JP4678814B2 - Image processing method and apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing method and apparatus for encoding image data obtained by changing a predetermined area of image data of a basic image.
[0002]
[Prior art]
In recent years, with the spread of personal computers and mobile terminals, digital data communication (data communication) via the Internet has been widely performed. The representative digital data distributed in this data communication is still image data. Since such still image data has a large amount of data, normally, transmission and storage are performed in advance, and the data amount of still image data is kept small. As an encoding method of such still image data, for example, ISO / iEC10918-1 | iTU-TT 81 is a JPEG encoding method. In this JPEG encoding method, a sequence of image data is converted by discrete cosine conversion.
[0003]
Currently, a discrete wavelet transform is attracting attention as a method for transforming a series of image data. This is because, for example, the decoded image of the image encoded data generated by the discrete wavelet transform and the decoded image of the image encoded data generated by the discrete wavelet transform tend to generate less noise in the former. Because there is an advantage of.
[0004]
In addition, by encoding an image by discrete wavelet transform, it is possible to perform stepwise display (progressive display) of the decoded image so that the resolution and resolution of the decoded image improve as decoding progresses. With such progressive display, a viewer who receives and browses a still image can know rough information of the image at an early stage of receiving or decoding the image.
[0005]
Various applications of image data encoded by discrete wavelet transform capable of progressive display can be considered. One of them is information distribution from companies to customers according to the capacity of the communication line and the resolution of the data display device on the receiving side. In this information distribution, a company may send information having a partial area (change area) to be changed for each customer to many customers (FIGS. 2A to 2C).
[0006]
[Problems to be solved by the invention]
Conventionally, in such information distribution, when the information to be distributed is image data encoded by discrete wavelet transform, it is common to re-encode all of the image data. However, since such image data has a basic image in common and changed image data is added to each customer, the entire changed image is encoded again and transmitted. It was not efficient.
[0007]
The present invention has been made in view of the above conventional example, and an image processing method capable of performing more efficient image encoding when encoding image data in which a certain area of a basic image is changed, and An object is to provide an apparatus.
[0008]
[Means for Solving the Problems]
  In order to achieve the above object, the image processing apparatus of the present invention comprises the following arrangement. That is,
  An image processing apparatus for encoding a derivative image obtained by changing a basic image,
  First conversion means for converting image data of the basic image into conversion coefficients of a plurality of frequency bands; first quantization means for quantizing each of the conversion coefficients obtained by the first conversion means; First code block dividing means for dividing the transform coefficient of each frequency band quantized by one quantization means into first code blocks of an arbitrary size; and the first code block dividing means divided by the first code block dividing means. First entropy encoding means for entropy encoding each of one code block; and basic image encoding for generating encoded data of the basic image from encoded data generated by encoding by the first entropy encoding means A pixel for generating a second code block related to a change area having a pixel value different from that of the basic image in the derived image; Calculating means for outputting, a second converting means for converting the pixels calculated by the calculating means into coefficients of a plurality of frequency bands, and a first quantizing coefficient of each frequency band obtained by the second converting means. 2 quantizing means, second code block dividing means for dividing the coefficient of each frequency band quantized by the second quantizing means into code blocks of an arbitrary size, and dividing by the second code block dividing means A second entropy encoding means for entropy-encoding the generated code block, and a code block for generating encoded data of the second code block from the encoded data generated by encoding by the second entropy encoding means An encoded data generating unit; and an encoded data of the second code block generated by the code block encoded data generating unit. Derived image encoded data for generating encoded data of the derived image by replacing the corresponding encoded code block of the encoded data of the basic image generated by the basic image encoded data generation means And generating means.
[0010]
  In order to achieve the above object, the image processing method of the present invention comprises the following steps. That is,
  An image processing method for encoding a derivative image obtained by changing a basic image,
  A first conversion step of converting image data of the basic image into conversion coefficients of a plurality of frequency bands; a first quantization step of quantizing each of the conversion coefficients obtained in the first conversion step; A first code block dividing step of dividing the transform coefficient of each frequency band quantized in one quantization step into first code blocks of an arbitrary size; and the first code block dividing step divided in the first code block dividing step A first entropy encoding step for entropy encoding each code block; and basic image encoding for generating encoded data of the basic image from the encoded data generated by encoding in the first entropy encoding step A data generation step, and a calculation step of calculating a pixel for generating a second code block related to a change area having a pixel value different from that of the basic image in the derived image A second conversion step of converting the pixels calculated in the calculation step into coefficients of a plurality of frequency bands, and a second quantization step of quantizing the coefficients of each frequency band obtained in the second conversion step; , A second code block dividing step of dividing the coefficient of each frequency band quantized in the second quantization step into code blocks of an arbitrary size, and the code block divided in the second code block dividing step A second entropy encoding step for entropy encoding, and a code block encoded data generation step for generating encoded data of the second code block from the encoded data generated by encoding in the second entropy encoding step And the basic image encoded data by the encoded data of the second code block generated in the code block encoded data generating step. And a derived image encoded data generation step of generating encoded data of the derived image by replacing corresponding encoded code blocks of the encoded data of the basic image generated in the generating step. To do.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings.
