JP4528043B2 - Video signal conversion apparatus, conversion method, and video signal recording apparatus using the same - Google Patents

Description

  The present invention relates to a video signal conversion apparatus and conversion method for converting an encoding system of an input compressed video signal into a different encoding system, and a video signal recording apparatus using the same.

  In recent years, techniques for compressing and encoding video signals, particularly moving picture signals, and recording them on a recording medium, or techniques for transmitting via a communication medium such as the Internet or wireless, are used in various fields. In these applications, in order to save the recording capacity of the recording medium or to save the bandwidth at the time of transmission, a moving image signal compression / expansion technique for reducing the amount of video data is used. For example, in digital television broadcasting, the MPEG-2 (ISO / IEC 13818) standard standardized by ISO / IEC is adopted as a compression / decompression method. Also, MPEG-4 (ISO / IEC 14496) system is used in streaming video distribution over the Internet or wireless. As described above, there are many compression encoding methods depending on the application, but there is a case where these compression methods are not compatible. Even in the same compression encoding system, there are cases where compatibility cannot be maintained due to differences in signal formats such as image size and bit rate.

  In such a situation, a so-called transcoding technique for converting a certain compression coding system into a different compression coding system or a different signal format in the same compression coding system is required. When performing transcoding, for example, as disclosed in Patent Document 1, by extracting and using information contained in a transcoded original compressed encoded signal (stream), the transcoded compressed encoded signal A method for reducing image quality degradation has been proposed.

  One feature of Patent Document 1 is that image quality degradation caused by transcoding is suppressed using information such as motion vectors, prediction errors, orthogonal transform coefficients, and quantization values included in an input stream.

Japanese Patent Laid-Open No. 2000-244921

  However, in the method disclosed in Patent Document 1, since transcoding is performed using only stream information of frames that correspond temporally (that is, at the same time), a moving image such as a scene change in which the code amount changes suddenly. The image cannot be sufficiently dealt with, and the image of the sudden change portion may be deteriorated.

  An object of the present invention is to provide an apparatus and a method for performing video signal conversion with good follow-up and little image quality degradation even when the code amount of an input compressed video signal changes suddenly.

  In order to solve the above problems, a video signal conversion apparatus according to the present invention converts a compression-coded first video signal into a second video signal having a different compression encoding method or a different signal format. A code amount analysis unit that acquires a change in the code amount of the first video signal on the time axis, and an encoding unit that compresses and encodes the first video signal into the second video signal. The encoding unit controls the code amount of the second video signal based on the code amount acquired by the code amount analyzing unit when compressing and encoding the first video signal.

Alternatively, the video signal conversion apparatus of the present invention includes a code amount analysis unit that analyzes a change in the code amount of the first video signal on the time axis, and a code amount that stores the change in the code amount analyzed by the code amount analysis unit A storage unit; a video signal storage unit that stores the first video signal for a predetermined time; a decoding unit that decodes the first video signal stored in the video signal storage unit; A decoding unit that compresses and encodes the video signal into the second video signal. The encoding unit controls the code amount of the second video signal based on the code amount stored in the code amount storage unit at a future position on the time axis when compressing and encoding the first video signal.
Alternatively, the video signal converter according to the present invention includes a first input unit to which the first video signal that has been compression-encoded is input, and the first video signal that is input from the first input unit , After being stored for a predetermined time in the first output unit for output to the outside and the storage unit connected to the outside so that the data is temporarily stored in the storage unit connected to, and then supplied to the external decoding unit It said the read second image signal obtained by decoding the first video signal at the outside of the decoding unit is input and a second input unit inputted from the outside, from the first input unit A code amount analysis unit that analyzes a change in the code amount of one video signal on a time axis, a code amount storage unit that stores a change in the code amount analyzed by the code amount analysis unit, and the second input unit The second video signal input from the first video signal is different from the first video signal in a different compression encoding system or different signal format. An encoding unit that compresses and encodes the third video signal; and a second output unit that outputs the third video signal compressed and encoded by the encoding unit. When the second video signal is compression-encoded, the code amount of the third video signal is controlled based on the code amount stored in the code amount storage unit at a future position on the time axis.

  Also, the video signal conversion method of the present invention acquires the change in the code amount of the first video signal on the time axis, and acquires it at a future position on the time axis when compressing and encoding the first video signal. Based on the code amount, the code amount of the second video signal is controlled.

