JPH10271498A - Compression data conversion method, information quantity conversion method for image compression data, and image recorder adopting the method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To record image furthermore by reserving an idle capacity in a record medium even when the record medium is fully occupied. SOLUTION: Image compression data via quantization and variable length coding are recorded in a record medium 40, and compression degree distribution information that integrates quantization coefficients at compression for each macro block is recorded in the record medium 40. In the case of re-compression processing, coded data in the record medium 40 are subject to variable length decoding and inverse quantization. Then the resulting data is again quantized by using a re-quantization coefficient and subject to variable length coding and the resulting data are returned to the record medium 40. The requantization coefficient is decided at a re-compression control section 46 based on a relative magnitude of the quantization coefficient obtained from the compression degree distribution information. An idle capacity of the record medium is extended while keeping the entire image quality level as high as possible by deciding the re-quantization coefficient so that a data reduction quantity is increased as a macro block has a relatively smaller quantization coefficient.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a compressed data conversion method for converting compressed data and changing the information amount (code amount) of the data. INDUSTRIAL APPLICABILITY The present invention can be applied to, for example, conversion of compressed image data, and can be particularly suitably applied to an image recording apparatus that records image information on a recording medium.

[0002]

2. Description of the Related Art As represented by the international encoding standard MPEG, a technology for compressing and encoding data such as a moving image and storing and providing the data in various storage media is being developed. Today, various manufacturers are focusing on the development of multimedia-related devices, and are trying to introduce MPEG-compliant products to the market.

FIG. 5 shows a configuration of a conventional general MPEG encoding apparatus 2. The DCT unit 4 performs a Discrete Cosine Transform on the input video, and expands the image into discrete spatial frequency components. In the quantization unit 6, each frequency component of the data after DCT is divided by the quantization coefficient, and the remainder is rounded. The quantization coefficient is sent from the quantization control unit 10 of the encoding control unit 8 to the quantization unit 6. Hereinafter, the data after quantization is referred to as quantized data. The quantized data is sent to a variable length coding unit (VLC) 12, and is subjected to variable length coding using a stochastic method so that the code amount becomes as small as possible. The encoded data is output to generate a bit stream. For example, MPEG2
, The bit stream is a sequence layer, GOP
(Group of Pictures) layer, picture layer, slice layer, macroblock layer, block layer, and the like.

The inverse quantization unit 14, the IDCT unit 16, and the motion compensation unit 18 are configured to perform predictive coding to convert a part of the video input into a differential video and increase the compression ratio. The inverse quantization unit 14 performs an inverse quantization process, which is a reverse process of the quantization, on the quantized data output from the quantization unit 6. ID
The CT unit 16 further performs an inverse DCT, which is an inverse process of the DCT, on the data after the inverse quantization process. Motion compensation unit 1
Reference numeral 8 includes a reference frame memory and a block matching unit, and performs a motion compensation process using the video almost restored by the inverse quantization and the inverse DCT. The motion compensator 18 sends a video signal for obtaining a difference to the differentiator 20 and converts the motion vector obtained by the motion compensation process into VL
Output to C section 12. The differentiator 20 is controlled by the encoding control unit 8 and operates only when an input video to be subtracted is input. Then, the input image and the motion compensation unit 18
The difference between the images sent from the
Send to The motion vector is used for generating a bit stream in the VLC unit 12. With these configurations, encoding using forward prediction, bidirectional prediction, and the like defined in MPEG is performed.

[0005] As the quantization coefficient used for quantization in the quantization section 6 is large, the code amount of data output from the VLC section 12 is small, and the degree of compression of the coded data is large. Utilizing this, the encoding control unit 8 controls the bit rate of the bit stream. Information indicating the code amount of the encoded data is sent from the VLC unit 12 to the encoding control unit 8. The quantization control unit 10 changes the quantization coefficient to be sent to the quantization unit 6 in macroblock units according to the code amount sent from the VLC unit 12. Bit rate control includes fixed control and variable control. In the case of the fixed rate control, the quantization coefficient is changed in macroblock units so that the code amount for each GOP is constant. In the case of variable rate control, for example, if the bit rate is 2 Mbps to 4 M
bps range and the average rate is 3Mbps
Is controlled so that

FIG. 6 shows a configuration of a moving picture recording apparatus equipped with an encoding / decoding device in which a decoding device is integrated with the encoding device of FIG. The imaging device 22 has a CCD, an A / D converter, and the like, and converts an image of a subject into a digital electronic signal. This digital electronic signal is input to the encoding / decoding device 24 as an input video. Encoding / decoding device 2
In step 4, as described above, a bit stream is generated by DCT, quantization, predictive coding using motion compensation, and variable-length coding, and the bit stream is sent to the recording medium 26 and recorded.

