JP2001257979A - Device and method for encoding image, signal transmitting method, device and method for decoding image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform image encoding capable of flexibly dealing with not only a still picture but also the set of still pictures or video moving images. SOLUTION: An encoding object image is sent to a sequence dividing part 1 and divided into plural sequences, and the inside of each sequence is divided into one or more tiles by a tile dividing part 2. An encoding part 3 encodes the image inside each of tiles on the basis of JPEG-2000 standard and sends a provided encoded bit stream to an output bit stream generating part 5. The output bit stream generating part 5 describes the number of tiles inside the relevant sequence in the header part of the encoded bit stream.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image encoding apparatus and method for encoding a still image, a continuous still image, a moving image, etc. and converting the encoded image into an output bit stream, a signal transmission method, and an image decoding apparatus and method. is there.

[0002]

2. Description of the Related Art As a conventional typical image compression method, I
SO (International Organization f)
or Standardization)
(JointPhotographic Coding Experts Group) There is a standard coding method. The encoding of the JPEG standard is DCT
(Discrete Cosine Transform) is a method that mainly compresses and encodes still images. When relatively high bits are allocated, good encoding and
It is known to provide decoded images.

In addition, the ISO is formulating the JPEG2000 standard as a next-generation international standard for still images, instead of the JPEG standard. Among them, JPEG2000 Part-1 is a standard for realizing a decoder having a minimum configuration of a still image.
On the other hand, JPEG2000 Part-3 is currently being developed as a standard for a still image continuous or moving image decoder based on the above Part-1.

At present, the most popular moving picture compression format is MPEG (Moving Picture Experts Groove).
up) -2 standard and DV (Digital Video) standard. The former is used to compress DVD video, while the latter is
Digital video camera for home, also called VC standard,
Popular as a movie compression format. Since all of these formats are recorded on an optical disk or a tape medium, there are cases where the file format and data structure are also specified. For example, in the case of the DV standard, except for some examples of professional broadcast editing equipment such as non-linear editing equipment, for recording on a tape, a compression format and a data structure according to the medium of the tape are required. taking it.

[0005] Although not standardized by ISO, as a de facto standard,
There is also a moving image compression technology called the Motion-JPEG standard. This means that every single still image is a JPEG
The compression is performed by a standard encoding method, and it may be considered that the compression is an extension of still image encoding rather than moving image encoding. Mo
The encoding method of tion-JPEG standard is video capture of PC expansion board, digital still camera (Digital Still Camera)
ra) video codecs, nonlinear editing equipment, etc.

[0006]

[Problems to be solved by the invention] By the way, JPEG2000 Part-3 (commonly known as Motion-JPEG2000), which is currently being standardized,
At the moment, as in Motion-JPEG above, multiple images are JPE
It is compressed by the means of G2000 Part-1. But Mo
Even in Motion-JPEG2000, no study has been done on the MPEG-2 standard or DV standard.
-In order to introduce the technology of JPEG2000, it is necessary to define the above data structure.

In view of the above-mentioned circumstances, the present invention provides a JPEG2000 standard in the same way as the MPEG-2 standard, DV standard, Motion-JPEG standard, etc., which stipulates the recording format and data structure of applied products. It is an object of the present invention to provide an image encoding device and method, a signal transmission method, and an image decoding device and method for defining a format and a data structure for recording on a tape-shaped recording medium or packetizing and serially transmitting using a standard And

[0008]

In order to solve the above-mentioned problems, the present invention divides an image to be encoded into a plurality of first division units, and divides each of the divided first division units. The image is divided into one or more second division units, and the image in each of the divided second division units is encoded based on a predetermined encoding standard, and the header of the encoded bit stream of the first division unit is encoded. The number of the second division units inside the first division unit is described in the section.

Here, the predetermined coding standard is JPEG2000.
The first division unit is a sequence corresponding to a track of a tape-shaped recording medium, and the second division unit is a tile. Note that the sequence is usually a rectangular area.

In this case, one picture of the input image is divided into a plurality of sequences, the divided sequence is divided into a plurality of tiles, and the obtained tile images are encoded by tile-based encoding based on the JPEG2000 standard. To output a bit stream. At this time, encoding control may be performed so that the target encoded bit amount is obtained. The coded bit stream of the JPEG2000 standard and necessary information such as tile numbers are multiplexed and described in the header section,
Generate an output bitstream.

Further, the present invention provides a method for transmitting an encoded bit stream of one or more pictures obtained by encoding based on a predetermined encoding standard, based on a predetermined serial transmission interface standard. As encoding based on the encoding standard, an encoding target image is divided into a plurality of first division units, and each divided first division unit is divided into one or more second division units. Image data corresponding to a plurality of blocks required for a packet defined by the predetermined encoding standard is encoded based on a predetermined encoding standard for an image in each of the divided second division units. Is included in the coded bit stream of the first division unit or the second coding unit therein.

The predetermined coding standard is the JPEG2000 standard, the first division unit is a sequence, the second division unit is a tile, and the predetermined serial transmission interface standard is an IEEE1394 standard. Is mentioned.

In this case, a coded bit stream of the JPEG2000 standard is cut out by a predetermined number of blocks, and a predetermined number of these blocks are grouped into a coded bit stream in a tile. The decoding is performed, a predetermined number of decoded images for each tile obtained by decoding are combined to generate a sequence, and the obtained sequences are combined together in a predetermined number to generate one picture.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. The form will be described.

