JP3882825B2 - Image signal encoding apparatus and image signal encoding method - Google Patents

Description

  The present invention relates to an image signal encoding apparatus and an image signal encoding method that enable additional recording of image information that has been highly efficient encoded on a recording medium.

  In recent years, information compressed by high-efficiency encoding of digitized image signals is recorded on a medium such as an optical disk or a magnetic medium, and the information reproduced from the recording medium is decoded to restore the image signal. Recording and playback systems that play back and play back have been developed and put to practical use. As a representative of such a system, DVD-VIDEO is currently commercialized. DVD-VIDEO uses MPEG2 which is an international standard for high-efficiency encoding of moving images, and records compressed data on a high-density optical disc, thereby recording image data over two hours on a single disc. Can do.

  As an encoding device that realizes high-efficiency encoding using prediction processing between image frames such as MPEG2, a configuration as shown in FIG. 5 is known. The input digital image signal is recorded in the frame memory 2 and delayed in order to be rearranged in the order of encoding according to the encoding syntax. The digital signal output from the frame memory 2 is converted into a two-dimensional block of N pixels in the vertical direction and M pixels in the horizontal direction (usually N and M are 8) by the two-dimensional block conversion circuit 4 in the reference frame. The transformed data is subjected to DCT transformation processing by the orthogonal transformation circuit 6 and sent to the quantization circuit 8. The encoding circuit 10 performs variable length or fixed length encoding on the DCT transform coefficient quantized by the quantization circuit 8 by referring to an address corresponding to the coefficient of the encoding table 12, and the multiplexer 14. As a result, the encoded data and additional information indicating the location of the block in the screen are multiplexed and output as a bit stream to the recording medium 16, for example, an optical disc.

  Along with the above encoding process, the DCT transform coefficient quantized by the quantization circuit 8 is subjected to inverse quantization and inverse DCT transform in the inverse quantization circuit 18 and the inverse orthogonal transform circuit 20, respectively. It is deblocked and restored to an image signal in a state where the encoded bit stream is decoded and stored in the prediction memory 24.

  Subsequently, in the prediction frame 24, a motion vector between the image output from the frame memory 2 and the image stored in the prediction memory 24 is obtained by the motion vector detection circuit 26. The motion vector detection circuit 26 is generally obtained by block matching, and each of the image to be encoded and the image in the prediction memory 24 is converted into a two-dimensional block, and the block having the smallest difference absolute value sum (or difference sum of squares) for each pixel is obtained. The motion component in the screen is output as a motion vector. Based on the output motion vector, the prediction block is cut out from the prediction memory 24 by the motion compensation prediction circuit 28. The motion compensation prediction circuit 28 selects a prediction mode, and sends the difference from the image block to be encoded to the two-dimensional block conversion circuit 4. Thereafter, the same processing as that of each block of the frame is performed, and the DCT transform coefficient is output as a bit stream together with the motion vector and the prediction mode.

  Regarding the control of the code amount, the code amount of the output stream is compared with the target code amount in the rate control circuit 30, and the quantization level of the quantization circuit 8 (in order to approach the target code amount ( Control the quantization scale. In the case of MPEG2, there are three types of picture types having different information amounts, ie, a reference frame (I picture) and two types of prediction frames (P picture: unidirectional prediction, B picture: bidirectional prediction). The target code amount for each frame is calculated using the properties and appearance frequency of the three picture types with respect to the conversion rate.

For example, R is the total code amount allocated for a certain time, NI, NP, NB is the number of frames of each picture of I, P, B within a certain time, and SI, SP, SB is the code amount of each previous picture. If the average value of the quantization scale at that time is QI, QP, QB, the multiplication symbol (multiplication operator) is set as *,
XI = SI * QI
XP = SP * QP
XB = SB * QB (Formula 1)
The above XI, XP, and XB are values representing the level of difficulty of encoding, and the target code amount can be set by allocating the code amount using this ratio for a picture to be processed from now on. When the target code amount of each picture type is TI, TP, TB,
TI = R * XI / {(NI * XI) + (NP * XP) / KP
+ (NB * XB) / KB}
TP = R * XP / {(NP * XI) * KP + (NP * XP)
+ (NB * XB) * KP / KB}
TB = R * XB / {(NI * XI) * KB + (NP * XP) * KB / KP
+ (NB * XB)}
... (Formula 2)
Can be calculated. Here, KP and KB are weights in consideration of encoding processing for P and B, respectively.