The moving picture recording apparatus shown in FIG. 6 has a display device 28 for displaying a reproduced image. At the time of reproduction, a bit stream is read from the recording medium 26,
It is input to a variable length decoding unit (VLD unit) 30. The data decoded by the VLD unit 30 is inversely quantized by the inverse quantization unit 14 and subjected to inverse DCT by the IDCT 16. The motion compensation unit 18 performs a process of returning the difference video data to a state before the difference is obtained. For these processes,
The bit stream includes data indicating a quantization coefficient, a motion vector at the time of predictive encoding, and the like. The image signal thus restored is sent to the display device 28 as an output video and displayed.

[0008]

SUMMARY OF THE INVENTION In a moving image recording apparatus,
Recording time is limited according to the capacity of the recording medium. In a still image recording device such as an electronic still camera, the number of images that can be recorded is limited. However, the user
Even when the recording medium becomes full, there are cases where further image recording is desired. In order to respond to such a request, the recorded image data is recompressed to reduce the data amount (hereinafter, such a process is referred to as a recompression process).
It is effective to return to the recording medium and secure the free space of the recording medium. Although the image quality of the recorded image is degraded by the recompression process, further image recording becomes possible.

Techniques for recompressing image data include, for example,
It is disclosed in Japanese Patent Application Laid-Open No. 7-312756. In the information amount conversion circuit of the publication, the bit stream is returned to the inverse quantization state. Then, quantization is performed again using a new quantization coefficient different from the first quantization coefficient, and further variable-length encoding is performed. By setting the new quantization coefficient small, the bit rate is reduced. Although the image quality of the final image data is lower than that before conversion, the code amount is reduced.

The apparatus disclosed in the above publication checks pixel data of a block after dequantization, and examines its frequency distribution.
If the value of the term of the high-frequency component is larger than the upper limit predetermined value, the quantization coefficient at the time of requantization is set relatively large.
On the other hand, if the value of the high frequency component is smaller than the lower limit predetermined value, the quantization coefficient is set to be smaller. Utilizing the property that the deterioration of the image quality is less noticeable even when the quantization coefficient is increased as the value of the high-frequency component is larger, the deterioration of the image quality at the time of recompression is suppressed. However, the device disclosed in the publication can focus on only one block at a time and determine only the quantization coefficient at the time of recompression for that block.

On the other hand, the present invention focuses on the fact that a series of image data includes a portion having a high image quality when the image is reproduced and a portion having a low image quality. For example, a part where a fast-moving image or a fine image is recorded is a part where the image quality at the time of reproduction is relatively largely lower than the original image, and is a part where the image quality is lower than the average image quality level. This part of the data
Compared to other parts, it is better not to reduce much. An object of the present invention is to provide a compressed data conversion method capable of reducing the amount of data while keeping the overall image quality level of image data in a sequence as high as possible in consideration of the existence of such image quality. To provide.

In addition, the apparatus disclosed in the above publication needs to provide a high-frequency information amount check circuit for checking pixel data after inverse quantization in order to determine which block can have a higher compression degree. 8x for each block
Since it is necessary to check the values of the high-frequency terms of the eight matrices, the circuit configuration becomes complicated accordingly. As the circuit becomes more complicated, the processing time for recompression also increases. Another object of the present invention is to provide a compressed data conversion method that can appropriately set the degree of compression when recompressing image data with a simple configuration.

A further object of the present invention is to provide an image recording apparatus to which a compressed data conversion method capable of achieving the above-mentioned object is applied, whereby an empty capacity can be ensured even when a recording medium is full, and user needs It is to respond to.

[0014] Such a situation applies not only to an image recording apparatus but also to an apparatus for recording other compressed data such as audio. When the capacity of the recording medium becomes full, the free space can be increased by recompressing the recorded data. Therefore, the present invention is not limited to image data,
An object of the present invention is to provide a compressed data conversion method capable of performing suitable recompression processing on other compressed data.

[0015]

[Means for Solving the Problems]

(1) The compressed data conversion method of the present invention is a method of converting compressed data so that the amount of information changes, based on the relative magnitude relationship of the degree of compression for each predetermined unit of data constituting the compressed data. The compression degree for recompression of the predetermined unit data is determined, and the predetermined unit data is converted according to the determined compression degree for recompression. Here, the compressed data includes moving image data, still image data, and compressed data such as audio data. The moving image data and the still image data include data compressed by various methods in addition to MPEG and JPEG. The data conversion processing according to the degree of compression for recompression includes (a) processing for expanding the data once and then compressing again, (b) processing for further compressing the data without expansion, and (c). Includes the process of decompressing data halfway and then compressing it again.