First Embodiment FIG. 1 is a block diagram showing an example of an image encoding device according to a first embodiment of the present invention. The image encoding device illustrated in FIG. 1 includes a sequence division unit 1 that divides an encoding target image into a plurality of sequences that are first division units,
A tile division unit 2 that divides each divided sequence into tiles that are one or more second division units, and an encoding unit 3 that encodes an image in each divided tile based on the JPEG2000 standard. , An output bitstream generation unit 5 that describes the number of tiles in the sequence in a header portion of the encoded bitstream of the sequence, and a control unit 4 that controls the operation of each unit.

The encoding unit 3 based on the JPEG2000 standard includes a wavelet transform unit 31 to which an input image is supplied, a coefficient quantizer 32 for quantizing the wavelet transform coefficients from the wavelet transform unit 31, and a coefficient quantizer 32. And an entropy encoding unit 33 that entropy encodes the quantized coefficient from.

Next, the operation will be described. Input image 10
0 is first input to the sequence division unit 1 and divided into one or more sequences. FIG. 2 illustrates this state, and shows an example in which a 720 (pixel) × 480 (line) image is vertically divided into ten sequences SQ0 to SQ9. Therefore, the size of one sequence SQ is 720 (pixels) × 48 (lines).

The sequence image 101 is further divided into one or more tiles by the tile division unit 2, and a tile image 102 is output. FIG. 3 illustrates this state, and shows that one sequence is equally divided into four tiles TL0 to TL3. In this case, the size of one tile TL is 240 (pixels) × 48 (lines).

Each tile image 102 is encoded by the JPEG2000 encoding unit 3 for each tile based on the JPEG2000 encoding rule, and a bit stream 103 for each tile is output. At the same time, a control signal is issued from the control unit 4 to the tile division unit 2, and the tile division number 105 is output from the tile division unit 2 to the output bit stream generation unit 5. The number of tile divisions 105 is multiplexed in the output bitstream generation unit 5 into a header portion of a bitstream of a sequence composed of all tiles in the sequence, and a final bitstream 107 is output for each sequence. .

Although the basic configuration has been described above, it is apparent that the sequence bit stream may be further multiplexed at the final stage and output as one bit stream.

Second Embodiment A second embodiment according to the present invention will be described. In the second embodiment, when recording an encoded bit stream generated by the encoding device described in the first embodiment on a tape medium, a bit stream for each sequence is recorded on each track of the tape. Record Hereinafter, a specific description will be given.

FIG. 4 is a diagram for explaining a recording format (525/60 Hz system) on a tape according to the DV standard currently used for home digital movies. 1
The image of the frame is divided into ten oblique recording tracks and recorded, and the data written on one recording track includes a subcode sector (Subcode Sector) SS, a video sector (Video Sector) VS, and an audio sector (Audi).
o Sector) AS and track information area (Track Informat)
ion Area) TIA. Each of the ten recording tracks constituting one frame corresponds to each of the above sequences. FIG. 5 is a diagram showing each data structure in detail.

As shown in FIG. 5, one frame is one frame.
It comprises zero sequences (DIF Sequence) SQ0 to SQ9, and each of the sequences SQ0 to SQ9 has a header of H ₀ , SC ₀ , SC ₁ , VA ₀ , VA ₁ , VA ₁ and VA ₂ , respectively, and audio data. a ₀ -A _8, the image data V ₀ -V ₁₃₄
It is composed of Each of these is a fixed-length block of 80 bytes and is called a DIF block. That is, each DIF block is composed of a header DIF block H ₀ , a subcode DIF block SC ₀ , SC ₁ , and a VAUX DIF block V
A ₀ , VA ₁ , VA ₂ , audio DIF block (Audio DI
F Block) A ₀ -A ₈ , video DIF block (Video DIF Bl
ock) is a V ₀ -V _134. Therefore, each sequence SQ0
-SQ9 image data is 80 bytes per sequence
X135 = 10,800 bytes.

FIG. 6 shows a data structure created for recording a Motion-JPEG2000 standard bit stream by applying the DV standard (525/60 Hz) system. As shown in FIG. 6, each sequence (DIF Sequence) SQ
Although the existing header part of FIG. 5 is used as it is as the header part of 0 to SQ9, the audio DIF block (Audio
DIF Block) A ₀ -A ₈ and Video DIF Block (Video DIF
Block) V ₀ -V ₁₃₄ are not divided into small DIF blocks, but are configured as areas ADB and VDB having an assigned byte length corresponding to each DIF block group. That is, the audio DIF block area ADB of FIG. 6 is 80 bytes × 9 = 720 bytes corresponding to the audio DIF blocks A _{0 to} A ₈ of FIG. 5, and the video DIF block area VDB of FIG. Block V ₀ -V
It is 80 bytes x 135 = 10,800 bytes corresponding to ₁₃₄ .

FIG. 6 shows that each sequence (DIF Sequence)
Since the coded bit stream of the JPEG2000 standard is written in the 10,800 byte video DIF block area VDB in SQ0 to SQ9, it is recorded together with other data on the recording track of the corresponding tape.

In the second embodiment, the encoding based on the JPEG2000 standard is controlled so that the data length of the encoded bit stream in the sequence is fixed.

As described in the first embodiment, if the sequence is divided into a plurality of tiles and encoding is performed based on the tile-based JPEG2000 standard, the video DIF block area shown in FIG. Obviously, the VDB can be divided into the number of tiles. Here, the encoding based on the JPEG2000 standard is controlled so that the data length of the encoded bit stream of the tile becomes a fixed length.

