JP2005065110A - File recording method, file recorder, computer program and computer readable recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To record the management information of media data such as moving image data and audio data in a file including MP4 by a method with little overhead and capable of maintaining compatibility with a conventional standard. <P>SOLUTION: File recording for shortening a time required for merge processing, with little risk of loss of the management information and excellent in compatibility with the conventional standard is performed by constituting a recording part 10 which executes the basic processing of a file output part 5 which multiplexes sample data and the management information and outputs them to an output file 8, a management information storage part 6 which temporarily stores the management information until it is outputted to the output file 8 and a data structure reallocation part 7 which transfers a data block recorded in the output file 8 to the other position in the file. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a file recording method, a file recording apparatus, a computer program, and a computer-readable recording medium. More specifically, the present invention is used to collectively output a data structure including management information of moving image data and audio data. And suitable.

  In recent years, apparatuses for digitizing and recording moving images have become widespread. For example, digital cameras that can digitally encode captured moving images using a video / audio encoding format with high compression efficiency and record them in an external storage medium such as a built-in memory or an SD card have become widespread. In addition, mobile phones that can exchange moving image data recorded in the same way via an IP network by e-mail are becoming widespread.

  For this video / audio encoding format, MPEG-4 (ISO / IEC 14496), which is being standardized mainly by ISO (International Organization for Standardization), is generally used. Compared with MPEG-2 used for DVD, MPEG-4 has the property that it can encode video and audio data with high compression efficiency while suppressing degradation of quality even in a low bandwidth environment. . Because of these characteristics, many of the recording devices for general consumers as described above employ MPEG-4.

  In the above-described device, generally, moving image data and audio data are multiplexed and finally output as a file. A typical example of the file format is the MPEG-4 File Format standard (ISO / IEC 14496-14, hereinafter referred to as “MP4”) standardized as part of the MPEG-4 standard. Hereinafter, the basic structure of the MP4 file will be briefly described.

FIG. 1 is a diagram showing a simplified structure of the MP4 file format.
In the MP4 file, all data is packed and recorded in a data structure called “Atom” (or “Box”). The header information common to all atoms having the following definition is added to the head of each atom.
class Atom {
unsigned int (32) size;
byte type [4];
byte data [];
};

  The “size” is a field indicating the data length of the atom. “Type” is a field indicating the type of atom, and a four-character identifier assigned to each atom is set. “Data” is a substance of data included in the atom, and the content differs depending on the type. Since atoms can have a nested structure and can include one or more other atoms, “data” can be another atom.

  The atom structure of the MP4 file can be roughly divided into two as shown in FIG. Media data such as video and audio is composed of a series of very small encoded data (hereinafter referred to as “samples”) divided in units such as durations and frames. Management information such as size and time stamp is also multiplexed and recorded.

  In the MP4 standard, the former is called “Movie Atom” (the value of the type field is “moov”) and is defined as an atom for storing so-called metadata information such as management information of moving image / audio samples. . The latter is defined as an atom called “Media Data Atom” (the value of the type field is “mdat”) for storing so-called media data such as moving image / audio sample data.

  Media data of Media Data Atom is stored in a state of being logically divided into data blocks called “chunks”. Each chunk is composed of temporally continuous sample data, and an array of chunks is stored in the data field of Media Data Atom. The size of the chunk and the number of included samples are not particularly limited, and a chunk having an arbitrary size and number of samples can be configured and stored according to the environment and situation.

  In MP4, it is not permitted to describe an atom physically divided into a plurality of parts. Therefore, Movie Atom must be recorded in a file as a single atom in which all management information to be recorded is stored. The same applies to the above “Media Data Atom”.

  It is not possible to know how long the recording time of a moving image shot by the above apparatus will be at the end of recording. Since the size of Movie Atom and Media Data Atom increases as the recording time becomes longer, MP4 can be arranged in the arrangement shown in FIG. 1 without dividing the atom unless the final size of these atoms is determined when recording is completed. It cannot be recorded to a file.

  Therefore, in order to output the MP4 file as shown in Fig. 1, management information stored in Movie Atom and sample data stored in Media Data Atom are stored or recorded separately, and finally the two are merged. It was necessary to perform the process of.

  As described above, when the output is performed in the format of FIG. 1, there is a problem that the MP4 file cannot be output unless the recording of all the moving image / sound sample data is correctly terminated. For example, if the recording process ends abnormally due to unforeseen circumstances such as power supply interruption or equipment malfunction during shooting, the data recorded up to that point may be erased or stored in an illegal state on the device or storage medium. It will remain. The remaining data cannot be reproduced because it has an illegal format for MP4. Therefore, if the process ends abnormally, substantially all the recorded contents up to that point are lost.