  Here, if the complexity (activity) of the preceding image is calculated in advance and the activity value is reflected in the allocation of the target code amount for each frame, the control capability can be enhanced.

For example, if the activity immediately before each frame type is AI, AP, AB, and the activity of the image to be encoded is CA, the encoding difficulty level of the frame type of the frame to be encoded is
XI = XI * CA / AI
XP = XP * CA / AP
XB = XB * CA / AB (Formula 3)
Thus, the complexity of the image to be encoded is reflected in the code amount allocation, and a stream with stable image quality can be output.

  The target code amount is limited so that the overflow / underflow of the buffer does not occur by virtually simulating the stream buffer of the decoding device. The quantization scale in the quantization circuit 8 uses the fact that the scale and the output code amount are generally in an inversely proportional relationship, and sets the quantization scale value CQ for the target code amount for each picture type, for example: Quantization processing is performed using such a calculation formula.

CQ = XI / TI (in the case of I picture)
= XP / TP (P picture)
= XB / TB (in case of B picture) (Formula 4)
Since there is no reference image at the start of encoding, a parameter considering a general image is set as an initial value. For this reason, the control capability of the code amount decreases at the start of encoding, but stable control can be realized as the processing proceeds.

  In each block in the picture, the coded bit stream is suppressed within the target code amount by changing the quantization scale in a direction approaching the target code amount for each block. In the case of fixed transfer rate encoding, if the amount of the encoded stream is less than the set encoding rate even if a fine quantization scale is used, a code that is insufficient before the header code indicating the delimiter of one picture By matching the amount of stuffing bits, it matches the set coding rate.

  A read-only recording medium such as a DVD video is produced by creating a recorded image and pressing it, so that image information is recorded in all data areas at once. However, in the read / write recording medium, it is not necessary to record in all the data areas at once, and additional recording can be performed as needed, and editing recording can be performed from the middle of the encoded stream. When used in applications such as video cameras, recording may be paused to change the composition even when shooting at the same location, and power may be temporarily turned off to replace batteries. Or need to turn off. In such a case, when it is desired to reproduce the already recorded image information and the image information to be recorded from now on as the same sequence, it is necessary to take over the control of the virtual stream buffer to the encoding process to be noted.

  When performing additional recording, in order to control the virtual stream buffer from the already recorded encoded stream, the size and encoding rate of the virtual buffer recorded in the stream are read and the last frame is read. Detects the beginning of the stream, reads the buffer position of the virtual buffer recorded in the header information, and further encodes from the amount of code required for the last frame, the encoding rate, and the size of the virtual buffer. It is necessary to calculate the position of the virtual buffer at the end and perform the encoding process using that value as an initial value. For this reason, there is a drawback that special processing is required at the start of encoding. In addition, since the control parameters used at the time of encoding the already encoded image signal are not recorded on the stream, the parameters are initialized at the time of urgent recording and before the parameters follow the characteristics of the input image signal. The frame encoding process is required, especially when performing additional recording by encoding an image that has similar characteristics to an already recorded image, and has the disadvantage of instantaneously degrading the image quality at the joint. there were.

  The present invention has been made to solve the above-described problem. When performing additional recording, it is possible to guarantee continuity with an already encoded image and to perform encoding processing with little image quality deterioration. It is an object of the present invention to provide an image signal encoding device and an image signal encoding method.

In order to achieve the above object, the present invention provides the following apparatus and method.
(1) In an image signal encoding apparatus for recording an encoded image signal obtained by subjecting an image signal to variable length encoding processing and a control parameter used for the variable length encoding processing on a recording medium ,
Coding for recording at least the position of the virtual buffer in a data area different from the recording area of the encoded image signal of the recording medium as the control parameter at the end of the variable length encoding process of the predetermined unit of the image signal An image signal encoding apparatus comprising a parameter processing means.
(2) In an image signal encoding method for recording an encoded image signal obtained by subjecting an image signal to variable length encoding processing and a control parameter used for the variable length encoding processing on a recording medium ,
Coding for recording at least the position of the virtual buffer in a data area different from the recording area of the encoded image signal of the recording medium as the control parameter at the end of the variable length encoding process of the predetermined unit of the image signal An image signal encoding method comprising a parameter processing step.