[0013]
For example, as shown in FIG. 1, image data of a basic image (basic image 1000) is input to the image processing apparatus according to the present embodiment. Then, as shown in FIGS. 2A to 2C, different difference data 200 to 202 are added to predetermined areas of the basic image 1000 shown in FIG. 1 to generate a plurality of derived images 1001 to 1003. The Each of the plurality of derived images 1001 to 1003 is an image to be encoded. An area to which such difference data is added is called a changed area.
[0014]
In the image processing apparatus according to the present embodiment that encodes the plurality of derived images 1001 to 1003, the basic image 1000 is divided into four rectangular areas (tiles) 30 to 33 as shown in FIG. Each of the tiles thus obtained includes a common tile 30 to 32 common to each derived image, and a modified tile 33 including a modified area to which difference data 200 to 202 is added in the example illustrated in FIGS. And are separated. The common tiles 30 to 32 are encoded only once. Further, the encoding of the change tile 33 is repeated by the number of derived image data added to the change tile 33. In the present embodiment, it is assumed that there are M types of derived image data (M = 3 in the example of FIGS. 2A to 2C). In the example of FIG. 3, the changed tile is the lower right tile 33 of 2 × 2 tiles, but the position of the changed tile is not limited to this.
[0015]
Details of the method will be described below.
[0016]
The image data to be encoded in the still image in this embodiment is assumed to be 8-bit monochrome image data. However, for example, a monochrome image represented by a bit number other than 8 bits, such as 4 bits, 10 bits, or 12 bits for each pixel, or a color that represents each color component (RGB / Lab / YCrCb) in each pixel by 8 bits. The present invention can also be applied to multi-valued image data. Also, the present invention can be applied to a case where the image data is multi-value information representing the state of each pixel constituting the image, for example, a multi-value index value representing the color of each pixel. When applied to these, each type of multi-value information may be monochrome image data to be described later.
[0017]
FIG. 4 is a flowchart showing processing until basic image data is input to the image processing apparatus according to the present embodiment, encoded, and encoded data of M types of derived images is output.
[0018]
First, in step S1, the variable i is initialized to “1”, and image data (basic image data) of the basic image 1000 is input. In step S2, the basic image data is divided into a plurality of tiles, and each of the plurality of tiles is classified into a common tile (a tile that is not changed) and a changed tile (a tile to which additional data is added). In step S3, processing is assigned to each corresponding to the common tile or the changed tile. In the case of a common tile, the process proceeds to step S4, where the image data (common tile data) of the common tile is encoded and stored as encoded common tile data in the first buffer 104 (FIG. 5). If the image data is changed tile image data (changed tile data), the process proceeds to step S6, and the changed tile data is output to the second buffer 105 (FIG. 5) as it is. In this state, the changed tile data is tile data obtained by dividing the basic image, and does not include difference data as indicated by 200 to 202 in FIGS.
[0019]
When the storage processing of the encoded data of the common tile data and the changed tile data is completed in this way, the process proceeds to step S7, and the difference data (corresponding to 200 in the example of FIG. 2A) added to the i th derivative image is obtained. get. This difference data may be stored in advance in a memory or the like, or may be generated as appropriate. In step S8, the changed tile data is copied from the second buffer 105 to the added changed tile data generation unit 106 (FIG. 5). In step S9, the i-th difference data is added to the changed tile data to generate i-th added changed tile data. This is executed by the added changed tile data generation unit 106. In step S10, the i-th added changed tile data is encoded. In step S11, the i-th derived image code is generated by combining the encoded i-th added changed tile data and the encoded common tile data stored in the first buffer in step S5. Generate and output digitized data. In step S12, the variable i is compared with the value of M to check whether the generation processing of M types of derived image encoded data has been completed. If not, the variable i is incremented by one. Returning to S7, the above-described processing is executed. Then, when all of the generation processes of the M types of derived image encoded data are finished, this process is finished.
[0020]
The detailed description of the processing of the image processing apparatus according to the present invention will be described below.
[0021]
FIG. 5 is a block diagram showing the configuration of the image processing apparatus according to the embodiment of the present invention.