  In order to prevent the above problem, it is necessary to periodically merge the sample data and the management information and leave an MP4 file in which the recorded contents up to that point are temporarily stored. However, since the merge process involves input / output of all accumulated sample data and management information, the time required for one merge process increases in proportion to the increase in the amount of data.

  In addition, in order to perform the merge processing, there must be an area for holding the output MP4 file and a temporary storage area for holding sample data to be merged and management information, which occupy the same size. I must. That is, in order to perform the merging process, the recording apparatus is required to provide a storage device or storage medium having a size approximately twice as large as the data size to be actually recorded.

  In order to obtain a necessary and sufficient temporary storage area, it is necessary to install a larger number of storage devices or storage media in the recording device, or to reduce the upper limit of the time that can be recorded by the recording device. There will be some cost in terms of performance.

  As a measure that can solve the above problem, the MP4 standard defines an atom called “Data Reference Atom” (type field value is 'dref') for referring to sample data of other files as shown in FIG. ing. If the sample data and management information are output as separate files using Data Reference Atom, the merge process is not required.

  Also, in the ISO Base Media File Format standard (ISO / IEC 14496-12) that is the basis of the MP4 standard, management information called “Movie Fragment Atom” (type field value is 'moof') is divided into multiple atoms. Atom is defined to describe as

  The mechanism using Movie Fragment Atom is currently being standardized in the direction to be introduced into the MP4 standard and the Motion JPEG 2000 standard (ISO / IEC 15444-3), which is also derived from the ISO Base Media File Format standard. In addition, a data recording method in which the basic principle is applied to other data formats has also been proposed (for example, see Patent Document 1).

  An example of an MP4 file when Movie Fragment Atom is used is shown in FIG. The management information is also recorded in the Movie Fragment Atom arranged behind the Movie Atom, and the contents stored in the Movie Atom and all the Movie Fragment Atoms are handled as one logical metadata area.

  Using this mechanism, the contents accumulated in the temporary storage are periodically merged, written at the end of the MP4 file in a format as shown in FIG. 3, and the temporary storage is released. It is possible to shorten the time required for the merge processing of one time, and to minimize the risk of losing management information.

JP 2003-22621 A

However, each of the above solutions has problems.
First, files recorded using the Data Reference Atom method are not guaranteed to be played correctly unless they are all in a complete state without being lost. Therefore, it seems that there are not a few users who want to output as a single MP4 file so that they can be exchanged easily and safely by mail or file transfer. In order for such users to output in the format they wanted, eventually all data had to be re-merged, which did not solve the problem.

  In addition, since Movie Fragment Atom has not yet been incorporated into the MP4 standard at the time of writing the description of the present application, it is guaranteed that the function of Movie Fragment Atom is supported by an already shipped MP4 processing device. There is no. Probably, when playing back an MP4 file containing Movie Fragment Atom on a non-supporting device, it is assumed that at least the portion of the data indicated by the management information recorded in the first Movie Atom is played back, and thereafter it cannot be played back. For this reason, there is a risk that sufficient compatibility with a shipped device cannot be maintained.

  According to the file recording method of the present invention, sample data composed of encoded data of moving images or audio, chunk data composed of a plurality of the sample data, and management information for managing the sample data and chunk data are recorded in a file. A file recording method, a reserved area generating process for generating a reserved area for collectively recording the management information in the file, and a data rearrangement for rearranging the chunk data at a different position in the file And the size of the reserved area is changed by rearranging the chunk data at a different position in the file.

  The file recording apparatus of the present invention records sample data composed of encoded data of moving images or sounds, chunk data composed of a plurality of the sample data, and management information for managing the sample data and chunk data in a file. A file recording apparatus, a reserved area generating means for generating a reserved area for collectively recording the management information in the file, and data rearrangement for rearranging the chunk data at different positions in the file Means for changing the size of the reserved area by rearranging the chunk data at a different position in the file.

  The computer program of the present invention is a file that records sample data consisting of encoded data of moving images or audio, chunk data consisting of a plurality of the sample data, and management information for managing the sample data and chunk data in a file A recording method, a reserved area generating process for generating a reserved area for collectively recording the management information in the file, and a data rearranging process for rearranging the chunk data at different positions in the file In the data rearrangement processing, the computer is caused to execute processing for changing the size of the reserved area.