  That is, by acquiring in advance the coding information of the first video signal (transition of the code amount including the future on the time axis), the allocation of the code amount of the second video signal can be optimally controlled.

  ADVANTAGE OF THE INVENTION According to this invention, even when the code amount of the input compressed video signal changes suddenly, the apparatus and method which perform video signal conversion with little image quality degradation can be provided.

  The best mode for carrying out the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing an embodiment of a video signal conversion apparatus according to the present invention (Embodiment 1). In this embodiment, a case where a coded video signal in the MPEG-2_MP @ HL format is converted and encoded in the MPEG-2_MP @ ML format is shown.

  An input unit 200 and an output unit 300 are connected to the video signal converter (transcoder) 100. The input unit 200 is, for example, a digital tuner that receives digital broadcasting, and the output unit 300 is, for example, a recording device that records a compression-coded video signal.

  The transcoder 100 includes a stream preprocessing unit 101 that separates an input stream into a video stream and an audio stream, a stream analysis unit 102 that analyzes a bit rate of the video stream, and a stream information storage memory 103 that stores the analyzed bit rate transition information. Have A stream storage memory 107 that temporarily stores the input stream for a predetermined time; a decoder 104 that decodes the MPEG-2_MP @ HL format; an encoder 105 that encodes the decoded signal in the MPEG-2_MP @ ML format; A stream post-processing unit 106 that multiplexes the encoded video stream and audio stream is included.

  The decoder 104 includes a video decoding unit 1041 and an audio decoding unit 1042. The encoder 105 includes a video encoding unit 1051 and an audio encoding unit 1052. The video encoding unit 1051 performs transform encoding using the bit rate transition information stored in the stream information storage memory 103. At that time, by supplying the video stream with a delay from the stream storage memory 107, not only the bit rate transition information at the time of conversion but also the bit rate transition information for the future can be used.

  Next, signal processing in this embodiment will be described in detail.

  First, an encoded video signal in the MPEG-2_MP @ HL format is input from the input unit 200 to the transcoder 100. Here, the input signal may be a program stream (PS) in which video and audio are multiplexed, a transport stream (TS), or an elementary stream (ES) including only video. Here, a description will be given assuming that a transport stream (TS) is input.

  FIG. 2 is a diagram illustrating a configuration of a transport stream (TS) packet. TS is a packet having a fixed length of 188 bytes. A TS packet includes a leading synchronization byte 2001, a program ID 2002, a header 2003, and a data body 2004. The program ID 2002 includes information for identifying desired image data from the multiplexed data. The header 2003 includes information indicating the size and attributes of the data body 2004 included in the packet.

  FIG. 3 shows a configuration example of the stream preprocessing unit 101. The stream preprocessing unit 101 includes a stream input buffer 1011, a system decoding unit 1012, and a demultiplexing unit 1013.

  The TS input to the stream preprocessing unit 101 is temporarily stored in the stream input buffer 1011. The TS is then read by the system decoding unit 1012, and only the TS packets designated by the user are collected by collating the program ID designated by the user with the program ID 2002 of the TS packet. Then, an audio stream, a video stream, and other user data are separated from the collected TS packets by the subsequent demultiplexing unit 1013 and output. At this point, the separated audio stream and video stream are both compression encoded.

  The video stream separated and output by the demultiplexing unit 1013 is input to the stream analysis unit 102. The stream analysis unit 102 analyzes the bit rate transition (code amount transition) of the input video stream. The analyzed bit rate transition information is stored in a stream information storage memory 103 connected to the stream analysis unit 102. Detailed operation of the stream analysis unit 102 will be described later.

  On the other hand, the audio stream and the video stream separated by the demultiplexer 1013 are temporarily stored in the stream storage memory 107. The stream storage memory 107 stores a stream for a predetermined time (tens of seconds to several minutes) using, for example, an SD-RAM. If a large capacity is required, a hard disk medium can be used.

  In the stream storage memory 107, the stream is delayed within the stored time, and the delayed stream is input to the decoder 104. The decoder 104 includes a video decoding unit 1041 and an audio decoding unit 1042.

  The audio stream input to the decoder 104 is input to the audio decoding unit 1042, decoded by a method according to the compression encoding method of the audio stream, and output as audio data. Similarly, the video stream input to the decoder 104 is input to the video decoding unit 1041 and decoded by a method according to the video stream compression encoding method (in this case, MPEG-2_MP @ HL format) as video data. Is output.