[0022]
In the figure, reference numeral 101 denotes a basic image data input unit for inputting basic image data. A tile dividing unit 102 divides the input basic image data into a plurality of tiles. A tile classification unit 103 classifies each tile into a common tile and a change tile. A first buffer 104 stores common tile data and encoded data obtained by encoding the common tile data. Reference numeral 105 denotes a second buffer for storing changed tile data. An added changed tile data generation unit 106 generates changed tile data in which the difference data generated by the difference data generation unit is added to the changed tile data. The difference data generation unit 107 generates difference data 200 to 202 as shown in FIGS. 2A to 2C, for example. Reference numeral 108 denotes a tile data encoding unit that inputs and encodes changed tile data to which common tile data and difference data are added. Reference numeral 109 denotes an i-th derived image data encoding unit, which is encoded data of changed tile data encoded by the tile data encoding unit 108 and encoded data of basic image data stored in the first buffer 104. And the encoded data of the derived image data is generated. Reference numeral 110 denotes an i-th derived image encoded image data output unit that outputs the i-th derived image encoded image data encoded by the i-th derived image data encoding unit 109.
[0023]
First, a case will be described in which i = 1st derivative image data is generated and encoded using i = 1st difference data among M types of difference data. Next, encoding of 1 <i ≦ Mth derivative image data will be described. Note that the difference between i = 1 and other cases is that when 1 <i ≦ M, the image of the common tile is encoded and stored. The processing is unnecessary.
[0024]
[When i = 1]
The basic image data input to the basic image data input unit 101 is divided into four tiles in the tile dividing unit 102 as described above with reference to FIG. Each of the four tiles thus generated is classified into a common tile and a changed tile by the tile classification unit 103. Next, among these classified tiles, common tile data is stored in the first buffer 104, and changed tile data is stored in the second buffer 105. The common tile data stored in the first buffer 104 is encoded by the tile data encoding unit 108 to be encoded common tile data, and is returned to the first buffer 104 and stored therein. The processing in the tile data encoding unit 108 will be described later.
[0025]
The changed tile data stored in the second buffer 105 is copied to the added changed tile data generation unit 106, and the first difference data generated by the difference data generation unit 107 is added to the changed tile data. First added modified tile data is generated. The first added changed tile data is encoded by the tile data encoding unit 108 to become encoded first added changed tile data, and is output to the i-th derived image encoded data generation unit 109. The i-th derived image encoded data generation unit 109 synthesizes the encoded first added modified tile data and the encoded image data of the common tile data stored in the first buffer 104, and further generates a header. . Then, as shown in FIG. 6, encoded data 603 of the first derivative image is generated from the header 600, the encoded data 601 of the common tile data, and the encoded first added modified tile data 602. The header 600 includes information such as the size of the basic image input to the basic image data input unit 101, information indicating whether the basic image is a binary image or a multi-valued image, and an image code to be transmitted. A character string indicating transmission / transmission apparatus, a transmission date and time, and the like are written.
[0026]
The encoded data of the first derivative image created in this way is output from the i-th derivative image encoded data output unit 110 to the outside.
[0027]
Next, the tile data encoding unit 108 will be described.
[0028]
FIG. 7 is a block diagram showing a configuration of tile data encoding section 108 according to the first embodiment.
[0029]
In the figure, 201 is a tile data input unit for inputting common tile data or first added changed tile data. A discrete wavelet transform unit 202 performs a discrete wavelet transform on tile data input from the tile data input unit 201. A buffer 203 stores the transform coefficient obtained by the discrete wavelet transform by the discrete wavelet transform unit 202. A coefficient quantizing unit 204 inputs these transform coefficients and quantizes them. Reference numeral 205 denotes an entropy encoding unit that entropy encodes the coefficient value quantized by the coefficient quantization unit 204. Reference numeral 206 denotes a tile encoded data output unit.
[0030]
The tile data input unit 201 receives common tile data or first added tile data. These tile data are supplied to the discrete wavelet transform unit 202. The discrete wavelet transform unit 202 performs discrete wavelet transform on the input tile data, and the generated discrete wavelet transform coefficients are divided into a plurality of frequency bands (subbands). Decompose. In the present embodiment, the discrete wavelet transform for the image data sequence x (n) is performed based on the following equation.