The computer-readable recording medium of the present invention is characterized by recording the computer program described above.

  As described above, according to the present invention, outputting management information to a file in a format that is compatible with the conventional standard can be realized with a small overhead.

  Embodiments of a file recording method, a file recording apparatus, a computer program, and a computer-readable recording medium according to the present invention will be described below with reference to the drawings. The present invention records moving image data management information without performing merge processing by rearranging the data structure in the file as necessary.

  In the embodiment of the present invention, the MP4 file format is used as the final output format of the recorded moving image. However, the object to which the present invention can be applied is not necessarily limited to the MP4 file, and it goes without saying that the present invention can also be applied to a file format having a similar structure to the MP4 file, such as the QuickTime format.

<First Embodiment>
FIG. 4 is a diagram showing a basic configuration example of the moving image recording apparatus of the present invention.
4, the moving image recording apparatus according to the present embodiment includes a moving image input unit 1 that captures a moving image to be recorded, an audio input unit 2 that also captures sound, and compression encoding processing of the captured moving image and sound. The video encoding unit 3 and the audio encoding unit 4 for generating management information such as sample data in digital encoding format and the size and time stamp of each sample, and multiplexing the sample data and management information to output file 8 The file output unit 5 that outputs data, the management information storage unit 6 that temporarily stores the management information until the management information is output to the output file 8, and the data block recorded in the output file 8 are stored in the file. The data structure rearrangement unit 7 is configured to move to another position. The file output unit 5, the management information storage unit 6, and the data structure rearrangement unit 7 constitute a recording unit 10 that executes basic processing in the present embodiment.

  The moving image and sound input to the moving image input unit 1 and the sound input unit 2 are input from an input device such as a camera or a microphone, an external recording / storage device, or a medium via some transmission means. . In FIG. 4, descriptions of these devices and media are omitted. The output file 8 to be output is held by any recording / recording device or medium.

  The parts not included in the recording unit 10, that is, the moving image input unit 1, the audio input unit 2, the moving image encoding unit 3, and the audio encoding unit 4 may have different configurations depending on the embodiment. For example, in FIG. 4, one moving image and one audio input system are provided, but any number of input systems may exist. That is, there may be a plurality of moving image / sound channel input systems for the recording unit 10 or only one of the moving image or sound input systems.

  In addition, pre-encoded data may be directly input to the recording unit 10. In this case, the moving image input unit 1, the audio input unit 2, the moving image encoding unit 3, and the audio encoding unit 4 are not necessary. In addition, as long as the intention of the present invention is correctly realized, the logical / physical arrangement and mounting form of each component may be any.

Next, a typical process flow performed in the recording apparatus of FIG. 4 will be described.
First, the recording apparatus takes in a moving image from the moving image input unit 1. Then, the captured video signal of the moving image is sequentially transferred to the moving image encoding unit 3. Similarly, the voice is taken in from the voice input unit 2, and the voice signal is passed to the voice encoding unit 4.

  The moving image encoding unit 3 encodes the input video signal in a predetermined compression encoding format such as MPEG-4 and outputs moving image sample data. Similarly, in the case of audio, the audio signal input by the audio encoding unit 4 is encoded, and audio sample data is output.

The sample data obtained by the encoding process is written to the output file 8 as chunk data included in the Media Data Atom via the file output unit 5.
The file output unit 5 outputs Media Data Atom from a position slightly away from the beginning of the file, and thereafter records additional chunk data at the end of the file. The area from the beginning of the file to the start position of Media Data Atom is used as a reserved area for outputting Movie Atom that stores management information, as shown in FIG.

  Even if the file output unit 5 immediately records the sample data in the output file 8 at the time of receiving the sample data, the file output unit 5 temporarily accumulates the passed sample data and collects it as a chunk when a certain amount of sample data is accumulated. It may be recorded. A plurality of sample data or chunk data may be collectively delivered to the file output unit 5.

  In the moving image encoding unit 3 and the audio encoding unit 4, management information such as sample size and time stamp obtained by the encoding process is also generated and transferred to the file output unit 5. The transferred management information is further transferred to the management information storage unit 6.

  The management information storage unit 6 refers to and updates the management information of the data structure written in the output file 8 and converts it to the Movie Atom format if necessary, and requests the MP4 file output unit 5 to output it. It has a function. The management information may be stored by some temporary storage unit depending on the implementation. This temporary storage will generally be realized as a relatively large capacity storage means such as an intermediate file or a RAM file.

