-coded from the upper left to the lower right as indicated by arrows in FIG. Also, the code stream by the Huffman encoded data continues from the encoded data of the leftmost block of the first block line of the original image to the encoded data of the block in the right direction shown in FIG. Next to the encoded data of the rightmost block, the encoded data of the leftmost block of the second block line continues from the encoded data of the rightmost block to the encoded data of the rightmost block of the nth block line. Connected to data. Here, as shown in FIG. 7, the “block line” means a line formed by blocks in a row in the horizontal direction from the left end to the right end of the image in the original image. In the example of FIG. 7, the original image is formed by first to nth (n is a natural number) block lines, and the first to mth (m is a natural number, m <n) block lines. It is assumed that the image can be displayed on the display device 106a without scrolling. Although detailed description of the Huffman code is omitted, as described above, since many values of the quantized data are “0”, the Huffman code can be highly compressed by Huffman coding. Since the quantized data for each block is compressed by Huffman coding, the data length of the Huffman coded data for each block is not constant.

  Finally, the codec 109 forms the code stream thus obtained into an image file (image data) in JPEG format.

  The restoration of the original image from the JPEG format image data generated in this way is the reverse of the procedure for generating the JPEG format image data, that is, the codec 109 decodes the Huffman encoded data with the Huffman encoding table. The quantization data is converted into quantized data, and the quantized data is inversely quantized with a quantization table to obtain a DC component and a frequency component in the frequency domain, and the DC component and the frequency component are subjected to inverse discrete cosine transform.

  By the way, in the present embodiment, data to be processed in each process of decoding Huffman encoded data and inverse quantization of quantized data in restoration of an original image from JPEG format image data. This reduces the required storage capacity of the RAM 103 and reduces the burden on the CPU 101.

  First, in the process of decoding the Huffman encoded data, the data amount of the Huffman encoded data to be decoded is reduced. That is, not all the Huffman encoded data forming the code stream is decoded, but only the Huffman encoded data corresponding to the block necessary for displaying the original image on the display device 106a. FIG. 8 shows the original image. In this figure, the displayable range (1) in which the original image can be displayed without scrolling on the display device 106a is a range delimited by the vertical broken line X1 and the horizontal broken line Y1. is there. In the example of FIG. 7, the range is from the first block line to the m-th block line.

  However, in anticipation that the original image is scroll-displayed, the original image in the range (image restoration range (2)) delimited by the vertical broken line X2 and the horizontal broken line Y2 is restored. Note that if the original image from the first block line to the xth block line is restored, m ≦ x ≦ n−1 may be satisfied. However, in consideration of a reduction in the second data amount described later, m ≦ x. Even if ≦ n, the effect of the present invention can be obtained. By the way, the Huffman encoded data for each block included in the code stream has a non-constant data length as described above. Cannot be distinguished. Therefore, in order to obtain the quantized data of the block corresponding to the image restoration range (2), the range delimited by the vertical broken line X3 and the horizontal broken line Y2 that exceeds the Huffman encoded data corresponding to the image restoration range (2) It is necessary to decode Huffman encoded data corresponding to (table decoding range (3)). Since the data length of the quantized data for each block is constant, it corresponds to the table decoding range (3) by monitoring and measuring the data amount of the quantized data converted and output by the Huffman code table conversion. It is determined whether the Huffman encoded data of the block has been decoded into quantized data. Note that the amount of quantized data of the block corresponding to the table decoding range (3) is calculated based on the size of the original image to be restored and the reduction ratio of the original image displayed on the display device 106a.

  Second, in the process of inverse quantization of quantized data, two types of data amounts are reduced. That is, the number of blocks to be dequantized is reduced, and the data to be dequantized in each block is reduced. That is, first, the quantized data of the block corresponding to the table decoding range (3) is generated by decoding the Huffman encoded data. Of the quantized data, the block corresponding to the image restoration range (2) is generated. Dequantize only the quantized data. Furthermore, in the inverse quantization of each block, the original image displayed on the display device 106a is not subjected to inverse quantization on the quantized data corresponding to the direct current component and the frequency component in the entire 8 × 8 range. Inverse quantization is performed only on quantized data corresponding to a DC component and a frequency component in a range corresponding to the reduction ratio of.

  In the frequency domain of FIG. 3, there is a characteristic that the higher frequency components are information related to the precise representation of the original image, and the lower frequency components are information related to the rough representation of the original image. Further, the direct current component is an average value of the values of all the pixels in one block of the original image, and has a great influence on the original image. When the original image displayed on the display device 106a is reduced, it is not necessary to express the image more precisely as the reduction ratio is higher. Accordingly, if the reduction ratio displayed on the display device 106a for the original image is increased, the frequency component in a wider frequency range from the high frequency to the low frequency is not required to restore the original image.

  Therefore, as shown in FIGS. 9 and 10, when the reduction ratio is 25% or less (including the case of enlargement), the quantization corresponding to the range (A) of the DC component and the frequency component of the 8 × 8 region Data, that is, quantized data of the entire range of the block is inversely quantized using a quantization table. When the reduction ratio is 26% or more and 50% or less, only the quantized data corresponding to the range (B) of the DC component and the frequency component of the 6 × 6 region is inversely quantized. When the reduction ratio is 51% or more and 75% or less, only the quantized data corresponding to the range (C) of the DC component and the frequency component of the 4 × 4 region is inversely quantized. Further, when the reduction ratio is not less than 76% and not more than 99%, only the quantized data corresponding to the range (D) between the DC component and the frequency component in the 2 × 2 region is inversely quantized. Note that all the values of frequency components for which no value is obtained without performing inverse quantization are set to “0”.