  The video data and audio data output from the decoder 104 are input to the encoder 105. The encoder 105 includes a video encoding unit 1051 and an audio encoding unit 1052.

  The audio data input to the encoder 105 is compressed and encoded by the audio encoding unit 1052 using a predetermined compression encoding method, and is output as a second audio stream. Similarly, the video data input to the encoder 105 is subjected to a filtering operation or a scaling operation in the video encoding unit 1051, and then compressed by a predetermined compression encoding method (in this case, MPEG-2_MP @ ML format). And output as a second video stream.

  Here, when performing compression encoding in the video encoding unit 1051, bit rate transition information (code amount transition information) analyzed by the stream analysis unit 102 and stored in the stream information storage memory 103 is used. That is, it is possible to optimally control the output bit rate of the video encoding unit 1051 by setting the code amount by pre-reading not only the time of encoding but also the future code amount variation. This bit rate control will be described later.

  The second video stream and second audio stream output from the encoder 105 are input to the stream post-processing unit 106.

  FIG. 4 shows a configuration example of the stream post-processing unit 106. The stream post-processing unit 106 includes a multiplexing unit 1061, a system encoding unit 1062, and a stream output buffer 1063.

  The second audio stream and the second video stream input to the stream post-processing unit are input to the multiplexing unit 1061, multiplexed, and input to the subsequent system encoding unit 1062 as a multiplexed stream. The system encoding unit 1062 performs system encoding on the input multiplexed stream. In this embodiment, output is performed in the MPEG-2 PS format, but other multiplexing formats may be used. The system-encoded PS is temporarily stored in the stream output buffer 1063. Thereafter, it is sent to output means 300 such as a recording device.

  Here, the operation of the stream analysis unit 102 will be described in detail. The stream analysis unit analyzes bit rate transitions on two types of scales, GOP (group of pictures) units and picture units.

  Taking MPEG-2 as an example, there are three types of picture encoding formats. When image data for one screen is encoded within the image (I picture) and when image data for one screen is encoded using a reference relationship with image data that is temporally previous (P picture) In this case, image data for one screen is encoded using a reference relationship between two temporally preceding and following image data (B picture). When the arrangement of pictures in the stream is, for example, “IBBPBBPBBPBBPBBIBB...”, The I picture to the B picture immediately before the next I picture is called GOP.

  FIG. 5 shows a configuration example of the stream analysis unit 102 and the stream information storage memory 103. The video stream input to the stream analysis unit 102 is temporarily stored in the buffer 1021. Then, the subsequent pattern search unit 1022 sequentially reads out the data. The pattern search unit 1022 searches the head of GOP, picture, and subpicture included in the stream, and counts the code amount in GOP units, the code amount in picture units, and the code amount in subpictures.

  FIG. 6 shows an example of the structure of a video stream input to the stream analysis unit. A video stream compressed and encoded in the MPEG-2 format has a header at the head of a GOP and a picture. In the case of FIG. 6, since the GOP header is “0x000001B8” and the picture header is “0x00000100”, the video search stored in the buffer 1021 is sequentially read while searching for these headers in the pattern search unit. The code amount in units of pictures can be obtained.

  The code amount in GOP unit or picture unit thus obtained is stored in the stream information storage memory 103 based on a predetermined format. Since the information to be stored is only the code amount, the code amount can be continuously recorded for a long time even if the memory capacity of the stream information storage memory 103 is small. For example, when 4 bytes are used for time information and 4 bytes are used for code amount information, bit rate transition information of about 4 minutes can be recorded if the memory capacity is 1 Mbyte. This is an amount of information sufficient to be used for rate control at a later stage. If the storage time of the stream information storage memory 103 is matched with the storage time of the stream storage memory 107, the time that can be used as future information can be maximized, which is most efficient.

  Next, compression coding using the bit rate transition information analyzed by the stream analysis unit 102 will be described. It is assumed that when the video encoding unit 1051 performs encoding, bit rate transition information for several tens of seconds to several minutes is stored in the stream information storage memory 103. That is, the future information on how the generated code amount transitions after the time of encoding is stored.