According to the present invention, the compression degree for recompression is determined based on the relative magnitude relation of the compression degree for each predetermined unit data. The degree of compression of the predetermined unit data is correlated with the resilience (image quality for an image) of the predetermined unit data. Therefore, the relative degree of the compression degree is determined, and the compression degree for recompression is determined so that the data amount of the predetermined unit data that has little influence on the overall resilience is preferentially reduced. As the difference between the degrees of compression before and after recompression is increased, the data amount can be significantly reduced. Thus, according to the present invention,
By performing the recompression processing based on the relative magnitude of the degree of compression, it is possible to appropriately adjust the restorability of each part of the recompressed compressed data. Then, the data amount can be reduced while keeping the overall level of the resilience of the compressed data as high as possible. Here, the overall level means the overall level of all compressed data and the overall level of data having a certain size in all compressed data. Shall be considered. In the following, the same applies to the case of the overall level of image data.

For example, when it is considered that the predetermined unit data having a higher compression degree has already been greatly compressed and the restoring property when the data is expanded and reproduced is degraded, the predetermined unit data having a relatively higher compression degree is considered. For the unit data, the same compression degree as before compression is adopted as the compression degree for recompression.
For the predetermined unit data having a relatively small compression degree, a larger compression degree than before the recompression is adopted as the compression degree for recompression.

(2) The present invention also relates to an information amount conversion method for converting the information amount of compressed image data, wherein a compression degree parameter indicating the degree of compression for each predetermined unit of data constituting the compressed image data is summarized. It is stored as compression degree distribution information, and the recompression degree of the predetermined unit data is determined based on the correlation between the relative size of the compression degree indicated in the compression degree distribution information and the image quality at the time of image reproduction. Is converted into data compressed with the recompression degree. The predetermined unit data is, for example, a macroblock, slice,
Or a picture.

In the present invention, the size of the high frequency component of the pixel data of each block is not checked unlike the prior art.
The compression degree for recompression is determined based on a criterion completely different from the conventional one, that is, the relative magnitude of the compression degree of the predetermined unit data. It is considered that the predetermined unit data having a relatively small compression degree has a sufficiently high image quality at the time of reproduction, and the image quality is higher than the average level. This part
Even if the code amount is significantly reduced during recompression, the influence on the average image quality is small. On the other hand, data with a relatively high compression degree is considered to have an image quality during reproduction lower than the average level.
Based on such a correlation between the degree of compression and the image quality, for example, the degree of recompression is determined so as to increase the data reduction amount of the predetermined unit data having a relatively small degree of compression. By performing the re-compression processing based on the relative magnitude of the degree of compression, the image quality of each predetermined unit data constituting the image data can be appropriately adjusted. Therefore, the data amount can be reduced while maintaining the overall level of the image quality of the series of image data as high as possible.

The compression degree parameter indicates the degree of compression of the predetermined unit data, and is, for example, a quantization coefficient at the time of quantization processing. The larger the quantization coefficient, the higher the degree of compression. The quantization coefficient is data originally included in header information of a bit stream in MPEG or the like. It is preferable to store the quantized coefficients at locations other than the bitstream as compression degree distribution information. By looking at the compression degree distribution information, the quantization coefficient, which is a parameter for determining the recompression degree, can be immediately known. Unlike the related art, it is not necessary to check the value of the high-frequency term, which is the content of the pixel data, so that the circuit configuration is not complicated, and the processing time is correspondingly short.

In the present invention, for example, it is preferable to set the compression degree for recompression as follows. That is, when the current compression degree of a certain predetermined unit data is relatively large, the compression degree for recompression is made substantially the same as the current compression degree. When the current compression degree of the predetermined unit data is relatively small, the compression degree for recompression is made larger than the current compression degree. As a result, the overall image quality level is substantially maintained before and after recompression, and the total code amount is reduced.

(3) Further, according to the present invention, in an image recording apparatus for compressing image information and recording it as compressed image data on a recording medium, a compression degree parameter indicating a degree of compression for each predetermined unit of image compression data is summarized. Compression degree distribution storage means for storing as compression degree distribution information, and recompression degree determining means for determining the degree of recompression of predetermined unit data according to the relative magnitude of the compression degree indicated in the compression degree distribution information; Data conversion means for reading compressed image data from the recording medium, converting predetermined unit data into data compressed with the recompression degree, and returning the data to the recording medium.