The sequence (DIF Sequence) shown in FIG.
In SQ0 to SQ9, in each sequence SQ, an audio DIF block area ADB is recorded together with a video DIF block area VDB, and in the same sequence, an image coded bit stream based on the Motion-JPEG2000 standard and audio data (Audio coded bit stream).

Third Embodiment A third embodiment of the present invention will be described. In the third embodiment, the header of the coded bit stream in the sequence described in the above embodiment includes moving image information such as image resolution, the number of scanning lines or color difference component information, and interlaced / progressive identification information. Means of describing information,
And a specific example for realizing the format will be described.

FIG. 7 shows a header DIF block H ₀ , sub-code DIF blocks SC ₀ and SC in the header portion of the above sequence.
Is a diagram illustrating an example of _a data structure. The undefined (Reserved) part in each code corresponds to the currently unused bit. Therefore, by using these unused bits, information of a moving image such as the above-mentioned image resolution, the number of scanning lines or color difference component information, and interlace / progressive identification information may be recorded. Also, when the sequence is divided into tiles, these undefined (Reserve)
The number of tiles may be described using the bits in d).

FIG. 8 shows a specific example in the case where a fixed bit length is assigned to each piece of information. That is, FIG.
(A) shows an example in which the resolution information is represented by 8 bits.
FIG. 8B shows an example in which the interlaced progressive information is represented by 1 bit, FIG. 8C shows an example in which the color difference component information is represented by 4 bits, and FIG. , And an example in which the scanning line information is represented by 8 bits.

Here, if the number of tiles is made variable for each sequence, encoding control can be performed in consideration of, for example, local characteristics of an image, and subjective image quality can be improved. In addition, since the tiles are coded completely independently, there is an advantage that the decoder can perform random access to each tile.

Fourth Embodiment In the fourth embodiment, the encoding unit 3 and the output bit stream generation unit 5 of the JPEG2000 standard described in the first embodiment are used to output one or more pictures. The encoded bit stream is transmitted to a transmission line of a predetermined serial transmission interface standard, for example, an IEEE1394 standard. This IEEE
IEEE of the 1394 standard is the Institute of Electrical and Electronics Engineers
gineers, Inc.), and the official name of the IEEE1394 standard is IE
EE Std. 1394-1995 IEEE Standard for a High Perforanc
e serial Bus standard. The serial bus cable conforming to the IEEE1394 standard adopts a format in which data is packetized and transmitted.

FIG. 9 is a diagram showing an example of a packet structure in the case of transmitting the data of the DV standard. One data packet is composed of a packet header, a header CRC (Header CRC), and a CIP header ( CIP Heade
r), six DIF blocks (DIF blocks), and a CRC.

In the case of the DV standard, as described above with reference to FIG. 5, nine audio DIF blocks (Audio DIF blocks) are used.
ck) A ₀ -A ₈ , 135 video DIF blocks (Video D
IfBlock) There is V ₀ -V _134, 6 pieces of the header portion of the DIF blocks (DIF Block), present a total of 150 amino DIF blocks (DIF Block) is. These six DIF blocks are extracted in order, and the six DIF blocks (DIF blocks) shown in FIG.
Block).

On the other hand, the motion described in the above embodiment has been described.
According to the JPEG2000 standard, as shown in FIG. 6, one sequence SQ includes an audio DIF block area ADB of 720 bytes and a video DIF block area V of 10,800 bytes.
The number of tiles in the sequence determines the division number of each area ADB and VDB. Also, the encoded bit stream of one tile may extend over a plurality of packets in FIG. That is, the coded bitstream of a tile is the sum of all six
Occupies a DIF block or the DIF of an adjacent packet
It is also conceivable that it extends to a block. In this case, the realizing means described in the following fifth embodiment determines which DIF
It is possible to determine which tile the block corresponds to.

Fifth Embodiment In the fifth embodiment, a specific example of writing a JPEG2000 tile-coded bit stream in a DIF block of a packet described in the fourth embodiment will be described. . Hereinafter, a case where necessary information is embedded in a CIP header (CIP Header) in a packet based on the IEEE1394 standard will be described.

FIG. 10A shows a packet header (Packet).
10B shows a CIP header in the case of the DV standard, and blank portions 25 and 26 in FIG. 10B are empty bits. Using the bits of these blank portions 25 and 26, the tile number or sequence number of the source of the data of the DIF block in the packet may be described.

The block multiplexed in the IEEE1394 standard packet is referred to as a DI defined by the DV standard.
F block.

The block multiplexed in the packet of the IEEE1394 standard has a fixed bit length.

Sixth Embodiment In the sixth embodiment, a case will be described in which the block length multiplexed in the IEEE1394 standard packet becomes variable depending on the encoding bit rate.

First, FIG. 11 is a diagram for explaining an example of transmitting a data stream of the MPEG standard based on the existing IEEE 1394 standard. In FIG. 11A, 18
8-byte fixed MPEG transport stream (Trans
port Stream) 4 byte timestamp (Tim
We are making a Source Packet SP with a total length of 192 bytes that includes an e Stamp. FIG. 11B
Shows an example of a packet configuration of the IEEE1394 standard, and shows either (a) a combination of a plurality of SPs, or (b) a division of one SP, as an MPEG packet in an IEEE1394 packet.
Multiplexed as t. This MPEG Packet has a variable length depending on the encoding bit rate.