  Therefore, even when the generated code amount suddenly increases due to a scene change or the like in the input signal, the video encoding unit 1051 predicts this in advance and secures the necessary code amount. Therefore, the generated code amount can be controlled with good follow-up regardless of any increase or decrease in the code amount of the input signal.

  FIG. 7 shows an example of control of the generated code amount at the time of transcoding according to this embodiment. The vertical axis represents the code amount at each time, and the horizontal axis represents the passage of time. Based on transcode processing and signal storage, the post-conversion signal is shifted on the time axis with respect to the pre-conversion signal, but this is ignored in the figure, and the corresponding position of each signal is aligned.

  The transition of the code amount of the original signal stored in the stream information storage memory is assumed to increase rapidly at time T0 as indicated by a curve (broken line) 601.

  On the other hand, the code amount transition in the case of transcoding by the conventional method becomes a curve (dotted line) 602. That is, when the amount of generated code increases due to a scene change or the like in the original signal and the increase in the amount of code exceeds the processing capability of the video encoding unit 1051, the code amount control cannot be followed by conventional transcoding (code The actual increase in the code amount is delayed until time T1. For this reason, the section from time T0 to T1 is encoded with a small code amount where a large amount of code is originally required, and the image quality deteriorates.

  However, in the system of this embodiment, since the code amount transition of the stream is acquired in advance, the code amount is ensured in advance in preparation for the expected increase in the code amount of the image data. Therefore, the generated code amount after transcoding can be changed with good followability as shown by a curve (solid line) 603. As a result, it is possible to reduce deterioration in image quality even in scenes such as scene changes.

  As described above, by using the bit rate transition information stored in the stream information storage memory 103, the video encoding unit 1051 can perform efficient rate control with little deterioration in image quality.

  FIG. 8 is a block diagram showing another embodiment of the video signal converter according to the present invention (Embodiment 2). The basic configuration is the same as that of the first embodiment (FIG. 1), but the decoding information processing unit 1053 is added to the basic configuration.

  Although the overall operation of this embodiment is also the same as that of the first embodiment, the decoding information processing unit 1053 receives decoding intermediate information output from the video decoding unit 1041. The decoding intermediate information includes motion vector information, quantized values, DCT coefficients, and the like. The decode information processing unit 1053 processes the decoding intermediate information supplied from the video decoding unit 1041 and converts it into a format that can be used for compression encoding of video data in the video encoding unit 1051.

  For example, an operation for searching for motion vector information in the video encoding unit is an operation with a high processing load even in the MPEG encoding, but the motion vector information included in the decoding intermediate information is processed by the decoding information processing unit 1053 and encoded. By using this, the processing in the video decoding unit 1041 can be greatly reduced. The processing of motion vector information in this case specifically refers to vector scaling and complementation in accordance with the image size to be encoded.

  According to the present embodiment, it is possible not only to perform efficient coding using rate transition information analyzed in advance as in the first embodiment, but also to reduce the processing load of video data compression coding. Become.

  FIG. 9 is a block diagram showing another embodiment of the video signal converter according to the present invention (third embodiment). The present embodiment is an encoder 700 with a stream processing function to which the first embodiment (FIG. 1) is applied, and does not have the stream storage memory 107 and the decoder 104 in the first embodiment, which are executed by connecting an external device. To do.

  The stream input unit 200 analyzes the bit rate information in the stream analysis unit 102 as in the first embodiment, and the result is stored in the stream information storage memory 103. On the other hand, the video stream and the audio stream that have been system-decoded and demultiplexed by the stream pre-processing unit are temporarily stored in the stream storage memory 400 connected to the outside. Thereafter, the data is supplied to an external video decoder 501 and an audio decoder 502.

  When the video stream input to the video decoder 501 is decoded and output as video data, it is supplied again to the encoder 700 with a stream processing function. The audio stream supplied to the audio decoder 502 is similarly decoded and output as audio data, and then supplied again to the encoder 700 with a stream processing function.

  The video data and audio data supplied to the encoder 700 with a stream processing function are supplied to the encoder 105 and encoded, multiplexed and system encoded by the stream post-processing unit, and then sent to the output unit 300.

  The encoder 700 with a stream processing function of this embodiment inputs a compression-encoded stream and data obtained by decoding the stream, and re-encodes the data into a different system. As the stream storage memory 400, the video decoder 501, and the audio decoder 502, existing memories and decoders can be used. In particular, if a large-capacity hard disk medium is used as the stream storage memory 400, a long-time stream can be stored, and thus encoding using long-time stream information is possible, and a more efficient transcoder can be realized.