[0031]
d (n) = x (2n + 1) -floor {(x (2n) + x (2n + 2)) / 2}
r (n) = x (2n) + floor {(d (n-1) + d (n + 1) +2) / 4}
Here, r (n) and d (n) are conversion coefficients, r (n) is a low-frequency subband, and d (n) is a high-frequency subband. In the above formula, floor {X} represents the maximum integer value not exceeding X. This conversion formula is for one-dimensional data. By applying this conversion in the order of horizontal direction and vertical direction to perform two-dimensional conversion, LL, HL, LH, and so on as shown in FIG. It can be divided into 4 sub-bands of HH. Here, L indicates a low-frequency subband, and H indicates a high-frequency subband. Next, the LL subband is similarly divided into four subbands (FIG. 8B), and the LL subband is further divided into four subbands (FIG. 8C), for a total of ten subbands. make. Each of these 10 subbands is called HH1, HL1,..., HH3 as shown in FIG. Here, the number in the name of each subband is the level of each subband. That is, the level 1 subbands are HL1, HH1, and LH1, and the level 2 subbands are HL2, HH2, and LH2. The LL subband is a level 0 subband. Decoded data obtained by decoding subbands up to level n is referred to as level n decoded data. A decoded image obtained from level n decoded data is referred to as a level n decoded image. Here, the higher the level of the decoded image, the higher the resolution.
[0032]
It is assumed that the encoded image data transmitted to the decoding device in the present embodiment has 10 subbands as described above.
[0033]
The decoded data obtained by decoding all these subbands is called complete decoded data. A decoded image obtained by displaying this complete decoded data on the image display device is referred to as a complete decoded image. In the present embodiment, the level 3 decoded image is a completely decoded image.
[0034]
The ten subbands are temporarily stored in the buffer 203, and are in the order of LL, HL1, LH1, HH1, HL2, LH2, HH2, HL3, LH3, and HH3, that is, in order of the subbands having the lowest level to the higher level. And output to the coefficient quantization unit 204. The coefficient quantization unit 204 quantizes the wavelet transform coefficient of each subband supplied from the buffer 203 in a quantization step determined for each frequency component, and entropy-encodes the quantized value (coefficient quantization value). The data is output to the unit 205. Here, when the coefficient value is X and the value of the quantization step for the frequency component to which the coefficient belongs is q, the quantized coefficient value Q (X) is obtained by the following equation.
[0035]
Q (X) = floor {(X / q) +0.5}
However, in the above formula, floor {X} represents the maximum integer value not exceeding X.
[0036]
FIG. 9 is a diagram illustrating a correspondence example between each frequency component and the quantization step in the present embodiment. As shown in the figure, a larger quantization step is given to the high frequency subbands (HL1, LH1, HH1, etc.) than the low frequency subbands (LL, etc.).
[0037]
The entropy encoding unit 205 generates an entropy encoded value by entropy encoding the input coefficient quantization value by arithmetic encoding. The entropy encoded value is output to the tile encoded data output unit 206. As shown in FIG. 10, the tile encoded data output unit 206 arranges the input entropy encoded values in units of subbands, adds a tile header to the head thereof, and generates tile encoded data. .
[0038]
FIG. 11 is a diagram illustrating an array of three common tile encoded data.
[0039]
In this way, the encoded common tile data is sent from the tile encoded data output unit 206 to the first buffer 104 and stored. Further, in the case of the encoded first added tile data, it is output to the i-th derived image data encoding unit 109.
[0040]
[When 1 <i ≦ M]
In the encoding of the derived image data in the case of 1 <i ≦ M, the encoding process of the common tile data has already been completed and stored in the first buffer 104. The added changed tile data generation unit 106 adds the i-th difference data to the changed tile data to generate the i-th added tile data, and the tile data encoding unit 108 encodes the i-th added tile data. . The i-th derived image encoded data generation unit 109 receives the encoded common tile data stored in the first buffer 104 and the encoded first added tile data from the tile data encoding unit 108. Then, the i-th derivative image encoded data is generated and output from the i-th derivative image encoded data output unit 110 to the outside.
[0041]
As described above, according to the image processing apparatus of the first embodiment, basic image data is divided into a plurality of tiles, and the obtained tiles are classified into common tiles and changed tiles, and are encoded independently. At this time, the common tile data is encoded only once, and the changed tile data is encoded a number of times according to the type of image data added to the tile data. In this way, efficient encoding can be achieved without encoding all the data of the derived image data including the changed tile data.
[0042]
[Embodiment 2]
In the first embodiment, the basic image data is divided into a plurality of tiles, the obtained tiles are divided into common tiles and modified tiles, the common tile data is encoded only once, and the modified tile data is derived from the derived image. Only the types were encoded.
[0043]
On the other hand, in the second embodiment, encoding is performed without dividing the basic image into a plurality of tiles. However, in the second embodiment, as shown in FIG. 12, in the encoding process, each subband obtained by performing discrete wavelet transform on the image data is divided into code blocks and then entropy encoded. . Further, after determining where to add the difference data in the basic image, the pixel (change code block generation pixel) for generating the code block (change code block) related to the change area of each derived image is calculated and the change is performed. The generated pixel of the code block is encoded. The encoded data of the changed code block generated in this way is exchanged with the encoded data of the corresponding code block which is the encoded data of the code block (corresponding code block) spatially corresponding to the code block. Thereby, a plurality of derivative images are generated.