Finally, the DC component value (F (0,0)) thus obtained and the frequency component values (F (0,1), F (0,2),..., F (1 , 0), F (1,1), ..., F (7,6), F (7,7)) and the two-dimensional inverse discrete cosine transform defined by Equation 2 The value (f (x, y), (x, y = 0,1,2,3,4,5,6,7)) of each pixel of the original image block to be restored is obtained. Note that the codec 109 performs two-dimensional inverse discrete cosine transform by a known matrix calculation. Finally, the blocks obtained by the two-dimensional inverse discrete cosine transform are sequentially arranged by thinning out images corresponding to the reduction ratios and displayed on the display device 106a to restore the original image.

  The Internet facsimile machine 1 uses the above-described image data restoration method to perform the JPEG method every time JPEG image data is displayed on the display device 106a and every time the reduction ratio of the image displayed on the display device 106a is changed. The original image is obtained by restoring the image data.

  As described above, according to the image data restoration method and Internet facsimile apparatus 1 of the present embodiment, when restoring an original image from JPEG image data, the Huffman code is converted into quantized data by Huffman code table conversion. Reduce the amount of data. In the inverse quantization process of the quantized data, the inverse quantization is performed only on the blocks corresponding to the display range and scroll range of the original image on the display device 106a. Further, in the inverse quantization of each block, only the quantized data corresponding to the direct current component and the frequency component in the range corresponding to the reduction ratio in the display device 106a of the original image is performed. Therefore, according to the image data restoration method and the Internet facsimile apparatus 1 of the present embodiment, the amount of data to be handled is reduced, so that the storage capacity of the RAM 103 required for the restoration process of the original image is reduced and the CPU 101 is involved. The burden can be reduced.

The embodiment of the present invention is not limited to the above-described embodiment, and can be variously modified within the scope of the idea of the invention. For example, in the above-described embodiment, the amount of Huffman encoded data to be converted into quantized data, the amount of quantized data to be dequantized, and the amount of data in a block to be dequantized are reduced. However, instead of reducing all three types of data amount, reduce any one type of data amount or reduce any two types of data amount. You may make it perform.

  The present invention provides, for example, an image data restoration method for restoring image data compressed by the JPEG method to an original image, and an image processing device for restoring image data compressed by the JPEG method to an original image and displaying the original image on a display device It is applicable to.

1 is a block diagram illustrating a configuration example of an Internet facsimile machine 1 according to the present invention. It is explanatory drawing which shows the block (space area) which is a processing unit in the image compression of a JPEG system. It is explanatory drawing which shows the direct current | flow component and frequency component which are obtained by carrying out the two-dimensional discrete cosine transform of the block. It is explanatory drawing which shows a quantization table. It is explanatory drawing which shows conversion of the quantization data by Huffman code table conversion, and Huffman encoding data. It is explanatory drawing which shows the conversion order by the Huffman code table conversion of quantization data. It is explanatory drawing explaining the block and block line which form an original image. It is explanatory drawing which shows the range of the process target at the time of decompress | restoring an original image. It is explanatory drawing which shows the range of the reduction | decrease rate of the original image displayed on the display apparatus 106a, and the quantization data to inverse-quantize. It is explanatory drawing which shows the range of the reduction | decrease rate of the original image displayed on the display apparatus 106a, and the quantization data to inverse-quantize.

Explanation of symbols

1 Internet facsimile machine (image processing device)
109 codec (decoding means, inverse quantization means, inverse discrete cosine transform means)

Claims (4)

An image data restoration method for restoring image data compressed in JPEG format to an original image,
The original image is formed by n (n is a natural number) block lines, and the display range from the first block line to the m-th (m is a natural number and m <n) block line of the original image is a display range. If
The quantized data is decoded by decoding the Huffman encoded data corresponding to the block from the beginning of the first block line of the original image to the last block of the block line that is greater than or equal to the m-th block line and less than or equal to the (n-1) -th block line. Obtaining step A;
And a step B of restoring an original image based on the quantized data obtained by decoding in the step A. Step B is
Step B-1 of obtaining the DC component and the frequency component by dequantizing the quantized data of the block from the left end of the original image to the right end of the display range of the original image in the display device and less than the right end of the original image. When,
And B-2, which performs an inverse discrete cosine transform on the basis of the direct current component and the frequency component obtained by inverse quantization in the step B-1, and obtains an original image. The image data restoration method described. In the step B-1,
The quantized data corresponding to the DC component and the quantized data corresponding to the frequency component of the frequency up to the height corresponding to the reduction ratio of the original image displayed on the display device are inversely quantized. Item 3. The image restoration method according to Item 2. An image processing device that restores image data compressed in JPEG format to an original image and displays the image on a display device,
When the original image is formed by n (n is a natural number) block lines, and the range from the first block line to the m-th (m is a natural number) block line of the original image is a display range in the display device,
The quantized data is decoded by decoding the Huffman encoded data corresponding to the block from the beginning of the first block line of the original image to the last block of the block line that is greater than or equal to the m-th block line and less than or equal to the (n-1) -th block line. Obtaining decryption means;
Among the quantized data obtained by decoding by the decoding means, the quantized data of blocks from the left end of the original image to the right end of the display range of the original image on the display device and less than the right end of the original image is inverted. Inverse quantization means for quantizing to obtain a DC component and a frequency component;
Inverse discrete cosine transform means for obtaining an original image by performing inverse discrete cosine transform based on a direct current component and a frequency component obtained by inverse quantization by the inverse quantization means, and
The inverse quantization means inversely quantizes the quantized data corresponding to the DC component and the quantized data corresponding to the frequency component of the frequency up to the height corresponding to the reduction ratio of the original image displayed on the display device. An image processing apparatus.

Images