The recording medium is, for example, a flash memory,
It is a hard disk, an optical disk, or the like, and is a medium from which data can be read and written by being accessed by an electrical, magnetic, optical method, or the like.

According to the above configuration, the amount of compressed image data recorded on the recording medium is reduced, so that the free space of the recording medium is increased. For example, even when the capacity of the recording medium becomes full, an empty capacity can be generated according to a request of a user or the like, and further image recording can be performed. Then, at the time of recompression for reducing the data amount, the same effects as those of the above-described compressed data conversion method and the information amount conversion method of image compressed data can be obtained.

Also in this embodiment, as described above, it is preferable to set the compression degree for recompression as follows. When the current compression degree of certain predetermined unit data is relatively large, the compression degree for recompression is made substantially the same as the current compression degree. When the current compression degree of the predetermined unit data is relatively small, the compression degree for recompression is made larger than the current compression degree. As a result, the overall image quality level is substantially maintained before and after the recompression, the total code amount is reduced, and the free space of the recording medium is secured. This is the same in the following embodiments.

(4) Further, the present invention provides an encoding means including a quantization section for compressing and encoding image information into image compression data, a storage means for storing image compression data, and an image including an inverse quantization section. And a decoding means for decompressing and decoding the compressed data. In the image recording apparatus, a quantized coefficient distribution storage for collectively storing quantized coefficients used for quantization for each predetermined unit data so as to know the quantized coefficient distribution information. Means, re-quantization coefficient determination means for determining a re-quantization coefficient of predetermined unit data according to the relative size of the quantization coefficient indicated in the quantization coefficient distribution, and reading the image compression data from the recording medium. Data conversion control means for causing the decoding means and the coding means to convert predetermined unit data into data compressed using the requantization coefficient, and to return the data to the recording medium. Here, the quantization coefficient distribution storage means is:
It may be a partial area of the recording medium.

According to the above arrangement, the encoding means and the decoding means provided in the image recording apparatus can be used as a constitution for recompression processing. Therefore, it is possible to add an empty capacity increasing function to the image recording apparatus at a low cost by using the existing configuration. Then, at the time of recompression for reducing the data amount, the same effects as those of the above-described compressed data conversion method and the information amount conversion method of image compressed data can be obtained.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention (hereinafter, referred to as embodiments) will be described below with reference to the drawings.

[Embodiment 1] FIG. 1 is a block diagram showing an overall configuration of an apparatus (bit rate conversion apparatus) for implementing a method of converting compressed data and a method of converting an amount of compressed image data according to the present invention. This embodiment is based on MPEG,
In FIG. 1, the same members as those in FIGS. 5 and 6 described above are denoted by the same reference numerals, and description thereof will be omitted.

The recording medium 40 is composed of a flash memory, and stores a bit stream storage section 42 for recording a bit stream of compressed image data, and a quantization coefficient Q for each macro block of compressed image data as compression degree distribution information. And a compression degree distribution storage unit 44, which is an area to be compressed. FIG. 2 partially shows a bit stream and compression degree distribution information.

As described above, the bit stream has a hierarchical structure such as a sequence. The bit stream in FIG. 2 is an overview of the macroblock layer. The data of each macroblock includes pixel data of four blocks that constitute the macroblock, and also includes motion vector and data of a quantization coefficient (quantization step value) Q as header information.

On the other hand, the compression degree distribution information includes the data of the quantization coefficient Q of each macroblock used when generating the bit stream. The quantization coefficient Q in the compression degree distribution information is stored together with information indicating which macroblock in the bit stream the quantization coefficient Q corresponds to.

In this embodiment, the compression degree distribution storage unit 44
, The quantization coefficient Q of each macroblock is stored in the order of the macroblock, and this data string is used as compression degree distribution information. However, the compression degree distribution information may be any information as long as the relative magnitude relation between the quantization coefficients for each macroblock can be understood, and the format is not limited to this embodiment. For example, the quantization coefficient may be stored only for the changing point of the quantization coefficient Q in the bit stream.

The recompression control section 46 controls each component of the present apparatus to perform recompression processing of image data.
A recompression quantization coefficient setting unit 48 is provided. The coefficient setting unit 48 operates based on the above-mentioned compression degree distribution information.
That is, the coefficient setting unit 48 obtains the relative size of the quantization coefficient Q of each macroblock. This makes it possible to determine whether the quantized coefficient Q of each macroblock when compressed is larger or smaller than the other macroblocks in the bit stream sequence. The coefficient setting unit 48 calculates, based on the relative size of the quantization coefficient Q,
The requantization coefficient Q1 of each macroblock is determined according to the following criteria. The requantization coefficient Q1 is used in a quantization processing step when recompressing image data.

