JPH1028226A - Method for converting compressed image data format and device therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for converting a compressed image data format for format conversion in a short processing time, without requiring a large memory area. SOLUTION: An offset address in a Huffman code length for the (n) line block of luminance components is calculated for unit blocks, constituted of the (n)×(n) small blocks of original compressed picture data (step 902), the offset address in the Huffman code length of color difference components is calculated for each unit block (step 908), and the original compressed image data are stored, based on the offset address in the Huffman code length of the luminance components and the offset address in the Huffman code length of the color difference components (step 909). Thus, compressed image data with the desired format can be generated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a compressed image data format conversion method and apparatus for converting a format of compressed image data into a different format.

[0002]

2. Description of the Related Art JPEG (Joint Photographic Coding Exper)
Compressed image data based on the ts Group) standard or the DCT (Discrete Cosine Transform) method has the same data content but different headers, and there are data with the same file format but different data structures.

Therefore, when it is desired to convert data having a different header or a different data structure into another file format and save it, the compressed image data cannot be temporarily stored in a memory. In such a manner that the image data is expanded (decoded) and converted to compressed image data of a desired file format.

[0004]

However, according to such a method, when converting to another file format, a free memory of a size corresponding to a developed image is required, so that a sufficient memory capacity cannot be prepared. Is not supported, and after being expanded on the memory, it is converted to compressed image data of the desired file format again,
The conversion process takes a long time, and further, there is a problem that the data accuracy of the newly generated desired file format is reduced.

The present invention has been made in view of the above circumstances, and provides a compressed image data format conversion method and apparatus capable of performing format conversion in a short processing time without requiring a large memory area. The purpose is to:

[0006]

According to the first aspect of the present invention,
The image data encoded for each block is configured in a predetermined format for each predetermined unit, and the compressed image data configured in the predetermined format is converted to another format configured by another unit different from the predetermined unit. The method for converting a compressed image data format to be converted includes a step of obtaining an address of each block of the predetermined unit, and a step of configuring the block into the other unit based on the obtained address.

According to a second aspect of the present invention, in the first aspect, the step of obtaining an address includes a step of obtaining a code length of the block and a step of obtaining an address by adding the code length. ing.

According to a third aspect of the present invention, in the first aspect, the step of configuring the other unit includes the step of converting the block of the encoded image data into a plurality of frequency components, and And configuring the converted image data into another unit for each frequency component based on the address.

According to a fourth aspect of the present invention, in the first aspect, the step of obtaining an address includes a step of obtaining an address of compressed image data for each block of one block line. The step of composing comprises sequentially compressing the compressed image data of one block line for each line based on the obtained address.
And converting to a GB signal.

According to a fifth aspect of the present invention, the image data encoded for each block is formed in a predetermined format for each predetermined unit, and the compressed image data formed in the predetermined format is different from the predetermined unit. In a compressed image data format conversion device for converting to another format constituted by another unit, a means for obtaining an address for each block of the predetermined unit, and the block for converting the block based on the address obtained by the means for obtaining the address. Conversion means for converting to another unit.

According to a sixth aspect of the present invention, in the fifth aspect, the means for obtaining the address includes: a block code length detecting means for obtaining a code length of the block; and a code length obtained by the block code length detecting means. And an adding means for obtaining an address by adding

According to a seventh aspect of the present invention, in the fifth aspect, the conversion means converts the coded image data block into a plurality of frequency components, and the address obtaining means. Means for converting the image data converted by the frequency component conversion means on the basis of the address obtained by the above into other units for each frequency component.

According to an eighth aspect of the present invention, in the fifth aspect, the means for obtaining the address includes means for obtaining an address of the compressed image data for each block of one block line, and the converting means includes: There is provided means for sequentially converting the compressed image data for one block line into RGB signals for each line based on the address obtained by the address obtaining means. As a result, according to the present invention, the capacity of the memory used for format conversion can be reduced, and the processing time for format conversion can be significantly reduced.