  Further, the management information storage unit 6 holds the data size of the management information written in the Movie Atom format. This data size is recalculated each time management information data is transferred to the management information storage unit 6 and updated to the latest state.

  When the sample output is repeated and the end of the management information exceeds the size of the reserved area at the beginning of the file, the management information storage unit 6 requests the data structure rearrangement unit 7 to rearrange the output chunk data. .

  The data structure rearrangement unit 7 receives the position and size of the chunk stored as management information from the management information storage unit 6, and rearranges the first chunk data at the end of the file. At the same time, the management information of the rearranged chunk is updated according to the position after the rearrangement. The area with the rearranged chunk can be used as a reserved area reserved for Movie Atom. That is, by executing the rearrangement process, the reserved area at the beginning of the file is expanded.

  The data structure rearrangement unit 7 repeats the above rearrangement process until a reserved area large enough to output the management information accumulated in the management information storage unit 6 is obtained. When the reserved area becomes sufficiently large, the rearrangement process is terminated, and the control is returned to the management information storage unit 6.

  The data structure rearrangement unit 7 also rearranges or changes data blocks other than the chunk data so that the file format does not contradict the standard due to the chunk rearrangement, and also performs a process of maintaining the file structure correctly.

  By performing the above processing in the recording apparatus of FIG. 4, a state is realized in which a reserved area having a sufficient size for recording management information is secured at the head of the file. Since this data area is followed by Media Data Atom, an MP4 file as shown in FIG. 1 can be created simply by outputting the stored management information to the reserved area in the Movie Atom format.

  The reserved area that is initially reserved can be arbitrarily sized in principle, but in practice it may be better to have a certain size in order to reduce the number of relocations. However, if the size is too large, a useless area is generated in the MP4 file and the size of the MP4 file increases. Therefore, it is desirable to set an appropriate size according to the form to be applied.

  It is also possible to set the size of the free area to 0, that is, not to reserve the free area in the initial state. However, in this case, chunk rearrangement occurs when Movie Atom size calculation is performed for the first time, and therefore, processing that is substantially equivalent to the case of allocating a free area first is performed.

<Metadata structure of MP4>
Subsequently, in describing the details of the present embodiment, how the management information of samples and chunks is held in the MP4 file format will be briefly described. It should be noted that unnecessary portions are omitted in this specification to explain the contents of the invention, and therefore the contents shown in the description and the drawings do not completely reflect the actual structure of MP4. .

FIG. 5 is a diagram for explaining what data structure the metadata has in the MP4 file and how to refer to the media data.
In FIG. 5, sample S1 indicates a sample, and C1 indicates a chunk including the sample S1. Since metadata information about all samples and chunks is recorded in Movie Atom, information on samples S1 and C1 can be obtained by accessing the contents of Movie Atom.

  In the Movie Atom, an atom “Track Atom” (the value of the type field is “trak”) for storing metadata information of the data track in the MP4 file is stored. Each data track is identified by a unique identifier in an MP4 file called “track ID”, and Track Atom describes information on the data track such as a track ID. Also, since all chunks and samples belong to one of the data tracks, the metadata information of all chunks and samples that constitute each data track is also described in Track Atom. Yes.

  Atoms that store metadata information about chunks and samples in Track Atom are "Sample Size Atom" (type field value is 'stsz'), "Sample to Chunk Atom" (type field value is 'stsc') , “Chunk Offset Atom” (type field value is 'stco'), etc.

  Sample Size Atom manages the size of the sample (entry-size), Sample to Chunk Atom manages information about the chunk containing the sample, and Chunk Offset Atom manages the offset position in the chunk file.

  In order to obtain the data of sample S1 from Media Data Atom, first, check the sample number indicating which sample S1 is the first sample from the beginning, and sample from the entry of Sample Size Atom corresponding to the sample number Get the sample size of S1.

  Then, using the sample number, the number of the chunk containing the sample S1 is acquired from Sample to Chunk Atom, and further, the start offset position of the chunk indicated by the chunk number is acquired from Chunk Offset Atom. At this time, if the sample S1 is not the first sample of the chunk C1, the size of each sample existing between the samples S1 from the beginning of the chunk is obtained from the Sample Size Atom, and the sum of the sizes is calculated. A position obtained by adding the total size to the offset position of C1 becomes the start offset of the sample S1. Using the start offset and size of the sample S1 obtained as described above, the sample data of the sample S1 can be extracted from the Media Data Atom.