[0044]
The operation in the image processing apparatus according to the second embodiment will be described in detail below.
[0045]
FIG. 13 is a block diagram showing the configuration of the image processing apparatus according to the second embodiment.
[0046]
In the figure, 1801 is a derived image data encoding unit, 1802 is an external output unit, and 1803 is a control unit. Here, the plurality of derived images are encoded by the derived image data encoding unit 1801, then temporarily stored in the external output unit 1802, and output to the outside. The control unit 1803 controls the output of the derived image encoded data stored in the external output unit 1802 to the outside.
[0047]
FIG. 14 is a block diagram showing a configuration of derived image data encoding section 1801 in the image processing apparatus according to the second embodiment.
[0048]
In the figure, 1701 is a basic image input unit, 1702 is a buffer, 1703 is a basic image data encoding unit, 1704 is a buffer, 1705 is a difference data position determination unit, 1706 is a modified code block generation pixel calculation unit, and 1707 is an i-th difference. A data generation unit, 1708 is an i-th changed code block generation pixel encoding unit, 1709 is a code block replacement unit, 1710 is an i-th derivative image encoded data output unit, and 1711 is a difference data erasing unit.
[0049]
FIG. 15 is a flowchart showing processing in the image processing apparatus according to the second embodiment.
[0050]
In the figure, first, in step S21, basic image data is input by the basic image input unit 1701. At this time, the variable i is initialized to “1”. In step S22, the input basic image data is sent to the buffer 1702 and temporarily stored. In step S23, the basic image data stored in the buffer 1702 is supplied to the basic image data encoding unit 1703. Thus, in step S24, the basic image data encoding unit 1703 encodes the basic image data to generate encoded basic image data. In step S25, the basic image encoded data is output to the buffer 1704 and temporarily stored.
[0051]
In step S26, the i-th difference data generation unit 1707 generates the i-th difference data (i-th difference data), and sends the area information to the difference data position determination unit 1705. Thereby, the difference data position determination unit 1705 determines the position of the difference data in the basic image data from the area information as shown in FIG. The position information thus determined is sent to the changed code block generation pixel calculating unit 1706, and a changed code block (a code block including difference data) is calculated as shown in FIG. 22 (step S27). The calculation method of the change code block generation pixel will be described in detail later.
[0052]
Next, the process proceeds to step S28, and based on the calculation result of step S27, the pixel of the basic image data determined to be the generated pixel of the changed code block is read from the buffer 1702, and the i-th difference data is further converted to the i-th difference data. The i difference data generation unit 1707 outputs the result to the i-th changed code block generation pixel encoding unit 1708. In step S29, the i-th changed code block generation pixel encoding unit 1708 encodes the generated code of the changed code block to generate encoded data of the i-th changed code block. Thereafter, the process proceeds to step S30, in which the encoded data of the i-th changed code block and the encoded data of the i-th corresponding code block in the basic image encoded data are replaced, and the i-th derived image of the i-th derived image is replaced. Encoded data is generated and input to the i th derivative image encoded data output unit 1710. Accordingly, the encoded data of the i-th derivative image is output to the external output unit 1802 (FIG. 13) via the i-th derivative image encoded data output unit 1710. In step S31, the difference data deletion unit 1711 deletes the i-th difference data existing in the i-th difference data generation unit 1707. In step S32, the variable i is compared with a value M indicating the number of types of difference data to check whether the processing of the M types of difference data has been completed. If not, the variable i is incremented by one. Returning to step S26, the above-described processing for the i-th difference data is repeated. Thus, when the processing of the M types of difference data is completed, this processing is terminated.
[0053]
FIG. 16 is a block diagram showing a configuration of basic image data encoding section 1703 according to the second embodiment.
[0054]
In the figure, 1301 is a basic image data input unit, 202 is a discrete wavelet transform unit, 203 is a buffer, 204 is a coefficient quantization unit, 1305 is a code block division unit, 1306 is an entropy encoding unit, and 1307 is basic image encoded data. It is an output unit. In FIG. 16, parts common to those in FIG. 7 according to the first embodiment are denoted by the same reference numerals, and description thereof is omitted.
[0055]
FIG. 23 is a block diagram showing a configuration of the i-th modified code block generation pixel encoding unit 1708 according to the second embodiment, and here, the same symbols are assigned to the same parts as those in FIG. 7 according to the first embodiment. The description thereof is omitted.
[0056]
In FIG. 23, 1901 is an i-th changed code block generation pixel input unit, 1305 is a code block dividing unit, 1306 is an entropy encoding unit, and 1902 is an i-th changed code block encoded data output unit.