[0014]

Embodiments of the present invention will be described below with reference to the drawings. In the following embodiment, 8 × 8
A dot is defined as one block, one line in the row direction of this block is defined as one line block, a set of minimum blocks based on a predetermined format is defined as a unit block, and one line in the row direction of one dot is simply defined as one line. (First Embodiment) FIG. 1 shows a schematic configuration of a format conversion apparatus applied to a compressed image data format conversion method of the present invention. In FIG. Are input unit 2, output unit 3, ROM4
And RAM5. Here, the input unit 2
The data of the file format before conversion is input, and the control unit 1 performs a format conversion process on the input data of the file format by the control program of the ROM 4, and outputs the converted file format data from the output unit 3. I am trying to do it. The RAM 5 stores data for format conversion control.

Next, an example of converting the file format of the non-interleaved compressed image data into the JPEG format file format by using such a format converter will be described.

Here, the file format structure of the compressed image data based on the non-interleave method includes a header 11, a quantization table 12, and compressed image data 13 as shown in FIG. In this case, the compressed image data 13 follows the luminance Y component, the color difference Cb, and the Cr component in one image unit according to the non-interleave method.

FIGS. 3 (a), 3 (b), 3 (c) and 4 illustrate the procedure for generating such compressed image data 13. As shown in FIG. 3, the original image data is composed of 480 × 240 dots. First, assuming that the row block number is i and the column block number is j, the luminance Y component is set to j = 0 in step 401 in the flow shown in FIG.
At step 02, 8 × 8 dot (one block) image data is encoded from the i row and j column. At step 403, the j column is shifted one block at a time in the column direction (while +8 is reached). The same operation is repeated until <M (= 480) is denied. Then, step 404
If j <M (= 480) is denied, step 40
5, the i-th row is not shifted in the direction of one block row (while +8).
Until the result of step 0) is negative, the operations in steps 401 to 404 are repeated to generate a luminance Y component.

For the color difference Cb and Cr components, M (= 240), N
By executing the flow shown in FIG. 4 from the state of (= 120), the color difference Cb and Cr components are generated.

On the other hand, as shown in FIG. 5, the JPEG format file format has a header 1 including a quantization table.
4 and the compressed image data 15. As shown in FIG. 6, the compressed image data 15 in this case has luminance Y (YM) for each MCU (Minimum Coded Unit) unit which is the minimum data unit.
1, Y2, Y3, Y4) components, color difference Cb, Cr components, followed by an interleaving system.

FIGS. 7 (a), 7 (b), 7 (c) and 8 illustrate the procedure for generating such compressed image data 15. In this case, as shown in FIG.
In the case of × 240 dots, the number of MCUs is 30 × 15 = 45
Assuming that the MCU number is k and k is 0, in the flow shown in FIG.
2, the luminance Y (Y1, Y2, Y) of the image data for one MCU
3, Y 4) and the color differences Cb and Cr in this order, and in step 803, while k + 1, k <4 in step 804.
The same operation is repeated until 50 is denied to generate the luminance Y component and the color differences Cb and Cr.

Thus, the file format of the compressed image data based on the non-interleave described in FIG.
In order to convert to the JPEG format compressed image data file format described in FIG. 5, the flow shown in FIG. 9 is executed.

In this case, assuming that the original image data shown in FIG. 7A is 480 × 240 dots, 8 × 8 dots are set as a small block, and a small block consisting of 4 × 2 × 2 is a unit block. This unit block is divided horizontally by 30
The arrangement of the large blocks is referred to as a large block, and the vertical numbers of these large blocks are w (w = 1, 2,..., 15). Also, the number of the unit block in the row direction (for the luminance Y component) is i (i = 1, 2,..., 30), and similarly, the number in the row direction (for the color difference Cb, Cr components) is k (k = 1, 2, 3). ,..., 30), and j (j = j)
1, 2), and the block number i of the luminance Y component.
And the offset address of the Huffman code length for the j-th two books in the unit block of the same block number i is PYj
[I], the offset address of the k-th Huffman code length of the color difference Cb component is PCb [k], and the offset address of the k-th Huffman code length of the color difference Cr component is PCr [k].

First, at step 901, w = 1, PYj
[I] = 0, PCb [k] = 0, PCr [k] = 0, and in step 902, i = j = k = 1 and step 90
Proceed to 3.