  Next, a specific example of the sample data acquisition process will be described with reference to FIG. As an example, assume that the sample S1 in FIG. 5 is the first sample (sample number = 1) of the data track with track ID = 2.

(1) Search for Track Atom with track ID = 2, and obtain the first entry E1 corresponding to sample number 1 from Sample Size Atom included therein. In the case of the sample S1, the value 0x2801 of the entry E1 is the sample size. If all samples have the same size, Sample Size Atom stores only one entry for the default sample size. In that case, the size of the sample S1 is the default sample size.

(2) The chunk number recorded as the first-chunk of the first entry E2 corresponding to sample number = 1 is obtained from Sample to Chunk Atom.
In the Sample to Chunk Atom entry, if chunks with the same entry contents are consecutive, the first chunk number of consecutive chunks is recorded in first-chunk, and the attributes of the second and subsequent chunks are not recorded. Omitted.

  That is, in FIG. 6, information of (4-1 = 2) chunks and (4 * (4-1) = 8) samples are implicitly recorded between the entry E2 and the next entry. It will be. Therefore, if the entry corresponding to the sample number does not exist in Sample to Chunk Atom, the chunk number of the omitted entry must be obtained. In this example, the value 1 of the first-chunk of the entry E2 corresponding to the sample number = 1 is the chunk number of the chunk C1 including the sample S1.

(3) The offset position of the chunk is acquired from the first entry E3 of Sample Size Atom corresponding to the chunk number = 1. In this example, 0x05a2a is the offset position of chunk C1. A value obtained by adding the total size of the samples existing between the acquisition target samples from the beginning of the chunk to the offset position becomes the offset position of the acquisition target sample. Since the sample S1 is the first sample of the chunk, the offset position of the sample S1 is (0x05a2a + 0 = 0x05a2a). By the above processing, it is derived that the data of the sample S1 is recorded between the offset position 0x05a2a and 0x2801 bytes.

<Relocation of chunks>
In searching for metadata information about samples and chunks, the numbers indicating the order of their appearance are very important, so the MP4 atom structure entries are basically the same as the order of appearance of the chunks and samples. Must be recorded in chronological order.

  In addition, in an atom such as Sample to Chunk Atom where the contents of an entry depend on the value of the previous entry, changing the value of an existing entry affects the subsequent entries, so the entry value is changed. I can't do it.

  However, in the case of Chunk Offset Atom, the entry has no dependency relationship with other entries in the atom, so changing the contents of each entry does not affect other entries. The contents of the Chunk Offset Atom entry are absolute positions in the file, and are not necessarily arranged in ascending order.

  Therefore, it is possible to set the position of the chunk indicated by the entry to an arbitrary place simply by changing the value of the Chunk Offset Atom entry. For example, in the case of FIG. 6, in order to change the offset position of the chunk C1, the value of the entry E3 of Chunk Offset Atom may be changed. In the present embodiment, chunk rearrangement processing is performed using the characteristics of the above Chunk Offset Atom.

<Relocation processing flow>
Hereinafter, a specific procedure of the chunk rearrangement process performed in the present embodiment will be described with reference to the flowchart of FIG.

(Step S71) When a recording operation is started in the recording apparatus of FIG. 4, first, the output file 8 is created in the initial state by the file creation unit 5.

  FIG. 8 is a diagram showing the configuration of the output file 8 created by the file creation unit 5. The file creation unit 5 creates an output file 8 in a state where a free area of an arbitrary size is set in advance as a reserved area for writing Movie Atom before Media Data Atom. The free space can be set by arranging an atom indicating a free space called “Free Space Atom” (the value of the type field is “free” or “skip”) in the MP4 standard. Hereinafter, the leading offset position of this empty area is represented as Fp.

  In FIG. 8, the first offset position of Movie Atom is represented as Mp, and the size of Movie Atom is represented as Ms (= Fp−Mp). In the initial state, Fp is at the same position as Mp, but each time Movie Atom is written to the output file 8, the position of Fp is pushed down to the position of (Fp + Ms).

  Media Data Atom is placed immediately after the free space, and thereafter, chunk data is written sequentially at the end of the file. In FIG. 8, the head offset of Media Data Atom is represented as Dp. That is, the size of the available free space is (Dp−Fp).

