JP4672734B2 - Digital broadcast reception signal processing apparatus, signal processing method, signal processing program, and digital broadcast reception apparatus - Google Patents

Description

  The present invention relates to a technique for processing a plurality of digital broadcast reception signals, and particularly to a technique for processing a plurality of digital broadcast reception signals transmitted by simulcast.

  Japanese terrestrial digital broadcasting is defined by the ISDB-T (Integrated Services Digital Broadcasting Terrestrial) standard, and employs OFDM (Orthogonal Frequency Division Multiplexing) modulation and hierarchical transmission schemes. According to the ISDB-T standard, as schematically shown in FIG. 1, the transmission bandwidth of one channel (about 5.6 MHz) is divided into 13 subbands, that is, OFDM segments S1 to S13. Segments S1-S13 can be further divided into a maximum of three groups or hierarchies. Different transmission characteristics such as carrier modulation scheme, inner code coding rate, and time interleave length can be set for each layer, and different broadcast programs can be transmitted for each layer. For example, 12 OFDM segments S1 to S6, S8 to S13 are used to provide high quality high definition television (High Definition Television) for fixed receivers, or 12 OFDM segments S1 to S6 and S8 to S13 are provided. Divided into three layers to provide standard definition television (Standard Definition Television) for fixed receivers for each layer, or digital broadcasting for mobiles (simple video broadcasting using only one segment S7 arranged in the center) ). As a compression encoding / decoding standard, MPEG-2 (Moving Picture Experts Group phase 2) standard is adopted for digital broadcasting for fixed receivers, and H.264 is used for digital broadcasting for mobiles. H.264 (MPEG4 AVC) standard is adopted.

  At the beginning of full-scale digital terrestrial broadcasting, so-called simulcast (simultaneous broadcasting), in which broadcasters provide programs with the same content on both fixed receiver digital broadcasting and mobile digital broadcasting simultaneously. Broadcasting) is scheduled. Thus, several techniques have been proposed for receiving high-quality digital broadcasting as much as possible using a simulcast in a moving body such as a vehicle. For example, in Patent Document 1 (Japanese Patent Application Laid-Open No. 2004-312361) and Patent Document 2 (Japanese Patent Application Laid-Open No. 2004-166173), depending on reception quality, digital broadcasting for a mobile unit and digital broadcasting for a fixed receiver are used. A receiving device that automatically switches between receiving broadcasts is disclosed.

However, in the receiving apparatuses disclosed in Patent Documents 1 and 2, when the received broadcast is switched from one of the digital broadcast for the mobile body and the digital broadcast for the fixed receiver to the other, the image quality of the display video is Since it changes suddenly, there is a problem that visually unnatural images are displayed, which gives a sense of discomfort to those who listen to broadcast programs. For example, when switching from a high-resolution video of a digital broadcast for a fixed receiver to a low-resolution video of a digital broadcast for a moving body, image information such as subtitles, a human face image, or a pattern with a high spatial frequency may suddenly become unidentifiable. .
Japanese Patent Laid-Open No. 2004-312361 JP 2004-166173 A

  In view of the above, an object of the present invention is to provide a signal processing device, a signal processing method, and a signal that can generate a display image that is as high-quality or visually natural as possible from a plurality of digital broadcast reception signals provided by simulcast. It is to provide a processing program and a digital broadcast receiving apparatus.

  A signal processing apparatus according to an aspect of the present invention processes first and second encoded bit streams respectively generated from reception signals of both a first digital broadcast and a second digital broadcast that have been simulcast. An error monitoring unit that detects an error in the first encoded bitstream, obtains a first syntax element by analyzing the first encoded bitstream, and A first inverse in which an intra-frame encoded image of one encoded bit stream is decoded to reproduce a decoded image, and a predicted encoded image of the first encoded bit stream is decoded to reproduce a first difference image It should be added to the first difference image by executing the first motion compensated prediction using the first syntax element with reference to the transformation processing unit and the decoded image A first motion compensation unit that generates one predicted image; and a second syntax element obtained by analyzing the second encoded bitstream, and intra-frame encoding of the second encoded bitstream A second inverse transform processing unit that decodes an image to reproduce a decoded image and decodes a predicted encoded image of the second encoded bitstream to reproduce a second difference image; and the second syntax A syntax element conversion unit that converts the element into a syntax element suitable for the image format of the first encoded bitstream, and a second motion using the converted syntax element with reference to the decoded image A second motion compensator that generates a second predicted image to be added to the second difference image by executing compensated prediction; and when the error is not detected, the first difference When the error is detected, the second predicted image is added to the second difference image while the first decoded image reproduced by adding the first predicted image to the image is output. And an output control unit that outputs the reproduced second decoded image instead of the first decoded image.