In step 903, first, j (=
1) Obtain an offset address for a line block. In this case, the block number i (= 1) of the luminance Y component
And j in the unit block of the same block number i (= 1)
The offset address PYj [i] of the Huffman code length for the (= 1) second book is obtained. In this case, PYj [i]
Is assigned to the offset address PYj [i-1] of the previous unit block of the unit block having the block number i (= 1),
It is determined from the sum of the Huffman code lengths of the next two small blocks.

Next, at step 904, while incrementing for each unit block (while i + 1), at step 905, until one line block ends (i ≦
Until step 30 is denied), the operation of step 903 is repeated. Then, if i ≦ 30 is denied in step 905, in step 906, j is incremented (j +
1) At the same time, the last address of the previous block is set as the start address of the current block.

In step 9061, i = 1, and an offset address is obtained for the j (= 2) line block in the same manner as described above. Then, offset addresses for two line blocks of the luminance Y component are obtained, and if j ≦ 2 is denied in step 907, the process proceeds to step 908.

In step 908, the offset address of each color difference component for each unit block is obtained. In this case, the offset address PCb [k] of the kth Huffman code length of the chrominance Cb component is obtained from the sum of PCb [k-1] and the Huffman code length of one block, and the kth of the chrominance Cr component is obtained.
Offset address PCr [k] of the Huffman code length
Is calculated from the sum of PCr [k-1] and the Huffman code length for one block.

Next, in step 909, the compressed image data based on the non-interleave of the original image data is
Store in F system. In this case, the compressed image data is stored at the offset addresses PY1 [k], PY2 [k], PCb
[K] and PCr [k] in this order up to the next offset address, and the luminance Y (Y1, Y1) corresponding to one MCU shown in FIG.
2, Y3, Y4) components, color difference Cb, Cr components, and subsequent interleaved compressed image data 15 are generated.

Then, the operations of steps 908 and 909 are repeated while k + 1 is determined in step 910 until k ≦ 30 is denied in step 911. Then, if k ≦ 30 is denied in step 911, the address is updated in step 912, and
In step 3, while w is incremented (while w + 1), in step 914, until w ≦ 15 is negated.
The operation after step 902 is repeated. Then, if w ≦ 15 is denied in step 914, the format conversion process is completed for one screen.

Accordingly, in this manner, in order to convert the compressed image data of the file format based on the non-interleave shown in FIG. 2 to the JPEG format file format as shown in FIG. What has been expanded (decoded) and then converted to the JPEG file format again,
For a unit block consisting of n × n small blocks, an offset address of the Huffman code length of n line blocks of the luminance component is obtained, and an offset address of the Huffman code length of the chrominance component is obtained for each unit block. Compressed image data of a file format based on non-interleaving is stored based on the offset address of the Huffman code length and the offset address of the Huffman code length of the color difference component. Compared with the case using a memory, the capacity of the memory used for format conversion can be reduced, and the processing time for format conversion can be significantly reduced.

The reverse (from the JPEG format to the non-interleaved format) is also possible. This makes the method of obtaining the offset address slightly complicated. In the JPEG format, since it is stored for each MCU, an address pointer for two line blocks for luminance, an address pointer for one line block for each color difference, and a buffer corresponding to these are required. First, a start pointer for luminance and color difference is set to an address pointer. The data in the area indicated by the address pointer is read out and stored in the buffer. At this time, regarding the luminance, the first two blocks are stored in an area corresponding to the first line block, and the subsequent two blocks are stored in an area corresponding to the second line block. Then, the first and second line block addresses are updated, respectively. This is repeated for two line blocks to obtain compressed image data for two line blocks in terms of luminance. Then, each line block can be converted into a non-interleaved system by sequentially reading the blocks in the row direction.

In the above description, the blocks are 8 × 8 square blocks. However, the present invention is not limited to this, and M × N rectangular blocks may be used. (Second Embodiment) In this second embodiment, FIG.
The case where the file format of the compressed image data based on the non-interleave described above is converted into the file format of the progressively built-up JPEG format compressed image data by using the format conversion apparatus described in the above section is shown.

In this case, the progressive build-up
As shown in FIG. 10, the JPEG format file format has various definitions 16 and scan headers 17 of # 1 to #N.
And # 1 to #N scan compressed image data 18 respectively corresponding to. The scan compressed image data 18 has a luminance Y (Y1, Y1) in the MCU unit of the minimum data unit.
Y2, Y3, Y4) components, and color difference Cb, Cr components, followed by an interleave system. These # 1 to #N scan compressed image data 18 are DC components → A
The order is C low frequency component → AC high frequency component.