  In FIG. 8, the start offset positions of the N chunks stored in the Media Data Atom are Cp [1], Cp [2]... Cp [n], and the sizes are Cs [1], Cs [2]. ... Cs [n]. (Cp [1] −Dp) is the byte length of the header of Media Data Atom, that is, (sizeof (size) + sizeof (type) = 8). The offset position and size shown in FIG. 8 are initialized by the management information storage unit 6 and managed thereafter until the recording operation ends.

(Step S <b> 72) Upon receiving the sample data, the file output unit 5 outputs the sample data to the end of the output file 8. The output may be performed each time a sample is received, or a plurality of samples may be output together.

(Step S73) The management information storage unit 6 receives the management information of the sample passed in step S72 from the file output unit 5, and stores it internally. At the same time, the size when all the management information including the received management information is converted to the Movie Atom format is calculated, and the calculated size is stored as Ms.

(Step S74) The management information storage unit 6 checks whether or not the end offset position of Movie Atom indicated by Fp exceeds the start offset position Dp of Media Data Atom.
When Movie Atom overlaps with Media Data Atom, that is, when (Fp> Dp), the content of Media Data Atom is overwritten with the content of Movie Atom. move on. In the case of (Fp <= Dp), chunk rearrangement is unnecessary, and the process returns to step S72.

(Step S75) The management information storage unit 6 requests the data structure rearrangement unit 7 to rearrange chunks. The data structure rearrangement unit 7 acquires the offset position Cp [1] and the size Cs [1] of the chunk recorded at the head of the Media Data Atom from the management information stored in the management information storage unit 6, and Cp Data of the first chunk from [1] to Cp [2] is copied to the end of the file.

  At this time, the data structure rearrangement unit 7 first expands the area of the Media Data Atom by adding the size Cs [1] of the chunk to be copied to the value of the size field of the Media Data Atom, and then the first chunk at the end of the file. Add data.

  The management information storage unit 6 may store the position of the leading chunk in order to efficiently perform the Cp [1] acquisition process. The offset position of the first chunk is the end position of the file in the initial state, and is updated when relocation occurs thereafter.

(Step S76) When the data structure rearrangement unit 7 finishes copying the data of the first chunk, the Chunk Offset managed by the management information storage unit 6 so that the offset position of the copied chunk becomes the copy destination position. Update the Atom entry. When the entry is updated, the chunk data at the copy source position is not referenced from anywhere.

(Step S77) The data structure rearrangement unit 7 acquires the offset position Cp [2] of the second chunk from the management information storage unit 6, and copies the header of Media Data Atom immediately before Cp [2].

  When copying the header of Media Data Atom, the data structure rearrangement unit 7 updates the value of the size field so that the atom size is equal to the total size of Cs [2] to Cs [n].

  There are two Media Data Atom headers at the time of copying completion, but the newly copied header will be referenced for the first time when the processing of the subsequent step S78 is completed. Does not occur.

  However, if the size of the chunk to be rearranged is smaller than the Media Data Atom header, that is, if it is 7 bytes or less, the content of the existing Media Data Atom header is destroyed by the copy processing in step S77, and the step Until the processing of S78 is completed, the Media Data Atom cannot be referred to temporarily. Therefore, in the present embodiment, it is desirable to record the chunk as a data block of at least 8 bytes or more.

(Step S78) The data structure rearrangement unit 7 obtains the Fp from the management information storage unit 6 and the head offset position (Cp [2] -8) of the Media Data Atom copied in step S77, and frees the area between the two. Make it an area.

The data structure rearrangement unit 7 extends the free space by updating the value of the size field of the Free Space Atom existing at the position of Fp to (Fp + (Cp [2] −8)).
When the expansion of Free Space Atom is completed, the value of Dp stored in the management information storage unit 6 is updated to the head offset position (Cp [2] -8) of Media Data Atom.

  Thereafter, the process returns to step S74, and the rearrangement process is repeatedly performed until the contents of Movie Atom are within the reserved area. The chunk offset expressions Cp [2] to Cp [n + 1] before returning to step S74 are processed as Cp [1] to Cp [n] after returning to step S74. Chunk rearrangement is executed by the above steps, and the output file 8 always maintains a state where an area where management information can be sufficiently recorded is secured.

  In the flowchart of FIG. 7, the timing at which the management information is output to the output file 8 is not specified, but it is actually possible to write the management information at an arbitrary timing after step S72. If the management information is output periodically, an MP4 file that can be correctly played back at that time can be obtained.

  According to the method described above, when management information is output to a file, it is only necessary to output management information data without performing media data merging processing, so that the time required for file output can be shortened.