  A digital broadcast receiving apparatus according to an aspect of the present invention is a digital broadcast receiving apparatus that processes reception signals of both a first digital broadcast and a second digital broadcast that have been simulcast. A demodulating circuit for generating first and second encoded bit streams from the received signal of the digital broadcasting, and the signal processing device.

  A signal processing method according to an aspect of the present invention processes first and second encoded bit streams respectively generated from reception signals of both a first digital broadcast and a second digital broadcast that have been simulcast. A signal processing method comprising: (a) detecting an error in the first encoded bitstream; and (b) analyzing the first encoded bitstream to obtain a first syntax element. And (c) decoding an intra-frame encoded image of the first encoded bitstream to reproduce a decoded image, and decoding a predictive encoded image of the first encoded bitstream to obtain a first And (d) performing the first motion compensated prediction using the first syntax element with reference to the decoded image, thereby reproducing the first difference image. Generating a first predicted image to be added to the image; (e) analyzing the second encoded bitstream to obtain a second syntax element; and (f) the second Decoding an intra-frame encoded image of the encoded bitstream to reproduce a decoded image, and decoding a predictive encoded image of the second encoded bitstream to reproduce a second difference image; g) generating a second predicted image to be added to the second difference image by executing a second motion compensated prediction using the transformed syntax element with reference to the decoded image; (H) When the error is not detected, the first decoded image reproduced by adding the first predicted image to the first difference image is output. On the other hand, when the error is detected, the first difference image is output. Difference of two A step of outputting in place the second decoded image reproduced by adding the second predictive image to the image on the first decoded image, those having a.

  A signal processing program according to an aspect of the present invention includes first and second codes generated from received signals of both a high-quality first digital broadcast and a low-quality second digital broadcast that are simulcast, respectively. A signal processing program for causing a processor to execute processing of a coded bitstream, wherein the processing analyzes an error detection process for detecting an error in the first coded bitstream, and analyzes the first coded bitstream A first analysis process for obtaining a first syntax element, decoding an intra-frame encoded image of the first encoded bitstream to reproduce a decoded image, and the first encoded bitstream A first inverse transform process for decoding a predictive encoded image and reproducing a first difference image, and using the first syntax element with reference to the decoded image A first motion compensation process for generating a first predicted image to be added to the first difference image by executing a first motion compensation prediction; and a second motion analysis by analyzing the second encoded bitstream A second analysis process for obtaining a syntax element; an intra-frame encoded image of the second encoded bitstream; a decoded image is reproduced; and a predicted encoded image of the second encoded bitstream The second inverse transform process for reproducing the second difference image and the second motion compensated prediction using the transformed syntax element with reference to the decoded image and executing the second motion compensation prediction A second motion compensation process for generating a second predicted image to be added to the difference image, and a first reproduced by adding the first predicted image to the first difference image and reproducing when the error is not detected Output decoded image On the other hand, when the error is detected, output control is performed to output the second decoded image reproduced by adding the second predicted image to the second differential image instead of the first decoded image. Processing.

FIG. 1 is a diagram illustrating 13 OFDM segments. FIG. 2 is a functional block diagram showing a schematic configuration of the digital broadcast receiving apparatus according to the first embodiment of the present invention. FIG. 3 is a functional block diagram schematically showing configuration examples of the first decoder and the second decoder. FIG. 4 is a diagram for explaining an example of vector information conversion processing. FIG. 5 is a diagram for explaining another example of the vector information conversion process. FIG. 6 is a functional block diagram showing a schematic configuration of the digital broadcast receiving apparatus according to the second embodiment of the present invention. FIG. 7 is a functional block diagram showing a schematic configuration of the decoder block of the second embodiment.

Explanation of symbols

1A, 1B Digital broadcast receiver 2 Receiver circuit (front end)
3A, 3B Signal processing circuit 10 Antenna 11 Tuner 12 Demodulation circuit 20 Error monitoring unit 21 Demultiplexer (DMUX)
22 first decoder 23 second decoder 24 signal output unit 25 image conversion unit 26 syntax element conversion unit 27 control unit 28 input unit 29 decoder block

BEST MODE FOR CARRYING OUT THE INVENTION

  This application uses Japanese Patent Application No. 2005-349352 as the basis of the priority claim, and the contents of the basic application are incorporated in the present application.

  Hereinafter, various embodiments according to the present invention will be described.