FIGS. 11 (a), (b), (c) and FIG.
Explains the procedure for generating such compressed image data 18. In this case as well, the original image data shown in FIG.
If 0 × 240 dots, the number of MCUs is 30 × 15 = 4
Assuming that the MCU number is 50, the MCU number is k, and the scan number is i, in the flow shown in FIG.
In step 1202, i = 1, and k = 1. In step 1203, the luminance Y (Y) of the image data for one MCU for the bandwidth and the color component specified in the i-th scan.
1, Y2, Y3, Y4) and the color differences Cb, Cr are sequentially encoded, and in step 1204, while k is incremented (while k + 1), k <450 in step 1205.
Until the result is negative, the same operation is repeated.
When k <450 is denied in 05, i is determined in step 1206.
Is incremented (while i + 1), the same operation is repeated in step 1207 until i <N is denied, so that the number of scans N is divided for each MCU unit, and the luminance Y corresponding to each MCU unit is divided. Component, color difference Cb
, Cr are generated.

The file format of the compressed image data based on the non-interleave described in FIG.
The flow shown in FIG. 13 is executed in order to convert to the progressive built-up JPEG format compressed image data file format described in FIG.

Also in this case, if the original image data is 480 × 240 dots as shown in FIG. 11 (a), 8 × 8 dots are set as small blocks, and a unit block consisting of 2 × 2 small blocks is used as a unit block. As one screen, 30 unit blocks are arranged in the horizontal direction and 15 unit blocks are arranged in the vertical direction.

Assuming that the vertical numbers of these unit blocks are w, first, in step 1301, w = 1, and in step 1302, the DC component → AC low frequency component → AC high frequency component in the format shown in FIG. Consisting of #
The memory Si for storing the scan compressed image data 18 of 1 to #N is secured, and the pointer Psi for each memory Si is set to 0.

Next, in step 1303, for the first block (w = 1), Huffman decoding is performed for 16 lines of the luminance Y component, and the DCT coefficients are stored in the memory. The pointer PY of the luminance Y component is set, that is, the pointer PY of the data to be read next is advanced. Next, Huffman decoding is also performed on the eight lines of the color difference Cb and Cr components (which have been thinned out in advance to 1/2), and the DCT coefficient values are stored in the memory. , C
The pointers PCb and PCr of the r component are set, that is, the pointers PCb and PCr of the data to be read next are advanced.

Then, the process proceeds to step 1305, where the decoded DCT coefficient values of the luminance Y and chrominance Cb, Cr components are processed for each of the data handled in the scans # 1 to #N, that is, the DC component → AC low frequency component → Divide for each AC high frequency component and store it in the memory in ascending order of scan number.

Next, in step 1306, the values of the DCT coefficients corresponding to the scans # 1 to #N are Huffman-coded,
These are stored as scan compressed image data 18 in each memory S
i [Psi] (where i = 1, 2,... N). In the pointer Psi to the memory Si, the offset to be stored next is inserted.

Then, at step 1037, w is incremented (while w + 1), while step 130 is executed.
In step 8, the operation from step 1303 is repeated until w ≦ 15 is denied. Then, in step 1308, w
If .ltoreq.15 is denied, the processing for one screen is terminated, and the flow advances to step 1309 to convert the scan compressed image data 18 in the memory Si corresponding to each of the scans # 1 to #N into # 1 to # shown in FIG. Save after N scan headers 17,
Finish the conversion of progressive build-up to JPEG compressed data file format.

Accordingly, in order to convert the compressed image data of the file format based on the non-interleave shown in FIG. 2 into the compressed data file format of the progressive built-up JPEG format shown in FIG. Until now, DCT coefficient data was divided into three regions by the spectrum selection method and encoded, and each image was saved in the memory to expand one screen, and further compressed by the progressive method. Here, DC obtained by Huffman decoding each component of luminance Y and color difference Cb, Cr
The T coefficient is divided into regions in the order of DC component → AC low frequency component → AC high frequency component in correspondence with scans # 1 to #N,
Since the DCT coefficient values of these areas are Huffman-coded and saved for each scan header, compared to the above-described one requiring expansion of one screen and compression by the progressive method, the format conversion is required. In addition to reducing the capacity of the memory used for the data conversion, the processing time for format conversion can be greatly reduced. (Third Embodiment) In the third embodiment, a case is shown in which a compressed image data file based on non-interleaving of a captured image input by a digital camera or the like is directly displayed on a display screen.