  According to the image signal encoding apparatus and the image signal encoding method of the present invention, a control parameter including at least the position of the virtual buffer for recording the code amount in a specific area of a recording medium such as a disk at the end of encoding is recorded. Therefore, when encoding processing is performed by additional recording or editing processing, if the encoding processing is started by referring to the control parameter, virtual buffer control for handling the encoded stream in the same sequence is performed. The processing can be easily realized, and the rate control at the start of encoding can be stabilized.

[Embodiment 1]
FIG. 1 shows a first embodiment of an image signal encoding apparatus to which the present invention is applied. The image signal encoding apparatus includes a frame memory 2, a two-dimensional block transform circuit 4, an orthogonal transform circuit 6, a quantization circuit 8, a coding circuit 10, a coding table 12, a multiplexer 14, an inverse quantization circuit 18, and an inverse orthogonal. The conversion circuit 20, the deblocking circuit 22, the prediction memory 24, the motion vector detection circuit 26, the motion compensation prediction circuit 28, the rate control circuit 30, the encoding parameter register 32, and the encoding parameter recording / reproducing circuit 34 are configured. Among these elements, the device portion from the frame memory 2 to the rate control circuit 30 is the same as that of the conventional device described with reference to FIG. 5, and the description thereof is omitted.

  The multiplexer 14 outputs and records the multiplexed bit stream on the recording medium 16, counts the generated code amount of the recorded stream, and outputs it to the rate control circuit 30. The rate control circuit 30 calculates a target code amount for encoding the next frame from the set encoding rate, generated code amount, and picture type (I picture, P picture, or B picture), and determines the quantization scale. And output to the quantization circuit 8. Here, the parameters used for calculating the target code amount are stored in the encoding parameter register 32 according to the present invention. When the encoding control is performed in the manner shown in the conventional example, the stored parameters are the parameters XI, XP, XB for each picture P, I, B indicating the encoding difficulty, and a fixed time. R indicating the total code amount allocated to N, N indicating the total number of frames in a certain time (= NI + NP + NB), activity AI, AP, AB of the last picture of each frame type, virtual buffer size VS, current virtual buffer Position VC. The above parameters are referred to and updated as necessary for each picture.

  At the end of the encoding process, the encoding parameter recording / reproducing circuit 34 records the parameters in a specific area other than the encoded data area of the recording medium 16 and ends the process. An example of the data format of the area where the parameters of the recording medium are recorded is shown in FIG. Following the lead-in section 40 at the first position of the recording area, there is a data track 42 in which disc information and the like are written. Here, there is a header area for defining medium ID information, that is, disc identification information 420, file format information, that is, disc management information 422, reproduction start point information 424, and the like. Next to this header area, an encoding control information area 426 in which the above parameters are recorded is provided. In the encoding control information area 426 described here, following the header A1 (32 bits) for control parameters, an additional recording head position area A2 (32 bits) indicating the head position of the disk area where additional recording is performed, R area A3 (16 bits) indicating the total code amount R, VS area A4 (16 bits) indicating the size VS of the virtual buffer, VC area A5 (16 bits) indicating the position VC of the virtual buffer, each within a fixed time N area A6 (8 bits) indicating the total number of frames of the picture, XI area A7 (12 bits) indicating the encoding difficulty XI of the I picture, XP area A8 (12 bits) indicating the encoding difficulty XP of the P picture, XB area A9 (12 bits) indicating the encoding difficulty level XB of B picture, activity AI of the last I picture of each frame type AI area A10 indicating (12 bits), likewise AP area A11 (12 bits) that indicates the activity AP P-picture, is recorded in the order of AB area A12 indicating the activity AB B picture (12 bits). After this data area, there are encoded stream tracks 44, 46, and 48 in which encoded information is recorded.