First embodiment

  FIG. 2 is a functional block diagram showing a schematic configuration of the digital broadcast receiving apparatus 1A according to the first embodiment of the present invention. The digital broadcast receiving apparatus 1A includes an antenna 10, a receiving circuit (front end) 2, and a signal processing circuit 3A. The reception circuit 2 includes a tuner 11 and a demodulation circuit 12. The signal processing circuit 3A includes an error monitoring unit 20, a demultiplexer (DMUX) 21, a first decoder 22, a second decoder 23, a signal output unit 24, an image conversion unit 25, a syntax element conversion unit 26, a control unit 27, and An input unit 28 is provided. The control unit (output control unit) 27 can individually control the operations of the error monitoring unit 20, the first decoder 22, the second decoder 23, and the signal output unit 24.

  Referring to FIG. 2, the tuner 11 selects a broadcast station to be received by the antenna 10 in accordance with an instruction from the control unit 27, converts the frequency of the received broadcast wave signal, and generates an OFDM (Orthogonal Frequency Division Multiplexing) signal. Generate. The demodulation circuit 12 performs an FFT (Fast Fourier Transform) on the OFDM signal from the tuner 11 to generate an OFDM demodulated signal, and further performs decoding processing such as deinterleaving, demapping, and error correction on the OFDM demodulated signal. To generate a transport stream TS. The tuner 11 may have a function of receiving one or a plurality of broadcasts among terrestrial digital broadcasts, satellite digital broadcasts, and cable digital broadcasts. It may have a function of receiving a digital broadcast transmitted via the computer network.

  The demultiplexer 21 hierarchically divides the transport stream TS supplied from the demodulation circuit 12 via the error monitoring unit 20, and from the transport stream TS, an elementary stream ES1 for video of digital broadcasting for fixed receivers The video elementary stream ES2 for digital broadcasting for mobiles is separated. One separated elementary stream ES1 is supplied to the first decoder 22, and the other elementary stream ES2 is supplied to the second decoder. In the present embodiment, one elementary stream ES1 is a high-quality compression code bit string, and the other elementary stream ES2 is a relatively low-quality compression code bit string. In the case of Japanese terrestrial digital broadcasting, elementary stream ES1 conforms to the MPEG-2 (Moving Picture Coding Experts Group phase 2) standard (MPEG-2 standard). H.264 standard. Although the error monitoring unit 20 and the control unit 27 are separated, the error monitoring unit 20 and the control unit 27 can constitute the “error monitoring unit” of the present invention.

  The error monitoring unit 20 constantly monitors the transport stream TS from the demodulation circuit 12, detects an uncorrectable bit error in the transport stream TS, and indicates error information indicating the occurrence position of this bit error and the bit error rate. Is supplied to the control unit 27. Based on the error information from the error monitoring unit 20, the control unit 27 determines that an error exists if the bit error rate Ber for the elementary stream ES1 exceeds an allowable value (threshold), and the bit error rate is the allowable value. If it is below, it can be determined that there is no error. The allowable value can be appropriately set by the user operating the input unit 28. In this embodiment, it is preferable that the control unit 27 determines the presence / absence of an error based on the bit error rate Ber, but the present invention is not limited to this. For example, when the receiving circuit 2 detects the C / N ratio (carrier power to noise power ratio) of the received signal of digital broadcasting, the control unit 27 gives an error if the C / N ratio is less than a desired value. It is also possible to determine that it has occurred.

  The first decoder 22 analyzes the elementary stream (first encoded bit stream) ES1 at the time of entropy decoding, and thereby a parameter group indicating the syntax (rule) of the elementary stream ES1, that is, a syntax element Get (syntax elements). MPEG-2 and H.264 A compression coding / decoding standard such as H.264 defines a syntax (rule) for describing and decoding a structure of a coded or multiplexed bit string (bit stream). The second decoder 23 also analyzes the elementary stream (second encoded bit stream) ES2 at the time of entropy decoding, thereby obtaining the syntax element of the elementary stream ES2. Examples of such syntax elements include a motion vector, a quantization parameter, a macroblock type, and a reference frame index.

  The syntax element conversion unit 26 converts the syntax element acquired by the second decoder 23 into a syntax element including a parameter group suitable for the image format of the elementary stream ES1, and converts the converted syntax element. This is given to the first decoder 22. On the other hand, when the second decoder 23 operates in the motion compensation prediction mode (interframe prediction mode), the second decoder 23 generates a prediction image by executing motion compensation prediction and also generates a difference image to be added to the prediction image. . The image conversion unit 25 can generate a difference image by interpolation, resolution conversion, or the like based on the difference image supplied from the second decoder 23 and provide the difference image to the first decoder 22.