  In general MP4 files, media data is often larger in size than management information, so if merge processing is not required, the amount of data to be output and the time required for output may be less than half. I can expect.

  When chunk rearrangement occurs, media data re-input / output processing occurs, but the data size at that time is limited to the total amount of chunks to be processed, so re-entry compared to conventional merge processing. The output media data size can be small. If a sufficient reserved area is secured in advance, chunk rearrangement may not occur, and therefore, media data may not be re-input / output at all.

  Further, according to the above method, since the media data is directly output to the MP4 file, the temporary storage area only needs to have a capacity capable of holding the management information. As a result, it is possible to obtain an effect of reducing the cost by reducing the temporary storage installed in the apparatus or increasing the performance by increasing the upper limit value of the recordable time.

<Second Embodiment>
As a second embodiment, a process for performing rearrangement for returning the last chunk to the reserved area will be described.
In the first embodiment, if the size of the reserved area that is initially secured is too large, a large amount of unused area remains at the beginning of the file when recording is completed. As a result, the size of the MP4 file is reduced. It will increase unnecessarily. In the second embodiment, in order to make effective use of the file area, processing for reusing the remaining reserved area as the media data area by rearranging the last chunk after completion of recording will be described. To do.

  Depending on the environment, there is a function that expands / reduces the file size by specifying the end position of the file.In the execution environment of this embodiment, this function is used when recording is completed. It is possible to reduce the file size.

<Relocation processing flow>
The forward chunk rearrangement process executed after the recording of the media data is completed will be described below using the flowchart of FIG. In addition, symbols such as those shown in FIG. 8 are used for expressions such as offset positions and sizes used in the description, as in the first embodiment.

(Step S91) When the recording of the media data is completed in the recording apparatus of FIG. 4, the management information storage unit 6 determines the size Fs of the current free area and the size Cs [ n] and check whether the chunk can be rearranged.

As a result of this check,
In the case of (Fs> = Cs [n]), the process proceeds to step S92, and a process of moving the last chunk forward is performed. Otherwise, the chunk rearrangement process is unnecessary, and the process ends at that point.

(Step S92) The management information storage unit 6 requests the data structure rearrangement unit 7 to rearrange chunks. First, the data structure rearrangement unit 7 obtains the start position Dp of Media Data Atom and the size Cs [n] of the chunk n to be rearranged from the management information storage unit 6, and after the rearrangement of the header part of the Media Data Atom Copy to the offset position (Dp-Cs [n]).

  At this time, the data structure rearrangement unit 7 performs copying in a state where Cs [n] is added to the value of the size field of the Media Data Atom to be copied. That is, the Media Data Atom immediately after copying contains a chunk n to be rearranged.

  When copying of the header part of Media Data Atom is completed, the value of Dp stored in the management information storage unit 6 is updated to the head offset position (Dp-Cs [n]) of Media Data Atom. As described in step S77 of the first embodiment, in the case of (Cs [n] <8), the existing Media Data Atom header is overwritten, so the chunk size is 8 bytes. The above is desirable.

(Step S93) The data structure rearrangement unit 7 acquires Fp and Fs from the management information storage unit 6, and updates the value of the size field of the Free Space Atom existing at the position of Fp to (Fp−Dp). To reduce the size of the free space. At the same time, the value of Fs stored in the management information storage unit 6 is updated to (Fp−Dp).

(Step S94) Based on the offset position Cp [n] and the size Cs [n] acquired in step S92, the data structure rearrangement unit 7 stores the last chunk n of the Media Data Atom copied in step S92. Copy to the position (Dp + 8) immediately after the header.

(Step S95) When the data structure rearrangement unit 7 finishes copying the data of the end chunk n, it changes the offset position recorded in the entry corresponding to the end chunk n in the Chunk Offset Atom to the copy destination position. To do. When the entry is updated, the chunk data at the copy source position is not referenced from anywhere.

(Step S96) The data structure rearrangement unit 7 sets again a value obtained by subtracting the size Cs [n] of the rearranged chunk n from the value of the size field of the Media Data Atom copied in step S92. As a result, the data of chunk n before relocation is excluded from Media Data Atom, and the contents of the size field return to the state before relocation processing is performed.

(Step S97) The data structure rearrangement unit 7 uses the file size change function provided by the execution environment to change the file size to the offset position Cp [n] before the rearrangement of the chunk n. As a result, the file size is reduced by Cs [n] compared to before the rearrangement.