[0057]
Hereinafter, the basic image data encoding unit 1703 (FIG. 16), the modified code block generation pixel calculation unit 1706, the i-th modified code block generation pixel encoding unit 1708 (FIG. 23), the code block replacement unit 1709, the first illustrated in FIG. The process of the i-derived image encoded data output unit 1710 will be described.
[0058]
[Basic Image Data Encoding Unit 1703]
In FIG. 16, basic image data is input from a basic image data input unit 1301, and is sequentially processed by a discrete wavelet transform unit 202, a buffer 203, and a coefficient quantization unit 204. Since the processing in the discrete wavelet transform unit 202, the buffer 203, and the coefficient quantization unit 204 is the same as the processing in the first embodiment described above, description of these processing units is omitted.
[0059]
Each coefficient value quantized by the coefficient quantization unit 204 is input to the code block dividing unit 1305 in the form of each subband, and is divided into rectangular code blocks as shown in FIG. As the size of the code block, 2n × 2m (where n and m are positive constants and n = m = 4 in FIG. 12) or the like can be considered.
[0060]
The code block generated in this way is input to the entropy encoding unit 1306 and is entropy encoded. The encoded value obtained by the entropy encoding is output from the basic image encoded data output unit 1307 to the buffer 1704 (FIG. 14) and temporarily stored.
[0061]
[Changed code block generation pixel calculation unit 1706]
Processing in the changed code block generation pixel calculation unit 170 will be described with reference to FIGS. 17 to 22. Here, the description is simplified by using a one-dimensional pixel data string.
[0062]
FIG. 17 shows a state in which the low-frequency component L2 is generated by performing discrete wavelet transform on the derived image, and the low-frequency component L1 of the low-frequency component L2 is further generated. Here, as shown in FIG. 18, when the pixels painted in black are the change region, the coefficients of the subbands generated from the change region are the coefficients indicated by 1900 to 1904 in FIG. From this, code block division is further performed as shown in FIG. The resulting change code block is as shown in FIG. From there, as shown in FIG. 22, pixels 1900 to 1904 are determined as changed code block generation pixels.
[0063]
[I-th modified code block generation pixel encoding unit 1708]
The processing in the i-th change code block generation pixel encoding unit 1708 is that the i-th change code block generation pixel is input to the i-th change code block generation pixel input unit 1901 in FIG. The encoded data of the i-th modified code block is output from the encoded data output unit 1902 of the above, and is different from the processing of the basic image data encoding unit 1703 (FIG. 16) described above. However, the other processes are the same as those of the basic image data encoding unit 1703 (FIG. 16) described above.
[0064]
[Code block replacement unit 1709]
The code block replacement unit 1709 copies the encoded data of the basic image stored in the buffer 1704 to a work buffer existing in the code block replacement unit 1709. Further, the encoded data of the corresponding code block in the encoded data of the basic image is replaced with the encoded data of the i-th changed code block input from the i-th changed code block generation pixel encoding unit 1708. Further, subbands entropy-encoded in units of code blocks are arranged as shown in FIG. 24 to generate quasi-i-th derivative image encoded data and output to the i-th derivative image encoded data output unit 1710.
[0065]
[I-th derivative image encoded data output unit 1710]
As shown in FIG. 25, the i-th derived image encoded data output unit 1710 adds a header to the head of the input semi-i-th derived image encoded data, and generates i-th derived image encoded data. . This header includes information such as the size of the image input to the basic image input unit 1701, the type indicating whether the image is a binary image or a multi-valued image, and characters indicating the image processing apparatus to be transmitted. The column, transmission date and time, etc. are written. The i-th derivative image encoded data generated in this way is output to the external output unit 1802 (FIG. 13).
[0066]
As described above, according to the image processing apparatus according to the second embodiment, the subbands of the image data subjected to discrete wavelet transform are divided into code blocks. Thereafter, the modified code block of each derived image is encoded. In this way, efficient encoding of derived image data can be performed.
[0067]
[Embodiment 3]
The encoding method according to the second embodiment described above not only enables efficient generation of derived image data, but also achieves memory saving in the entropy encoding unit 1306 by processing in units of code blocks. However, memory saving in the entire coding has not been achieved.
[0068]
On the other hand, in the third embodiment, a method for efficiently generating derivative image data while saving memory of the entire encoding will be described.
[0069]
FIG. 26 is a block diagram showing the configuration of the image processing apparatus according to the third embodiment. This is obtained by replacing the derived image data encoding unit 1801 (FIG. 13) in the image processing apparatus of the second embodiment with a derived image data encoding unit 2201.