Similar to the case of FIG. 11, FIG. 12 shows an example in which a data stream of the Motion-JPEG2000 standard described in the embodiment of the present invention is transmitted based on the IEEE1394 standard. That is, FIG. 12 is a diagram for describing an implementation example in the case of transmitting a variable-length encoded bit stream of the Motion-JPEG2000 standard based on the IEEE1394 standard. In FIG. 12A, a 4-byte time stamp is added to a fixed-length (for example, 80-byte) packet (Motion-JPEG2000 Packet) determined in advance by the encoding bit rate, and the fixed-length packet is added. I am making a source packet SP. FIG. 12B shows the Motion-JPE in the packet of the IEEE1394 standard.
The following shows an example of a configuration in which a source packet SP of the G2000 standard is multiplexed, in which (a) a plurality of SPs is combined or (b) 1
Of SPs is divided into Motion-JPEG2000 packet groups (Motion-JPEG2000 P
acket group). This Motion-JPEG2000
The packet group has a variable length depending on the encoding bit rate.

The data in the Motion-JPEG2000 Packet is obtained by extracting the sequence described in the previous embodiment or the encoded bit stream generated by tile encoding in the sequence. Blocking and dividing (corresponding to (b) above) or combining a plurality of them (corresponding to (a) above) may be mentioned.

Next, FIG. 13A shows an MPEG-TS transport stream MPEG-TS data stream.
CIP header (CIP) for transmission based on the EEE1394 standard
If this structure is followed in the case of the Motion-JPEG2000 standard, the 23-bit undefined (Reserve) field in the 24-bit area FDF shown in FIG.
The tile number and the sequence number as described above may be described using the area 28 of d).

On the other hand, as shown in FIG. 12, by adding time stamp information to blocks necessary for a packet, there is an advantage that audio and video can be easily synchronized.

Seventh Embodiment Next, a seventh embodiment of the present invention will be described. In the seventh embodiment, the IEEE13 described in the above embodiment is used.
Motion-JPEG2 in a block multiplexed into 94 packets
A method of recording or extracting 000 coded bitstreams in order from low resolution to high resolution with scalability of resolution, and the order of low to high quality with scalability of image quality A method for recording or extracting the information in a row will be described.

Here, before describing the scalability of the resolution, the wavelet (Wavelet) transform of the encoding technique which is the basis of the JPEG2000 standard will be described with reference to FIG. This wavelet transform is based on the JP in FIG.
Wavelet transform unit 31 in EG2000 standard encoding unit 3
It is performed in.

FIG. 14 illustrates band components obtained as a result of band-dividing a two-dimensional original image having a size of X_SIZE in the horizontal direction and Y_SIZE in the vertical direction to level 2. That is, seven sub-bands generated by applying a horizontal / vertical filter to the input two-dimensional image and then performing a wavelet transform twice such as to perform 1/2 down-sampling thereafter are shown. ing. In FIG. 14, first, four components LL, LH, HL, and HH are divided by level 1 band division (horizontal and vertical directions). Here, LL means that the horizontal and vertical components are both L, and LH means that the horizontal component is H and the vertical component is L. Next, the LL component is band-divided again to generate LLLL, LLHL, LLLH, and LLHH. As described above, in addition to hierarchically dividing low-frequency components, the entire band is equally divided.

Next, a bit plane which is an indispensable technique for realizing the scalability of the image quality will be described with reference to FIG. A bit plane is a binary value from the MSB to the LSB that can be created when the sample value is expressed in binary.
Each plane is a set of 0s and 1s.

FIG. 15 (a) shows 16 quantization coefficients of 4 rows and 4 columns, and +13, -6, etc. represent coefficient values after quantization. These quantized coefficients are divided into their absolute value and positive / negative sign (+-), and the absolute value is MS
The data is expanded from B to the LSB bit plane. FIG. 15B shows each bit plane of an absolute value, and FIG. 15C shows a bit plane of a code. FIG.
The coefficient on the bit plane of the absolute value of (b) is either 0 or 1, and the coefficient on the bit plane of the sign of (c) in FIG. 15 is any of +, 0, and-. . FIG.
(B) and (c) are composed of four absolute value bit planes and one sign bit plane. As the subsequent processing, binary coding for each bit plane is performed.

By expressing the coefficient value obtained by performing the wavelet transform using the bit plane, the coefficient value can be restored in order from the most significant bit (MSB) to the least significant bit (LSB), for example. Thus, progressive decoding in which the image quality gradually improves while the resolution remains the same as the original image can be realized.

Therefore, the encoded bit stream of the Motion-JPEG2000 standard in the block multiplexed with the packet of the IEEE 1394 standard is obtained by dividing all the bands (subbands) generated by the wavelet division as shown in FIG. )
The transform coefficients are coded for the components in order from MSB to LSB. That is, MSB of all band components
After encoding the transform coefficients of the bit planes, the next bit planes of all band components are encoded, and the LSBs are sequentially encoded.
Are encoded up to the bit plane of.

On the other hand, in the scalability of the resolution, for each band (sub-band) component described with reference to FIG. 14, in the order of the lowest band to the highest band, the Motion- JPEG2000 standard coded bit stream is arranged. Therefore, after an encoded bit stream of one subband is recorded,
The configuration is such that the encoded bit stream of the next higher sub-band is added. Therefore, unlike the above-described scalability of image quality, there is no straddling of sub-bands. The above is an example of realizing the scalability of the resolution.

Eighth Embodiment Next, an eighth embodiment will be described. While the above-described embodiments relate to an encoding apparatus and method using the encoding technology of the JPEG2000 standard, the eighth embodiment uses a decoding technology based on the JPEG2000 standard. The present invention relates to an apparatus and a method.