  Thereafter, the process returns to step S91, and the rearrangement process is repeatedly performed until the free area becomes sufficiently small. Note that the chunk offset expressions Cp [1] to Cp [n−1] before returning to step S91 are Cp [2] to Cp [n] after returning to step S91, and Cp [n] is Cp [n]. 1].

  According to the method described above, the size of the output file that has become unnecessarily large in the first embodiment can be reduced to a realistic size.

<Other embodiments>
The present invention can be applied to a system constituted by a plurality of devices (for example, a host computer, an interface device, a writer, an encoder, etc.) or to a device (for example, a video camera device, etc.) comprising a single device. Good.

  Another object of the present invention is to supply a storage medium (or recording medium) in which a program code of software that realizes the functions of the above-described embodiments is recorded to a system or apparatus, and the computer (or CPU or CPU) of the system or apparatus. Needless to say, this can also be achieved by the MPU) 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. It goes without saying that a case where the function of the above-described embodiment is realized by performing part or all of the actual processing and the processing is included.

  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 expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the 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.

  When the present invention is applied to the storage medium, the storage medium stores program codes corresponding to the flowcharts described above.

It is a figure which simplifies and shows the structure of MP4 file format. It is a figure which shows the structure of MP4 file using Data Reference Atom. It is a figure which shows the structure of the MP4 file using Movie Fragment Atom. It is a figure which shows the structural example of the moving image recording device in embodiment of this invention. It is a figure which shows an example of the correlation with the metadata structure and media data in MP4 file. It is a figure which shows an example of the recording form of the sample data in MP4 file, and management information. It is a flowchart which shows an example of the chunk rearrangement process in the 1st Embodiment of this invention. It is a figure which shows the structure of the MP4 file output in embodiment of this invention. It is a flowchart which shows an example of the chunk rearrangement process in the 2nd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Video input part 2 Audio | voice input part 3 Video | video encoding part 4 Audio | voice encoding part 5 File output part 6 Management information storage part 7 Data structure rearrangement part 8 Output file 10 Recording part

Claims (8)

A file recording method for recording sample data consisting of encoded data of moving images or audio, chunk data consisting of a plurality of the sample data, and management information for managing the sample data and chunk data in a file,
A reserved area generating process for generating a reserved area for recording the management information in the file in a batch;
Data relocation processing for relocating the chunk data to a different position in the file,
A file recording method, wherein the size of the reserved area is changed by rearranging the chunk data at a different position in the file. A file recording apparatus for recording sample data consisting of encoded data of moving images or audio, chunk data consisting of a plurality of the sample data, and management information for managing the sample data and chunk data in a file,
Reserved area generating means for generating a reserved area for collectively recording the management information in the file;
Data rearrangement means for rearranging the chunk data at different positions in the file,
The file recording apparatus, wherein the data rearrangement means changes the size of the reserved area by rearranging the chunk data at a different position in the file. Size management means for managing the size of the management information and the reserved area;
The size management means requests the data relocation means to relocate the chunk data when the size of the management information is larger than the size of the reserved area;
The management information and the size of the reserved area are updated by performing a process of rearranging the first chunk data in the continuous chunk data at the end of the file by the data rearrangement unit. The file recording apparatus according to claim 2. 4. The file recording apparatus according to claim 3, wherein the data rearrangement unit repeatedly performs the chunk data rearrangement process until a reserved area large enough to output management information is obtained. . Size management means for managing the size of the management information and the reserved area;
The size management means manages the size of an unused area that is not used for recording the management information in the reserved area, and the size of the unused area is equal to or less than the size of chunk data arranged at the end of the file. If so, the data relocation means is requested to relocate the chunk data,
The size of the management information and the reserved area is updated by performing a process of rearranging the chunk data at the end of the file in the unused area by the data rearranging means. The file recording device described. The data rearrangement means also rearranges and changes data blocks other than chunk data, and keeps the file structure correct so that the file format does not contradict the standard by chunk rearrangement. Item 6. The file recording device according to any one of Items 2 to 5. A file recording method for recording sample data consisting of encoded data of moving images or audio, chunk data consisting of a plurality of the sample data, and management information for managing the sample data and chunk data in a file,
A reserved area generating process for generating a reserved area for recording the management information in the file in a batch;
A data relocation process for relocating the chunk data to a different position in the file;
A computer program for causing a computer to execute processing for changing the size of the reserved area in the data rearrangement processing. A computer-readable recording medium on which the computer program according to claim 7 is recorded.

Images