[0070]
FIG. 27 is a block diagram showing a configuration of derived image data encoding section 2201 according to Embodiment 3. This is obtained by replacing the basic image data encoding unit 1703 (FIG. 14) in the derived image data encoding unit 1801 according to Embodiment 2 with a basic image data encoding unit 2301.
[0071]
FIG. 28 shows a block diagram of basic image data encoding section 2301 according to Embodiment 3. In this, a tile division unit 1501 is inserted between the basic image data input unit 1301 and the discrete wavelet transform unit 1302, and the basic image encoded data output unit 1307 in FIG. 16 is replaced with a basic image encoded data output unit 1502. Is.
[0072]
When basic image data is input to the basic image data input unit 1301 in the basic image data encoding unit 2301 (FIG. 27), the basic image data is tiled (tiles 0 to 3) as shown in FIG. 29 by the tile dividing unit 1501. It is divided into. Each tile is independently encoded as in the second embodiment.
[0073]
As shown in FIG. 30, the basic image encoded data output unit 1502 generates basic image encoded data by arranging encoded tile data, and buffers the basic image encoded data. 1704 is output. Note that the processing after the basic image encoded data is output to the buffer 1704 is the same as that in the above-described second embodiment, and thus the description of the corresponding processing is omitted.
[0074]
As described above, according to the image processing apparatus according to the third embodiment, the image data is divided into a plurality of tiles, and each tile is encoded in the same manner as the encoding method in the above-described second embodiment. In this way, the derived image data can be efficiently generated while saving the memory of the entire encoding.
[0075]
In each of the above-described embodiments, image data is orthogonally transformed by discrete wavelet transform. However, the present invention is not limited to this. For example, image data is hierarchically encoded according to resolution. The method is also included in the scope of the present invention.
[0076]
Note that the present invention can be applied to a system (for example, a copier, a facsimile machine, etc.) composed of a single device even if it is applied to a system composed of a plurality of devices (for example, host computer, interface device, reader, printer, etc.). May be.
[0077]
Another object of the present invention is to supply a storage medium (or recording medium) on which a program code of software that realizes the functions of the above-described embodiments is recorded to a system or apparatus, and to computer (or CPU or MPU) of the system or apparatus ) Is also achieved by reading and executing the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the storage medium storing the program code constitutes the present invention. Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an operating system (OS) running on the computer based on the instruction of the program code. A case where part or all of the actual processing is performed and the functions of the above-described embodiments are realized by the processing is also included.
[0078]
Further, after the program code read from the storage medium is written in a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer, the function is determined based on the instruction of the program code. The case where the CPU of the expansion card or the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing is also included.
[0079]
As described above, according to the present embodiment, it is possible to efficiently apply various change processes to one image and perform encoding.
[0080]
【The invention's effect】
As described above, according to the present invention, when encoding image data in which a certain region of a basic image is changed, there is an effect that more efficient image encoding can be performed.
[Brief description of the drawings]
FIG. 1 is a diagram showing an example of a basic image according to the present embodiment.
FIG. 2 is a diagram showing an example of a derived image according to the present embodiment.
FIG. 3 is a diagram showing an example of tile division of a basic image according to the present embodiment.
FIG. 4 is a flowchart showing processing for generating encoded data of a plurality of types of derivative images in the image processing apparatus according to Embodiment 1 of the present invention.
FIG. 5 is a block diagram showing a configuration of an image processing apparatus according to Embodiment 1 of the present invention.
FIG. 6 is a diagram illustrating encoded data of a first derivative image.
7 is a block diagram showing a configuration of a tile data encoding unit 108 according to the first embodiment. FIG.
FIG. 8 is a diagram illustrating a conversion result by discrete wavelet transform.
FIG. 9 is a diagram for explaining a relationship between a frequency component of a transform coefficient and a quantization step in the present embodiment.
FIG. 10 is a diagram illustrating tile encoded data according to the present embodiment.
FIG. 11 is a diagram for describing quasi-image encoded data including encoded data obtained by encoding common tile data.
FIG. 12 is a diagram for explaining an example of code block division according to Embodiment 2 of the present invention;
13 is a block diagram showing a configuration of an image processing apparatus according to Embodiment 2. FIG.
14 is a block diagram showing a configuration of a derived image data encoding unit 1801 in the image processing apparatus according to Embodiment 2. FIG.
FIG. 15 is a flowchart illustrating a process of generating encoded data of a derived image in the image processing apparatus according to the second embodiment. Descriptive image
16 is a block diagram showing a configuration of a basic image data encoding unit 1703 according to Embodiment 2. FIG.
FIG.
FIG.
FIG.
FIG.
FIG.
FIG. 22 is a diagram for explaining a modified code block generation pixel according to the second embodiment.