FIG. 16 is a block diagram showing a configuration example of an image decoding apparatus according to the eighth embodiment. The image decoding apparatus includes a bit stream extraction unit 6 to which an encoded bit stream encoded by the encoding apparatus based on the JPEG2000 standard of the first embodiment is supplied,
A decoding unit (JPEG2000 decoding unit) 7 for decoding based on the EG2000 standard; and a tile synthesizing unit 8 for synthesizing a sequence as a first division unit by grouping a plurality of tiles as a second division unit of a decoded image. And a sequence combining unit 9 for combining a plurality of tile-combined images into one picture image. Next, the operation will be described.

The bit stream extraction unit 6 reads the coded bit stream 107 and reads conditions such as the number of tiles at the time of coding and the fixed byte length in the block. As a result, the encoded bit stream 1 for each tile
08 is extracted, and is subjected to tile-based decoding in the JPEG2000 decoding unit 7 to output a tile decoded image 109.

Next, in the tile decoded image 109, all tile images in the sequence are synthesized by the tile synthesizing unit 8, and the sequence image 110 is reproduced. continue,
The sequence image 110 is synthesized by the sequence synthesizing unit 9, and finally a decoded image 111 of one picture is output.

Also, the number of tiles in the sequence recorded in the header portion of the coded bit stream 107 in the sequence is decoded, and the number of tiles is detected from this. The length may be calculated.

The resolution, the number of scanning lines, the color difference component information, and the interlace / progressive identification information of the image recorded in the header of the coded bit stream 107 in the sequence can also be described. By this, JPEGF2
The 000 decoded image can be restored at a predetermined resolution using predetermined color information. For video images,
Since it is important to identify whether the image is interlaced or progressive, it is possible to display a proper video image by identifying the image.

As described with reference to FIG. 6, when the coded bit stream is multiplexed with audio data in addition to image data in the same sequence, a block of the coded bit stream corresponding to the audio data is multiplexed. , Can be extracted by the bit stream extraction unit 6.

Further, the image and audio data extracted by the above means can be reproduced in synchronization. FIG.
FIG. 7 is a diagram for explaining a specific example of such synchronous reproduction. That is, when the data configuration as shown in FIG. 6 described above is used, since the image and audio data streams are recorded as a pair in the same sequence SQ, they must be synchronized in the same time frame even during display and reproduction. Can be. That is, in the example of FIG. 17, the picture a and the corresponding sound are reproduced in synchronization with each other, and another picture b is reproduced.
And the corresponding sound are reproduced synchronously. In addition,
This example has an advantage of being superior in editability as compared with the example of synchronization using a time stamp described later with reference to FIG.

Ninth Embodiment In a ninth embodiment, a plurality of data blocks in a data packet of the IEEE1394 standard are collected in a JPEG format.
Decoded images are generated by decoding means of the 2000 standard, a predetermined number of the decoded images are collected to generate a sequence image, and a predetermined number of the sequence images are collected to generate one picture. This is a decoding device corresponding to the encoding device of the fourth embodiment, and the configuration is as follows.
It may be the same as FIG.

Although the packet structure of the IEEE1394 standard has already been described, a plurality of data blocks in the packet are subjected to tile-based decoding based on the JPEG2000 standard to generate a tile-decoded image. A predetermined number of tile decoded images are collected to generate a sequence image. Further, a predetermined number of sequence images are collected to obtain a decoded image of one picture.

At this time, a plurality of data blocks in the packet are collected, an encoded bit stream necessary for restoring the tile image is extracted, and tile-based decoding of the JPEG2000 standard is performed on the extracted bit stream. Can be restored. Further, as described above, a sequence image can be generated by combining a predetermined number of tile images.

Tenth Embodiment The tenth embodiment is directed to a partially coded bit stream formed by collecting a plurality of data blocks in a data packet of the IEEE1394 standard, or a partially coded bit stream formed by dividing. The stream is collected to a predetermined length, and the generated coded bit stream is generated by a decoding unit based on the JPEG2000 standard to generate a decoded image. This is a decoding device corresponding to the encoding device of the sixth embodiment.

In the tenth embodiment, first, IEEE13
A plurality of data blocks in a 94 standard packet are collected or divided to create a partially encoded bit stream. It is assumed that the data length of the partially coded bit stream is predetermined. Next, the partially coded bit stream is decoded based on the JPEG2000 standard to obtain a decoded image.

As shown in FIG. 12, Motion-JPEG2000
If the data of the standard partially coded bit stream and the time stamp information are multiplexed, the generated JPEG2000
The decoded image can be displayed in synchronization with the time, and the synchronized playback with the audio can be performed as described in the above embodiment.

FIG. 18 is a diagram specifically showing this.
By using a time stamp, an image and an audio having the same time stamp are simultaneously reproduced while treating the image and the audio completely separately, so that the images and the audio can be output in synchronization with each other.

Eleventh Embodiment Next, an eleventh embodiment of the present invention will be described.
In the eleventh embodiment, at the time of encoding, the encoding bit stream of the tile image selected from the plurality of sequences is set so that the data length of the encoding bit stream in the sequence becomes a constant value. Control is performed so that the sum of the data lengths is always constant.

In the second embodiment and the like, the description has been given of controlling the data length of the coded bit stream of the tile image to be a fixed value. However, the number of generated bits depends on the complexity of the texture in the tile image. Occasions arise where the amounts differ widely. In this case, it is difficult to always control the code amount in the tile to a fixed value.