FIG. 14 shows a schematic configuration of such a display block. In the figure, 21 is a digital camera,
This digital camera 21 is provided with a C
It is connected to PU23. The CPU 23 has a printer 25 via a driver 24 and a C
RT27 are connected respectively.

The storage unit 28, the program ROM 29, the work memory 30, and the output memory 31 are connected to the CPU 23. Here, the digital camera 21
Is for outputting a captured image as compressed image data in a file format based on the non-interleave shown in FIG. The storage unit 28 stores a compressed image data file from the digital camera 21. The output memory 31 stores the decoded data displayed on the CRT 27. And CP
U23 converts the compressed image data in the file format stored in the storage unit 28 into display data by the control program of the program ROM 29.

The image captured by the digital camera 21 is stored as compressed image data in a file format based on the non-interleave shown in FIG.
Is stored once. From this state, the flow shown in FIG. 15 is executed.

Also in this case, the original image data is 480 × 240
Assuming dots, 8 × 8 dots are small blocks in the same manner as described above, and a collection of four 2 × 2 small blocks is a unit block (for 16 lines). The arrangement is taken as a large block and the whole is divided into 15 blocks. Assuming that the numbers of these large blocks are w, first, in step 1501, w =
In step 1502, the luminance Y component, the color difference Cb,
The respective decode pointers PY, PCb,
Initialize PCr. Then, in step 1503, for the first block (w = 1), the decode pointer PY
, PCb, and PCr, and decodes 16 lines of the luminance Y component and 8 lines of the chrominance Cb and Cr components (which have been thinned out in advance to 1/2), and stores them in the output memory 31. Store.

Next, in step 1504, the components of the luminance Y and the color differences Cb and Cr stored in the output
After performing GB conversion, this is converted into image data for 16 lines by C
Partial display on RT27. In this case, the RGB converted 1
The image data for six lines may be temporarily stored and then displayed on the CRT 27.

Then, in step 1505, the decode pointers PY, PCb, PCr are updated, and in step 1506
Then, while incrementing w (while w + 1),
Until w ≦ 15 is denied in step 1507, the operations in and after step 1503 are repeated to display one screen.

Therefore, if the compressed image data of the file format based on the non-interleave shown in FIG. 2 is to be decoded and displayed as it is, a dedicated memory of about 230 K is required as a display memory. Here, the luminance Y, the color differences Cb, C
The decode pointers PY, PCb, and PCr are prepared for each color component of r, and the data is displayed on the CRT 27 while being converted into display data in block units in accordance with the instructions of the decode pointers PY, PCb, and PCr. Compared with the necessity of a memory, the capacity of the display memory can be reduced, and the time required for the conversion process into the display data can be significantly reduced, so that a quick image display can be realized.

In the present invention, a sequential non-interleaved JPEG format file can be displayed on a screen in a progressive manner. In this case, first, JP
The EG file is Huffman-decoded in block units, and the resulting DCT coefficient matrix is divided into a DC component region, a low-frequency AC component region, and a high-frequency AC component region, and stored in regions 1, 2, and 3. The operation is performed for one screen.
Is subjected to IDCT conversion, the three components of luminance and chrominance are stored in a memory, and RGB conversion is performed to be displayed on a CRT. Subsequently, the area 2 is subjected to IDCT conversion, and the luminance and color difference components are stored in a memory.
Similarly, the image is displayed on the CRT, and the same processing is performed on the area 3 at last, and the CRT display enables the progressive screen display.

In this way, the DCT coefficient is divided into three regions in the file of the sequential JPEG format, and expansion of the DC component → AC low frequency component → AC high frequency component and color component is added to these regions. As a result, progressive screen display can be realized.

[0052]

As described above, according to the present invention, the capacity of the memory used for format conversion can be reduced, and the processing time for format conversion can be greatly reduced.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of a format conversion apparatus applied to a compressed image data format conversion method according to a first embodiment of the present invention.