(1) A macroblock having a relatively small quantization coefficient Q; when an input video is encoded using the encoder as shown in FIG. , The generated code amount is small, the quantization coefficient is set to be small, and the compression degree tends to be low. Therefore, it is considered that a portion having a small quantization coefficient has sufficient image quality, and has an image quality higher than the average image quality of the entire sequence. Even if the code amount of this part is largely reduced, the influence on the average image quality is small. Therefore, for a macroblock having a relatively small quantization coefficient Q, the difference (Q1-Q) between the requantization coefficient Q1 and the quantization coefficient Q is increased in order to greatly reduce the code amount during recompression. , The requantization coefficient Q1 is set.

(2) A macroblock having a relatively large quantization coefficient Q; on the other hand, when an input image is coded, a portion having a fast motion or a portion having a fine image has a large amount of generated code and a large quantization coefficient. It tends to be set high and the degree of compression already high. Therefore, the portion where the quantization coefficient is large is
It is considered that the image quality is lower than the average image quality of the entire sequence. Therefore, for a macroblock having a relatively small quantization coefficient Q, the code amount is not reduced much during recompression. Specifically, the re-quantization coefficient Q1 is set so that the difference (Q1-Q) between the re-quantization coefficient Q1 and the quantization coefficient Q becomes small. Alternatively, the requantization coefficient Q1 is made equal to the quantization coefficient Q.

More specifically, the requantization coefficient Q1 is given by Q1 =
It is represented as Q + α + β. α is a constant value for all macroblocks. β changes according to the relative size of the quantization coefficient Q. For a macroblock having a relatively large quantization coefficient Q, β is reduced. Note that the size of α may be changed in accordance with a user's instruction.

Next, the operation of the information amount conversion device of this embodiment will be described. When an instruction to start recompression is input from an input unit (not shown), the recompression control unit 46 reads out the compression degree distribution information from the compression degree distribution storage unit 44 of the recording medium 40.
Then, the recompression quantization coefficient setting unit 48 determines the requantization coefficient Q1 of each macroblock in the sequence according to the above criteria (1) and (2). The determined requantization coefficient Q1 is sent to the quantization unit 6 via the quantization control unit 10.

The recompression control section 46 outputs a control signal to a read / write control section (not shown), whereby the encoded data is read from the bit stream storage section 42 of the recording medium 40 to the VLD section 30. The VLD unit 30 performs variable length decoding on the encoded data and sends the decoded data to the inverse quantization unit 14. The data sent to the inverse quantization unit 14 includes a quantization coefficient Q.
On the other hand, data that is not changed by the quantization processing, such as header information and motion vectors other than the quantization coefficients, is sent from the VLD unit 30 to the VLC unit 12.

The inverse quantization section 14 inversely quantizes the data of each block using the quantization coefficient Q, and outputs the result to the quantization section 6. The requantization coefficient Q1 corresponding to the data transmitted from the inverse quantization unit 14 is sent from the recompression control unit 46 to the quantization unit 6 via the quantization control unit 10 at the same timing. The quantization unit 6 quantizes the data using the requantization coefficient Q1 and outputs the data to the VLC unit 12. The VLC unit 12
The quantized data sent from the quantization unit 14 is subjected to variable-length encoding, combined with the requantized coefficient Q1 and the data sent from the VLD unit 30 to output encoded data. VLC unit 1
The output of 2 is written to the bit stream storage unit 42 of the recording medium 40 via a read / write control unit (not shown). At this time, in the bit stream storage unit 42, the image data after recompression is overwritten on the image data before recompression. Thus, the data amount of the image data in the bit stream storage unit 42 is reduced, and the bit rate is reduced.

Further, the quantization section 6 stores the requantized coefficient Q1 sent from the recompression quantized coefficient setting section 48 in the recording medium 4.
0 is stored in the compression degree distribution storage unit 44. This allows
The compression degree distribution information in the compression degree distribution storage unit 44 also corresponds to the image data after the recompression processing.

As described above, in the present embodiment, the distribution of the quantization coefficient which is a parameter indicating the degree of compression of each part in the data sequence is stored. From the relative magnitude of the compression of a macroblock, the appropriate requantization factor for that macroblock is known. For example, in a macroblock, the original image is fine, the amount of code generated for the macroblock is large, and it is highly likely that the image quality has dropped during the previous compression. Is better. Based on such a correlation between the quantization coefficient and the image quality, the data amount of a portion where the quantization coefficient Q is relatively small (that is, a portion that is considered to have little effect on the overall image quality even if the data amount is reduced). Are re-quantized so that is preferentially reduced. Therefore, at the time of recompression, a decrease in the overall level of image quality in the sequence can be suppressed. This makes it possible to minimize degradation in image quality while minimizing
The data amount of the image data once recorded can be reduced.