In order to solve this problem, a target tile may be selected over a plurality of sequences, and control may be performed so that the data length of the coded bit stream of all the tiles in the sequence is constant. FIG. 19 is a diagram for explaining this operation. FIG. 20 shows a specific example of the image encoding apparatus, in which the shuffling section 10 is inserted and connected between the encoding section 3 and the output bit stream generating section 5 in the image encoding apparatus having the configuration shown in FIG. are doing. The shuffling means that tiles are mixed between a plurality of sequences so that the data amount of each sequence is substantially equal to each other.

First, FIG. 19 shows each sequence SQ0, SQ1,... Of the encoded bit stream before shuffling.
, And each sequence SQ0s,... After shuffling. As is clear from FIG. 19, a plurality of different sequences, for example, each sequence SQ0, SQ
2, SQ5, SQ2, and tiles TLa,
TLb, TLc, and TLd are extracted, and these tiles TL are extracted.
a, TLb, TLc and TLd together constitute one (after shuffling) sequence SQ0s.

In FIG. 20, a shuffling unit 10 is inserted after the JPEG2000 encoding unit 3 in FIG. In the output bit stream generation unit 5, the shuffling unit 10
For the coded bit stream obtained by shuffling from, the sequence in the image and information on which tile from which shuffled,
The final output bit stream 113 is output.

In the example of FIG. 19, the sequence of the top line and the sequence of the bottom line in the screen may be simultaneously selected as the selected tile. It is necessary to have a means for storing the encrypted bit stream.

As specific application examples of the embodiment according to the present invention described above, digital image recording / editing equipment equipped with a JPEG2000 standard codec, broadcasting equipment, electronic still camera, video movie, surveillance Image recording / editing device or its software module.

It should be noted that the present invention is not limited to only the above-described embodiment, and it is needless to say that various modifications can be made without departing from the spirit of the present invention.

[0078]

According to the present invention, an image to be encoded is divided into a plurality of first division units, and each divided first division unit is divided into one or more second division units. Then, the image in each of the divided second divided units is encoded based on a predetermined encoding standard, and the header of the encoded bit stream of the first divided unit is stored in the header portion of the first divided unit. Second
By describing the number of division units, it is possible to flexibly cope with not only still images but also a group of still images and a video moving image.

Further, the predetermined encoding standard is set to the JPEG2000 standard, the first division unit is set to a sequence corresponding to a track of a tape-shaped recording medium, and the second division unit is set to a tile, thereby forming a tape. Even on a recording medium, a bit stream can be recorded for each sequence by associating the sequence with a recording track, so that a recording device and an editing device using a tape can be easily realized.

When the DV standard is followed, the DIF
Undefined bits (Reserved) in the block (DIF Block)
By embedding the tile number and other necessary information in the JPEG2000 standard (Motion-J
A coded bit stream of the PEG2000 standard can be easily recorded.

Further, when transmitting an encoded bit stream of one or more pictures obtained by encoding based on a predetermined encoding standard based on a predetermined serial transmission interface standard, the encoded bit stream conforms to the predetermined encoding standard. As the encoding based on, the encoding target image is divided into a plurality of first division units, each divided first division unit is divided into one or more second division units, and each divided The image in the second division unit is encoded based on a predetermined coding standard, and image data corresponding to a plurality of blocks necessary for a packet defined by the predetermined coding standard is converted into the image data. Transmission of a coded bit stream based on a predetermined serial transmission interface standard by including in a coded bit stream of one division unit or the second coding unit therein It can be easily realized.

Here, the predetermined encoding standard is set to JPEG2000.
Standard, the first division unit is a sequence, the second division unit is a tile, and the predetermined serial transmission interface standard is the IEEE1394 standard, so that the Motion-JPEG2000 standard can be transmitted via an IEEE1394 standard bus. When transmitting an encoded bit stream, IEEE13
It is easy to realize by arranging a bit stream in blocks in a 94 standard data packet, and also to achieve synchronization with voice.

Further, the resolution scalability and the image quality scalability can be realized by the wavelet transform used in the JPEG2000 standard and the bit plane expression of the coefficients.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of an image encoding device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating an example in which one picture of an image is divided into ten sequences.

FIG. 3 is a diagram showing an example in which one sequence is divided into four tiles.

FIG. 4 is a diagram for explaining a recording format on a tape according to the DV standard (525/60 Hz system).

FIG. 5 is a diagram for explaining each data structure of the DV standard (525/60 Hz system).

Fig. 6: Motion-JPEG2 to DV standard (525 / 60Hz system)
FIG. 9 is a diagram for explaining each data structure when the 000 standard is applied.

FIG. 7 is a diagram illustrating an example of a data structure of a header DIF block H ₀ and subcode DIF blocks SC ₀ and SC ₁ in a header portion of a sequence.

FIG. 8 shows an undefined bit (Reserved) in the header part of the sequence.
FIG. 14 is a diagram illustrating a specific example in a case where information required for ed bit) is allocated.

FIG. 9 is a diagram illustrating an example of a packet structure when data of the DV standard is transmitted.

FIG. 10: Packet header and DV
It is a figure showing the CIP header (CIP Header) in the case of a standard.

FIG. 11 is a diagram for describing an example of transmitting a data stream of the MPEG standard based on the IEEE 1394 standard.

FIG. 12 is a diagram for describing an implementation example in the case of transmitting a variable-length coded bit stream of the Motion-JPEG2000 standard based on the IEEE1394 standard.

FIG. 13: MPEG standard transport stream (MPEG
FIG. 2 is a diagram showing a structure of a CIP header (CIP Header) when a data stream of (-TS) is transmitted based on the IEEE1394 standard.