  Further, a system LSI like this embodiment may be configured. This system LSI can be connected to an existing MPEG-2_CODEC_LSI to form a transcoder, which is useful.

  FIG. 10 is a block diagram showing an embodiment of a video signal recording apparatus to which the video signal conversion apparatus according to the present invention is applied (Embodiment 4). The video signal recording apparatus 900 includes a decoder 1000 and a recording unit 1100 in addition to the transcoder 100 described in the first embodiment. The input unit 200 is a tuner that receives a digital broadcast, for example. The encoded video signal input from the input unit 200 is supplied to the transcoder 100 and the decoder 1000. In the transcoder 100, transcoding is performed by the operation described in the first embodiment, and the converted encoded video signal is recorded in the recording unit 1100. The recording unit 1100 is, for example, an optical disk or a hard disk (DVD, HDD) recording device. On the other hand, the decoder 1000 decodes the input video signal and outputs the decoded image to the display unit 1200 connected to the outside. The display unit 1200 displays the decoded image.

  Here, it is assumed that transcoder 100 and decoder 1000 operate completely independently. By adopting this configuration, the encoded video signal input from the input unit 200 is monitored in real time by the display unit 1200, and at the same time, the encoded video signal is converted into a desired format by the transcoder 100. And can be stored in a recording medium. For example, it is possible to select a method of displaying high quality on a monitor and giving priority to long time recording.

  The video signal conversion device (transcoder) of the present invention can be applied not only to a single unit but also to various video devices. For example, in addition to the recording apparatus of the above-described embodiment, the present invention can be widely applied to an imaging apparatus that converts and records or transmits an imaging signal, and a transmission apparatus that converts and transmits a reception signal.

  Further, the compression encoding format described in the present embodiment is merely an example, and does not limit the application range of the present invention. The present invention can be applied even if the destination compression encoding scheme is the same as the source compression encoding scheme.

1 is a block diagram showing an embodiment of a video signal converter according to the present invention (Embodiment 1). FIG. The figure which shows the structure of a transport stream (TS) as a video signal. The figure which shows the structural example of the stream pre-processing part 101 in FIG. 2 shows a configuration example of the stream post-processing unit 106 in FIG. FIG. 2 is a diagram illustrating a configuration example of a stream analysis unit 102 and a stream information storage memory 103 in FIG. 1. The figure which shows the structural example of the video stream of MPEG-2 as a video signal. The figure which shows the example of control of the generated code amount by a present Example. The block diagram (Example 2) which shows the other Example of the video signal converter concerning this invention. The block diagram (Example 3) which shows the other Example of the video signal converter concerning this invention. FIG. 9 is a block diagram showing an embodiment of a video signal recording apparatus to which the video signal conversion apparatus according to the present invention is applied (Example 4).

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Video signal converter (transcoder), 101 ... Stream pre-processing part, 102 ... Stream analysis part, 103 ... Stream information storage memory, 104 ... Decoder, 105 ... Encoder, 1053 ... Decoding information processing part, 106 ... After stream Processing unit 107... Stream storage memory 501... Video decoder 502. Audio decoder 700 700 Encoder with stream processing function

Claims (1)

A first input unit to which a compression-coded first video signal is input;
A first output for outputting the first video signal input from the first input unit to the outside so as to be temporarily stored in a storage means connected to the outside and then supplied to an external decoding unit And
Read after a predetermined time stored in the connected storage means to the external second second video signal by decoding the first video signal at the outside of the decoding unit is input from the outside An input section;
A code amount analyzing unit for analyzing a change in a code amount of the first video signal input from the first input unit on a time axis;
A code amount storage unit for storing a change in the code amount analyzed by the code amount analysis unit;
An encoding unit that compresses and encodes the second video signal input from the second input unit into a third video signal in a different compression encoding method or a different signal format from the first video signal; ,
A second output unit for outputting the third video signal compressed and encoded by the encoding unit;
With
When the encoding unit compresses and encodes the second video signal, the encoding amount of the third video signal is based on the encoding amount stored in the encoding amount storage unit at a future position on the time axis. A video signal converter characterized by controlling the video signal.

Images