The requantization coefficient Q1 of this embodiment may be set as follows, for example. That is, when the current quantization coefficient Q of a certain macroblock is relatively large, the requantization coefficient Q1 of the macroblock is set to be the same as the quantization coefficient Q. When the current quantization coefficient Q of a certain macroblock is relatively small, the requantization coefficient Q1 of the macroblock is made larger than the current quantization coefficient Q. As a result, data is reduced only in a portion where the quantization coefficient Q is relatively small. The image quality of the portion where the quantization coefficient Q is relatively large does not change before and after recompression. With this setting, the overall image quality level is substantially maintained before and after recompression, and the total code amount is reduced.
Further, for a macroblock in which the requantization coefficient Q1 is equal to the current quantization coefficient Q, the inverse quantization and the quantization processing may be omitted. As a result, the processing speed can be increased.

In this embodiment, the recording unit of the compression degree for obtaining the compression degree distribution is a macroblock. On the other hand, the compression degree may be recorded for each different unit such as each slice or each picture. An appropriate compression degree parameter may be recorded according to the recording unit. However, the smaller the compression degree recording unit, the finer the control of the code amount, but the larger the data amount of the compression degree distribution information.

When encoding the original input video, bit rate control is performed as follows. For example, in bit rate control on a macroblock basis, the code amount of coded data of a certain macroblock is sent to a quantization control unit,
Based on the code amount, a quantization coefficient for the next macroblock is determined. Therefore, when it is determined that the code amount of a certain macroblock is large, the quantization coefficient of the next macroblock is reduced. This means that a quantization coefficient suitable for quantization of a certain macroblock is used for quantization of the next macroblock. In view of this point, the following modifications can be made to the present embodiment.

That is, the requantization coefficient Q1 of a certain macroblock is determined based on the relative size of the quantization coefficient Q of the next macroblock. Alternatively, when determining the requantization coefficient Q1 of a certain macroblock, the quantization coefficient Q of the macroblock and the next macroblock is referred to. As a result, a quantization coefficient originally appropriate for the macroblock can be used in the recompression processing.

[Second Embodiment] FIG. 3 shows the overall configuration of the second embodiment. The second embodiment is a modification of the first embodiment. In FIG. 3, the same reference numerals are given to the same members as those in FIG. 1, and the description will be omitted.

In the second embodiment, the quantization control unit 10 is eliminated, and the quantization unit 6 and the inverse quantization unit 14 are eliminated. After determining the re-quantization coefficient Q1, the re-compression control unit 50 according to the second embodiment further obtains a transform coefficient Q1 / Q that is a ratio between the re-quantization coefficient Q1 and the quantization coefficient Q.

A data conversion section 52 is provided between the VLD section 30 and the VLC section 12. The data conversion section 52 receives the conversion coefficients Q1 / Q from the recompression control section 50. The data conversion unit 52 multiplies the data input from the VLD unit 30 by a conversion coefficient Q1 / Q,
Send to 2.

Next, the operation of the apparatus shown in FIG. 3 will be described. Here, differences from the first embodiment will be mainly described. The coded data is read from the recording medium 40 to the VLD 30 under the control of the recompression control unit 50. The VLD unit 30 performs variable-length decoding on the encoded data and sends it to the data conversion unit 52. Data that does not need to be changed, such as a motion vector, is sent from the VLD unit 30 to the VLC unit 12.

The re-compression quantization coefficient setting section 48 determines the re-quantization coefficient Q1 for each macroblock based on the compression degree distribution information, as in the first embodiment. Then, the recompression control unit 50 obtains a transform coefficient Q1 / Q for each macroblock using the quantization coefficient Q and the requantization coefficient Q1. When the data of the macroblock corresponding to the conversion coefficient Q1 / Q is input from the VLD unit 30 to the data conversion unit 52, the conversion coefficient Q1 / Q is adjusted to match the timing.
2 is output.

The data conversion section 52 multiplies the data sent from the VLD section 30 by a conversion coefficient Q1 / Q, and
Send to The re-quantization coefficient Q1 is sent from the re-compression control unit 50 to the VLC unit 12. The VLC unit 12 performs variable-length coding on the data sent from the data conversion unit 52, and outputs the data together with the requantization coefficient Q1 and the motion vector as coded data. The output of the VLC unit 12 is written to the bit stream storage unit 42 of the recording medium 40. In addition, the recompression control unit 50 sends the requantization coefficient Q1 of each macroblock to the recording medium 40. Thereby, the compression degree distribution information after recompression is stored in the compression degree distribution storage unit 44.