FIG. 2 is a view showing a file format structure of compressed image data based on non-interleaving according to the first embodiment;

FIG. 3 is a diagram showing each component of luminance and color difference of compressed image data based on non-interleaving according to the first embodiment;

FIG. 4 is a view showing a method for generating compressed image data based on non-interleaving according to the first embodiment;

FIG. 5 is a diagram showing a file format structure of JPEG format compressed image data according to the first embodiment.

FIG. 6 is a diagram showing a data structure in units of MCU of JPEG format compressed image data according to the first embodiment;

FIG. 7 is a diagram illustrating luminance and color difference components of JPEG format compressed image data according to the first embodiment;

FIG. 8 is a view showing a method of generating JPEG format compressed image data according to the first embodiment;

FIG. 9 is a flowchart illustrating a format conversion method according to the first embodiment;

FIG. 10 is a diagram showing a file format structure of progressive built-up JPEG format reduced image data according to the second embodiment of the present invention.

FIG. 11 is a diagram showing each component of luminance and color difference of JPEG format reduced image data of progressive build-up according to the second embodiment.

FIG. 12 is a view showing a method of generating JPEG format reduced image data of progressive build-up according to the second embodiment;

FIG. 13 is a flowchart illustrating a format conversion method according to the second embodiment;

FIG. 14 is a diagram showing a schematic configuration of a third embodiment of the present invention.

FIG. 15 is a flowchart illustrating a format conversion method according to a third embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Control part, 2 ... Input part, 3 ... Output part, 4 ... ROM, 5 ... RAM, 11 ... Header, 12 ... Quantization table, 13 ... Compressed image data, 14 ... Header including a quantization table, 15 ... Compressed image data, 16: Various definitions, 17: Scan header of # 1 to #N, 18: Scanned compressed image data of # 1 to #N, 21: Digital camera, 22, 24, 26: Driver, 23: CPU Reference numeral 25: printer, 27: CRT, 28: storage unit, 29: program ROM, 30: work memory, 31: output memory

Claims (8)

[Claims] 1. An image data encoded for each block is configured in a predetermined format for each predetermined unit, and compressed image data configured in the predetermined format is configured by another unit different from the predetermined unit. A method of converting the compressed image data format into another format, the method comprising: obtaining an address for each block of the predetermined unit; and configuring the block into the other unit based on the obtained address. A method for converting a compressed image data format. 2. The compressed image data according to claim 1, wherein the step of obtaining the address includes a step of obtaining a code length of the block, and a step of obtaining an address by adding the code length. Format conversion method. 3. The step of configuring the other unit includes: converting the block of the encoded image data into a plurality of frequency components; and converting the converted image data based on the determined address. 2. A method of converting a compressed image data format according to claim 1, further comprising the step of configuring another unit for each frequency component. 4. The step of obtaining an address includes the step of obtaining an address of compressed image data for each block of one block line, and the step of configuring the other unit is based on the obtained address. 2. The method according to claim 1, further comprising the step of sequentially converting the compressed image data for one block line into RGB signals line by line. 5. The image data encoded for each block is configured in a predetermined format for each predetermined unit, and the compressed image data configured in the predetermined format is configured by another unit different from the predetermined unit. Means for obtaining an address for each block of the predetermined unit, and converting the block to the other unit based on the address obtained by the means for obtaining the address. A compressed image data format conversion device, comprising: conversion means. 6. The means for obtaining an address comprises: a block code length detecting means for obtaining a code length of the block; and an adding means for adding an code length obtained by the block code length detecting means to obtain an address. 6. The compressed image data format conversion device according to claim 5, further comprising: 7. The frequency component conversion means for converting the encoded image data block into a plurality of frequency components, the frequency component conversion based on an address obtained by the address obtaining means. 6. The apparatus according to claim 5, further comprising means for converting the image data converted by the means into another unit for each frequency component. 8. The means for obtaining an address includes means for obtaining an address of compressed image data for each block corresponding to one block line, and the converting means based on the address obtained by the means for obtaining the address. 6. The compressed image data format converter according to claim 5, further comprising means for sequentially converting the compressed image data for one block line into RGB signals for each line.