  At the start of encoding, first, a header A1 indicating a control parameter is searched, the contents of the additional recording head position A2 are read, and the recording position is moved to that position. The encoding parameter recording / reproducing circuit 34 reads the control parameter and records it in the encoding parameter register 32. This encoding apparatus starts the encoding process with reference to the encoding parameter register 32 in the same manner as the normal encoding process.

In this way, by reading the control parameters of the already encoded image information and performing additional recording, continuity with the already encoded image can be ensured, and rate control at the start of encoding is possible. Ability can be improved.
[Embodiment 2]
Next, an encoding apparatus showing a second embodiment to which the present invention is applied will be described. In this embodiment, the encoding parameter recording / reproduction circuit 34 is separated into an encoding parameter recording circuit 36 and an encoding parameter reproduction conversion circuit 38. The encoding parameter recording circuit 36 records a signal for identifying the encoding apparatus together with the same parameters as those in the first embodiment. By recording different signals with encoding devices having different control methods, it is possible to know which encoding device has processed the already recorded image information. The identification signal is recorded following parameters common to the encoding device (total code amount R assigned to a predetermined time, virtual buffer size VS and position VC), and parameters specific to encoding are recorded thereafter. . At the start of encoding, the reproducing circuit reads the identification signal and then reads the control parameters. If the identification signal is the same as its own, the control parameter is stored in the encoding parameter register as it is, and the encoding process is started. In the case of an identification signal different from the self signal, control parameter conversion processing is performed, usable parameters are stored in encoding parameters, and initial values are set in other areas.

  When it is desired to continuously perform additional recording on an image signal recorded by an encoding device of a different algorithm, parameters common to the encoding device such as a virtual buffer can be reliably read, so that the continuity of the encoded stream is ensured. be able to. In addition, the recognition signal and parameter conversion rule of the encoding device of another algorithm is stored in advance in a memory or the like, and a parameter necessary for the encoding process is generated from the recognition signal and the parameter, so that it can be used between different devices. The stability of the encoding control process can be improved.

  In the example of the recording medium in FIG. 2, the parameter recording area is provided only in the head area of the recording medium, but a plurality of parameter recording areas may be provided. As shown in FIG. 4, parameter recording areas, that is, control information areas B1, B3, and B6 are provided in the editable stream unit, for example, the encoded stream 44, prior to the stream data B2, B4, and B7, respectively. By reading the parameters at points, it is possible to realize the continuity of the encoded stream and the stability of the encoding control even in the editing process.

It is a block diagram which shows 1st Embodiment of the encoding apparatus to which this invention is applied. It is a figure which shows an example of the data format of the recording medium which implement | achieves 1st Embodiment to which this invention is applied. It is a block diagram which shows 2nd Embodiment of the encoding apparatus to which this invention is applied. It is a figure which shows an example of the data format of a recording medium which owns multiple storage areas of control data. It is a block diagram which shows the conventional image coding apparatus.

Explanation of symbols

2 Frame memory 4 Two-dimensional block transform circuit 6 Orthogonal transform circuit 8 Quantization circuit 10 Encoding circuit 12 Encoding table 14 Multiplexer 16 Recording medium 18 Inverse quantization circuit 20 Inverse orthogonal transform circuit 22 Deblock circuit 24 Prediction memory 26 Motion vector Detection circuit 28 Motion compensation prediction circuit 30 Rate control circuit 32 Encoding parameter register 34 Encoding parameter recording / reproducing circuit 36 Encoding parameter recording circuit 38 Encoding parameter reproducing conversion circuit

Claims (2)

In an image signal encoding apparatus for recording an encoded image signal obtained by subjecting an image signal to variable length encoding processing and a control parameter used for the variable length encoding processing on a recording medium,
Coding for recording at least the position of the virtual buffer in a data area different from the recording area of the encoded image signal of the recording medium as the control parameter at the end of the variable length encoding process of the predetermined unit of the image signal An image signal encoding apparatus comprising a parameter processing means. In an image signal encoding method for recording an encoded image signal obtained by subjecting an image signal to variable length encoding processing and a control parameter used for the variable length encoding processing on a recording medium,
Coding for recording at least the position of the virtual buffer in a data area different from the recording area of the encoded image signal of the recording medium as the control parameter at the end of the variable length encoding process of the predetermined unit of the image signal An image signal encoding method comprising a parameter processing step.

Images