In the recompression processing of the present embodiment, the variable-length decoded data is converted using the conversion coefficients Q1 / Q, and is again subjected to variable-length coding. Also in this process, data similar to that of the first embodiment is obtained. Since a quantization unit and an inverse quantization unit are not required, the circuit configuration is simplified.

[Third Embodiment] In the third embodiment, the information amount converter shown in FIG. 1 is incorporated in an image recording apparatus. FIG.
Shows the configuration of the image recording apparatus of the third embodiment. In FIG. 4, the same members as those in FIG. 1 and the same members as those in the image recording apparatus in FIG.

The encoding / decoding device 24a of FIG. 4 has substantially the same configuration as the device of FIG. 6, except that the output side of the inverse quantization unit 14 is connected to the input of the quantization unit 6. It is connected to the side. Therefore, the data inversely quantized by the inverse quantization unit 14 can be sent to the quantization unit 6. The recompression control unit 46a is substantially the same as the recompression control unit 46 of the first embodiment, and is a component of the encoding / decoding device 24a and controls each configuration shown in FIG. And has a recompression quantization coefficient setting unit 48 similar to that of the first embodiment.

In FIG. 4, the recording data amount monitoring unit 54
Monitors the amount of data recorded in the bit stream storage unit 42 of the recording medium 40. The recording data amount monitoring unit 52
When the recording data amount reaches a predetermined value close to the capacity of the bit stream storage unit 42, the fact is displayed on the display unit 56. The display unit 56 may be integrated with the display 28.

The re-compression control unit 46a includes the user operation unit 5
8 are connected. By operating the user operation unit 58, the user can input an instruction to start recompression processing, a target of a data reduction amount by recompression processing, and the like. The goal is, for example,
A, B, and C are ranked. When rank A is selected, the data reduction amount is set larger than rank C.

Next, the operation of the image recording apparatus of FIG. 4 will be described. Here, a description will be given focusing on differences from the first embodiment and the image recording apparatus of FIG.

[Shooting and Reproduction] At the time of shooting, the image of the subject is converted into a digital electronic signal by the imaging device 22 as in the prior art, and the encoding / decoding device 24a
Is input to In the encoding / decoding device 24a, as described above, a bit stream is generated by DCT, quantization, predictive encoding using motion compensation, and variable-length encoding, and this bit stream is sent to the recording medium 40. Recorded.

The quantization control unit 10 of the third embodiment sends the same quantization coefficient Q to the compression unit 6 via a read / write control unit (not shown) in addition to sending the quantization coefficient Q to the quantization unit 6. Write to the degree distribution storage unit 44. Thereby, the compression degree distribution information corresponding to the image data recorded in the bit stream storage section 42 is recorded in the compression degree distribution storage section 44.

The processing at the time of reproduction is the same as that of the conventional image recording apparatus shown in FIG. 6, and a description thereof will be omitted.

As the “recompressed” image data is recorded,
The free space of the bit stream storage unit 42 decreases. The recording data amount monitoring unit 54 indicates that the free space falls below a predetermined value.
It is detected by. Then, the display unit 56 displays that the bit stream storage unit 42 is full. The user who sees this display operates the user operation unit 58 to instruct the start of the recompression process. At this time, the user inputs a target of the data reduction amount in the recompression processing. Even when the display 56 does not indicate that the capacity is full, the user can appropriately instruct the execution of the recompression processing for the purpose of effectively using the capacity of the medium.

A signal indicating an instruction for the recompression processing is sent from the user operation section 58 to the recompression processing section 46a. Subsequent recompression processing is the same as in the first embodiment. The recompression control unit 46a reads out the compression degree distribution and determines a requantization coefficient Q1 for each macroblock. Requantization coefficient Q1
Is sent to the quantization unit 6 via the quantization control unit 10 of the encoding control unit 8. On the other hand, under the control of the recompression processing control unit 46a, the encoded data of the recording medium 40 is read out, and V
The variable-length decoding is performed by the LD unit 30, further dequantized by the inverse quantization unit 14, and sent to the quantization unit 6. In the quantization unit 6, the inversely quantized data is re-quantized using the requantization coefficient Q 1 and sent to the VLC unit 12. In the VLC unit 12, the quantized data is subjected to variable-length encoding, and the encoded data is overwritten and recorded in the bitstream storage unit 42. The quantization control unit 10 also sends the requantization coefficient Q1 to the recording medium 40. Thereby, the compression degree distribution storage unit 4
The compression degree distribution information of No. 4 also corresponds to the image data after the recompression processing.