23 is a block diagram showing a configuration of an i-th modified code block generation pixel encoding unit 1708 according to Embodiment 2. FIG.
FIG. 24 is a diagram for describing quasi-th derivative image encoded data according to Embodiment 2;
25 is a diagram for describing i-th derivative image encoded data according to Embodiment 2. FIG.
FIG. 26 is a block diagram showing a configuration of an image processing apparatus according to Embodiment 3 of the present invention.
27 is a block diagram showing a configuration of a derived image data encoding unit 2201 according to Embodiment 3. FIG.
FIG. 28 is a diagram showing a block diagram of a basic image data encoding unit 2301 according to Embodiment 3.
FIG. 29 is a diagram showing an example of division of a basic image according to the third embodiment.
30 is a diagram for describing basic image encoded data according to Embodiment 3. FIG.

Claims (10)

An image processing apparatus for encoding a derived image changing the basic image image,
First conversion means for converting image data of the basic image into conversion coefficients of a plurality of frequency bands;
First quantization means for quantizing each of the transform coefficients obtained by the first transform means;
First code block dividing means for dividing the transform coefficient of each frequency band quantized by the first quantization means into first code blocks of an arbitrary size;
First entropy encoding means for entropy encoding each of the first code blocks divided by the first code block dividing means;
Basic image encoded data generating means for generating encoded data of the basic image from encoded data generated by encoding by the first entropy encoding means;
Calculating means for calculating a pixel for generating a second code block related to a change area having a pixel value different from that of the basic image in the derived image;
Second conversion means for converting the pixels calculated by the calculation means into coefficients of a plurality of frequency bands;
Second quantization means for quantizing the coefficients of each frequency band obtained by the second conversion means;
Second code block dividing means for dividing the coefficient of each frequency band quantized by the second quantization means into code blocks of an arbitrary size;
Second entropy encoding means for entropy encoding the code block divided by the second code block dividing means;
Code block encoded data generating means for generating encoded data of the second code block from encoded data generated by encoding by the second entropy encoding means;
The encoded data corresponding to the encoded data of the basic image generated in the basic image encoded data generating means by the encoded data of the second code block generated by the code block encoded data generating means Derived image encoded data generating means for generating encoded data of the derived image by replacing blocks;
An image processing apparatus comprising: The image processing apparatus according to claim 1 , wherein the first and second conversion units perform conversion by discrete wavelet transform. The image processing apparatus according to claim 2 , wherein the first and second conversion units recursively divide the low frequency band. Further comprising a dividing means for dividing the image data of the basic image into a plurality of tiles;
Said first converting means, the image processing apparatus according to claim 1, characterized in that converting the image data of each tile divided by the dividing means. An image processing method for encoding a derived image changing the basic image image,
A first conversion step of converting the image data of the basic image into conversion coefficients of a plurality of frequency bands;
A first quantization step for quantizing each of the transformation coefficients obtained in the first transformation step;
A first code block dividing step of dividing the transform coefficient of each frequency band quantized in the first quantization step into first code blocks of an arbitrary size;
A first entropy encoding step of entropy encoding each of the first code blocks divided in the first code block dividing step;
A basic image encoded data generation step of generating encoded data of the basic image from the encoded data encoded and generated in the first entropy encoding step;
A calculation step of calculating a pixel for generating a second code block related to a change area having a pixel value different from that of the basic image in the derived image;
A second conversion step of converting the pixels calculated by the calculation step into coefficients of a plurality of frequency bands;
A second quantization step of quantizing the coefficients of each frequency band obtained in the second conversion step;
A second code block dividing step of dividing the coefficient of each frequency band quantized in the second quantization step into code blocks of an arbitrary size;
A second entropy encoding step of entropy encoding the code block divided in the second code block dividing step;
A code block encoded data generation step for generating encoded data of the second code block from the encoded data generated by encoding in the second entropy encoding step;
The encoded data corresponding to the encoded data of the basic image generated in the basic image encoded data generating step by the encoded data of the second code block generated in the code block encoded data generating step Derived image encoded data generation step for generating encoded data of the derived image by replacing blocks;
An image processing method comprising: 6. The image processing method according to claim 5 , wherein the transform coefficient is obtained by discrete wavelet transform in the first and second transforming steps. The image processing method according to claim 6 , wherein in the first and second conversion steps, the low frequency band is recursively divided. Has the further division step of dividing the image data of the basic image into a plurality of tiles, in the first conversion step, claim, characterized in that converting the image data of each tile divided by the dividing step 5 An image processing method described in 1. A computer-readable storage medium storing a program for causing a computer to execute the image processing method according to any one of claims 5 to 8 . A program for causing a computer to execute the image processing method according to any one of claims 5 to 8 .

Images