FIG. 14 is a diagram for explaining band division (division level = 2) of a two-dimensional image.

FIG. 15 is a diagram illustrating an example in which coefficients are developed on a bit plane.

FIG. 16 is a block diagram illustrating a configuration example of an image decoding device according to an eighth embodiment of the present invention.

FIG. 17 is a diagram for describing a specific example of synchronous reproduction.

FIG. 18 is a diagram for explaining another specific example of synchronous reproduction.

FIG. 19 is a diagram for explaining a shuffling operation for setting the data length of an encoded bit stream in a sequence to a constant value.

FIG. 20 is a block diagram illustrating a schematic configuration of an image encoding device according to an eleventh embodiment of the present invention.

[Explanation of symbols]

1 sequence division section, 2 tile division section, 3 encoding section (JPEG2000 encoding section), 4 control section, 5 output bit stream generation section, 6 bit stream extraction section, decoding section (JPEG2000 decoding section), 8 tile synthesis section , 9 Sequence synthesis unit, 10 Shuffling unit

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) 5C053 FA07 FA22 GA11 GB11 GB18 GB22 GB26 GB32 GB36 GB37 JA03 JA07 LA01 LA14 5C059 KK22 LC03 MA00 MA23 MA24 MA32 MC11 ME01 ME13 PP01 PP04 PP16 SS01 SS13 SS14 SS15 SS30 UA02 UA05 UA21 CA00 DA00 DA01 DA02 5J064 AA01 BA01 BA09 BA15 BA16 BB09 BC02 BC16 BD02 BD03 9A001 BB03 BB04 EE04 HH27 HH30

Claims (40)