As described above, according to the present embodiment, when the capacity of the image recording apparatus is full, it is possible to secure an empty capacity and record an image further. The user can execute this even when it is necessary to record an image more than originally planned. In particular, in the present embodiment, similar to the first embodiment, it is possible to suppress a decrease in the overall level of image quality in a sequence during recompression. Therefore, the free space of the recording medium 40 can be secured while making the deterioration of the image quality inconspicuous. Further, in the present embodiment, the existing quantization unit, inverse quantization unit, VLC unit, VLD unit, and coding control unit are used as members constituting the information amount conversion device. Therefore, a recompression processing function can be added to the image recording apparatus at low cost.

As described in the first embodiment,
Also in the third embodiment, it is preferable to set the requantization coefficient Q1 as follows. That is, when the current quantization coefficient Q of a certain macroblock is relatively large, the requantization coefficient Q1 of the macroblock is set to be the same as the quantization coefficient Q. The current quantization coefficient Q of a certain macroblock
Is relatively small, the requantization coefficient Q1 of the macroblock is made larger than the quantization coefficient Q. As a result, the overall image quality level is substantially maintained before and after recompression, and the total code amount is reduced.
Free space is secured. Further, for a macroblock in which the requantization coefficient Q1 is equal to the current quantization coefficient Q, the inverse quantization and the quantization processing are omitted, and the overall processing speed can be increased.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an overall configuration of a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing recording data on a recording medium of the apparatus of FIG.

FIG. 3 is a block diagram illustrating an overall configuration of a second embodiment of the present invention.

FIG. 4 is a block diagram illustrating an overall configuration of a third embodiment of the present invention.

FIG. 5 is a block diagram illustrating a configuration of a conventional MPEG encoding device.

FIG. 6 is a block diagram illustrating a configuration of an image recording device including a conventional MPEG encoding / decoding device.

[Explanation of symbols]

4 DCT section, 6 quantization section, 8 coding control section, 10
Quantization control unit, 12 VLC unit, 14 inverse quantization unit, 1
6 IDCT section, 18 motion compensation section, 26, 40 recording medium, 30 VLD section, 42 bit stream storage section, 44 compression degree distribution storage section, 46, 50 recompression control section, 48 quantization coefficient setting section for recompression, 52 data conversion unit, 54 recording data amount monitoring unit, 58 user operation unit.

Claims (5)

[Claims] 1. A method for converting compressed data so that its information amount is changed, comprising a method for recompressing predetermined unit data based on a relative magnitude relationship of compression degrees of predetermined unit data constituting the compressed data. A compression data conversion method comprising: determining a degree of compression of a predetermined unit of data; and converting predetermined unit data according to the determined degree of compression for recompression. 2. An information amount conversion method for converting the amount of information of compressed image data, comprising: combining compression degree parameters so as to know the compression degree of each predetermined unit data constituting the image compression data; The recompression degree of the predetermined unit data is determined based on the correlation between the relative size of the compression degree obtained from the compression degree distribution information and the image quality at the time of image reproduction, and the predetermined unit data is recompressed. A method of converting the amount of information of compressed image data, wherein the method converts the information into data compressed in degrees. 3. The method according to claim 2, wherein the image compression data has been subjected to a compression process including a quantization process, and the compression degree parameter is a quantization coefficient at the time of the quantization process. A method for converting the amount of information of compressed image data. 4. An image recording apparatus for compressing image information and recording the same as compressed image data on a recording medium, wherein a compression degree parameter is collected so that a compression degree for each predetermined unit of image compression data can be known. Compression degree distribution storing means for storing as information, recompression degree determining means for determining a recompression degree of predetermined unit data according to a relative magnitude of the compression degree obtained from the compression degree distribution information, An image recording apparatus, comprising: data reading means for reading compressed image data, converting predetermined unit data into data compressed at the recompression degree, and returning the data to the recording medium. 5. An encoding unit including a quantization unit, which compresses and encodes image information into image compression data, a storage unit for storing the image compression data, and includes an inverse quantization unit to decompress the image compression data. In an image recording apparatus including a decoding unit, a quantization coefficient distribution storage unit that collectively stores quantization coefficients used for quantization for each predetermined unit data so as to know the quantization coefficient distribution information, and a quantization coefficient Requantization coefficient determining means for determining a requantization coefficient of predetermined unit data according to the relative size of the quantization coefficient obtained from the distribution information; and Means and a data conversion control means for causing the encoding means to convert predetermined unit data into data compressed by using the requantization coefficient and returning the data to the recording medium. That the image recording device.