[Claims] A first division unit that divides an image to be encoded into a plurality of first division units; and divides each divided first division unit into one or more second division units. A second dividing unit that encodes the divided image in each second division unit based on a predetermined encoding standard; and a header of the encoded bit stream of the first division unit. Means for describing the number of the second division units inside the first division unit in the first division unit. 2. The method according to claim 1, wherein the predetermined encoding standard is the JPEG2000 standard, the first division unit is a sequence corresponding to a recording track of a tape-shaped recording medium, and the second division unit is a tile. The image encoding device according to claim 1, wherein: 3. The apparatus according to claim 1, wherein said encoding means has means for controlling a data length of an encoded bit stream in said first division unit to be a fixed value. Image encoding device. 4. The apparatus according to claim 1, wherein said encoding means includes means for controlling a data length of an encoded bit stream of the image of the second division unit to be a fixed value. The image encoding device according to claim 1. 5. The image encoding apparatus according to claim 1, wherein the first division unit includes corresponding audio data in addition to the image data. 6. A header portion in an encoded bit stream in the first division unit includes at least one of image resolution, scanning lines in the case of a moving image, color difference component information, and interlace / progressive identification information. The image encoding device according to claim 1, wherein 7. The encoding means according to the JPEG2000 standard, wherein the encoded bit stream data of the tile image selected from the plurality of sequences is selected so that the data length of the encoded bit stream in the sequence becomes a constant value. 3. The image encoding apparatus according to claim 2, further comprising control means for controlling the sum of the lengths to be always constant. 8. The image encoding apparatus according to claim 7, wherein said control means has means for storing an encoded bit stream of tiles of the entire screen. 9. The control means includes means for multiplexing the position information or number of the sequence in which the selected tile exists and the position information or number of the tile into the coded bit stream. 8. The image encoding apparatus according to claim 7, wherein: 10. A first division step of dividing an encoding target image into a plurality of first division units, and dividing each of the divided first division units into one or more second division units. A second division step of encoding the image in each of the divided second division units based on a predetermined encoding standard; and a header of the encoded bit stream of the first division unit. A step of describing the number of the second division units inside the first division unit in the first division unit. 11. The method according to claim 11, wherein the predetermined encoding standard is the JPEG2000 standard, the first division unit is a sequence corresponding to a recording track of a tape-shaped recording medium, and the second division unit is a tile. 11. The image encoding method according to claim 10, wherein: 12. A signal transmission method for transmitting an encoded bit stream of one or more pictures obtained by encoding based on a predetermined encoding standard based on a predetermined serial transmission interface standard. The encoding based on the standard divides an encoding target image into a plurality of first division units, divides each divided first division unit into one or more second division units,
This is performed by encoding the image in each of the divided second division units based on a predetermined coding standard, and corresponds to a plurality of blocks necessary for a packet defined by the predetermined coding standard. The image data to be converted to the first
A signal transmission method characterized by being included in a coded bit stream of the second coding unit or a division unit of the above. 13. The predetermined coding standard is the JPEG2000 standard, the first division unit is a sequence, the second division unit is a tile, and the predetermined serial transmission interface standard is an IEEE1394 standard. 13. The signal transmission method according to claim 12, wherein: 14. The signal transmission method according to claim 12, wherein said encoding is controlled such that a data length of an encoded bit stream in said first division unit becomes a fixed value. 15. The signal transmission method according to claim 12, wherein said encoding is controlled such that a data length of an encoded bit stream of the image of the second division unit becomes a fixed value. 16. A sequence or tile information to which a block in a packet belongs is recorded by using unused bits in a CIP header (CIP Header) of the packet defined in the IEEE1394 standard. The signal transmission method according to claim 13, wherein 17. Blocks required for a packet defined by the IEEE1394 standard are defined by the DV standard.
14. The signal transmission method according to claim 13, wherein the signal transmission method is a DIF block. 18. The signal transmission method according to claim 13, wherein a block required for a packet defined by the EEE1394 standard has a fixed bit length. 19. The signal according to claim 13, wherein a block required for a packet defined in the EEE1394 standard has a variable bit length according to an encoding bit rate of the encoded bit stream. Transmission method. 20. The apparatus according to claim 13, wherein a block required for a packet defined in the EE1394 standard is obtained by dividing the extracted data into blocks and dividing or combining a plurality of blocks. Signal transmission method. 21. The signal transmission method according to claim 13, wherein time stamp information is added to a block required for a packet defined by the EE1394 standard. 22. A signal transmission method for transmitting, based on a predetermined serial transmission interface standard, an encoded bit stream of one or more pictures obtained by encoding based on a predetermined encoding standard. Image data corresponding to a plurality of blocks required for a packet defined in the standard is extracted from the coded bit stream in order from low resolution to high resolution with scalability of resolution. Signal transmission method. 23. The signal transmission method according to claim 22, wherein said predetermined encoding standard is JPEG2000 standard, and said predetermined serial transmission interface standard is IEEE1394 standard. 24. A signal transmission method for transmitting, based on a predetermined serial transmission interface standard, an encoded bit stream of one or more pictures obtained by encoding based on a predetermined encoding standard, Image data corresponding to a plurality of blocks required for a packet defined in the standard is extracted from the coded bit stream in order from low image quality to high image quality with scalability of image quality. Signal transmission method. 25. The signal transmission method according to claim 24, wherein said predetermined encoding standard is JPEG2000 standard and said predetermined serial transmission interface standard is IEEE1394 standard. 26. An image decoding apparatus to which an encoded bit stream encoded based on a predetermined encoding standard is supplied, means for cutting out the supplied encoded bit stream by a predetermined number of blocks, Means for grouping a predetermined number of data into a coded bit stream in the second division unit; means for performing decoding based on the predetermined coding standard for each of the second division units; Means for generating a first division unit by grouping a predetermined number of the obtained decoded images for each second division unit; and means for generating one picture by grouping a predetermined number of the first division units. An image decoding apparatus characterized by the above-mentioned. 27. The predetermined encoding standard is the JPEG2000 standard, the second division unit is a tile, and the first division unit is a tile.
27. The image decoding apparatus according to claim 26, wherein the division unit is a sequence corresponding to a track on a tape-shaped recording medium. 28. The image according to claim 26, further comprising means for decoding the number of tiles in the sequence recorded in the header of the coded bit stream in the sequence and detecting the number of tiles. Decoding device. 29. A means for describing the resolution of an image, the number of scanning lines in the case of a moving image, color difference component information, or interlace / progressive identification information, in a header portion in the coded bit stream. The image decoding apparatus according to claim 26, wherein: 30. The image decoding apparatus according to claim 26, wherein audio data is recorded in the same sequence. 31. The image decoding apparatus according to claim 30, wherein the image data and the audio data in the same sequence are reproduced in synchronization. 32. The image decoding apparatus according to claim 26, wherein said coded bit stream is decoded from low resolution to high resolution in order of scalability of resolution to generate a decoded image. . 33. An image decoding apparatus according to claim 26, wherein said coded bit stream is decoded from low image quality to high image quality with scalability of image quality to generate a decoded image. . 34. An image decoding method in which a coded bit stream coded based on a predetermined coding standard is supplied, wherein a step of cutting out the supplied coded bit stream by a predetermined number of blocks, Combining a predetermined number of data into a coded bit stream in a second divided unit; performing decoding based on the predetermined coding standard for each of the second divided units; A step of generating a first division unit by grouping a predetermined number of the obtained decoded images of each second division unit; and a step of generating one picture by grouping a predetermined number of the first division units. An image decoding method characterized by the following. 35. The method according to claim 35, wherein the predetermined encoding standard is the JPEG2000 standard, the second division unit is a tile, and the first division unit is a sequence corresponding to a track of a tape-shaped recording medium. The image decoding method according to claim 34, wherein: 36. An image decoding apparatus in which an encoded bit stream encoded based on a predetermined encoding standard is transmitted and supplied based on a predetermined serial transmission interface standard, wherein: A decoding means for decoding a plurality of data blocks in the data packet based on the predetermined coding standard; a predetermined number of decoded images generated by the decoding means; An image decoding apparatus comprising: means for generating an image of the first division unit; and means for collecting a predetermined number of images of the first division unit to generate one picture. 37. The predetermined encoding standard is JPEG2000 standard, the predetermined serial transmission interface standard is IEEE1394 standard, and the first division unit is a sequence corresponding to a track of a tape-shaped recording medium. 37. The image decoding device according to claim 36, wherein: 38. An image decoding apparatus in which an encoded bit stream encoded based on a predetermined encoding standard is transmitted and supplied based on a predetermined serial transmission interface standard, wherein A partially coded bit stream obtained by collecting a plurality of data blocks in a data packet or a partially coded bit stream obtained by dividing the data block is collected to a predetermined length, and the generated coded bit stream is transmitted to the predetermined bit stream. An image decoding apparatus, wherein a decoded image is generated by decoding means based on the coding standard of (1). 39. The image decoding apparatus according to claim 38, wherein said predetermined encoding standard is JPEG2000 standard, and said predetermined serial transmission interface standard is IEEE1394 standard. 40. The JPEG based on tom stamp information added to a data block in the data packet.
39. The image decoding apparatus according to claim 38, wherein the decoded image of the 2000 standard is reproduced in synchronization.