  FIG. 3 is a functional block diagram schematically showing each configuration example of the first decoder 22 and the second decoder 23. In the example of FIG. 3, when the control unit 27 does not detect an error, the first decoder 22 performs decoding based on the normal MPEG-2 standard, and when the control unit 27 detects an error, the first decoder 22 The decoder 22 uses the converted syntax element from the syntax element conversion unit 26 and the difference image from the image conversion unit 25 to perform H.264 decoding. Decryption based on the H.264 standard may be performed.

  As shown in FIG. 3, the first decoder 22 has an entropy decoding unit (ED) 30, an inverse quantization unit (IQ) 31, and an inverse orthogonal transform unit (IOT) 32 as a general MPEG-2 decoder configuration. , A first motion compensation unit (MC1) 34A, a frame buffer memory 35, and an adder 36. The hybrid motion compensation unit 34 includes a first motion compensation unit 34A and a second motion compensation unit 34B. The entropy decoding unit 30, the inverse quantization unit 31, and the inverse orthogonal transform unit 32 may constitute the “first inverse transform processing unit” of the present invention.

  While the first decoder 22 operates as an MPEG-2 decoder, the first switch 33A connects between the entropy decoding unit 30 and the hybrid motion compensation unit 34 according to the control signal SW1, while the second switch 33B. Connects between the inverse orthogonal transform unit 32 and the adder 36 in accordance with the control signal SW1.

  Further, the first decoder 22 is connected to the H.264 decoder. As a configuration conforming to the H.264 standard, a second motion compensation unit (MC2) 34B, an intra prediction unit (IP) 37, and a third switch 38 are provided. The hybrid motion compensation unit 34 includes a first motion compensation unit 34A that performs motion compensation prediction based on the MPEG-2 standard in response to the control signal SW1 when the control unit 27 does not detect an error. In response to the control signal SW1 when H. is detected. The second motion compensation unit 34B executes motion compensation prediction based on the H.264 standard. When the control unit 27 does not detect an error, the third switch 38 always connects the hybrid motion compensation unit 34 and the adder 36 according to the control signal SW2 from the control unit 27, while the control unit 27 detects an error. When the third switch 38 detects H.E. It is possible to selectively connect the adder 36 to either the intra prediction unit 37 or the hybrid motion compensation unit 34 according to the operation mode conforming to the H.264 standard. When the control unit 27 detects an error, the entropy decoding unit 30 changes the operation mode of the first decoder 22 based on the syntax element to H.264. The third switch 38 determines whether the motion compensation prediction mode or the intra prediction mode based on the H.264 standard should be selected, and the third switch 38 responds to switching information (not shown) of the operation mode supplied from the entropy decoding unit 30. Thus, the hybrid motion compensation unit 34 and the intra prediction unit 37 are switched.

  On the other hand, as shown in FIG. As the configuration of the H.264 decoder, an entropy decoding unit (ED) 40, an inverse quantization unit (IQ) 41, an inverse orthogonal transform unit (IOT) 42, a motion compensation unit (MC) 43, an intra prediction unit (IP) 44, a switch 45, a frame buffer memory 46, an adder 47, and a deblocking filter 48. The entropy decoding unit 40, the inverse quantization unit 41, and the inverse orthogonal transform unit 42 can constitute the “second inverse transform processing unit” of the present invention.

  Referring to FIG. 2, the signal output unit 24 selects one of the decoded image signal DO1 from the first decoder 22 and the decoded image signal DO2 from the second decoder 23 in accordance with an instruction from the control unit 27. The selected signal can be output as the output image signal CS.

  The operation of the digital broadcast receiving apparatus 1A having the above configuration will be described below. Whether or not the two digital broadcasts are simulcast can be determined by, for example, obtaining EPG (electronic program guide) information and determining the control unit 27 based on this information. When the control unit 27 determines that the simulcast digital broadcast is not received, the control unit 27 causes the signal output unit 24 to selectively output one of the decoded image signal DO1 and DO2 as the output image signal CS. Can do. Alternatively, the control unit 27 can cause the signal output unit 24 to selectively output one of the decoded image signals DO1 and DO2 as the output image signal CS according to the operation of the input unit 28 by the user. is there.

  The operation described below is based on the premise that the digital broadcast receiving apparatus 1A is receiving simultaneous broadcast digital broadcasts for fixed receivers and mobile broadcasts simultaneously. First, each operation of the first decoder 22 and the second decoder 23 when the control unit 27 does not detect an error will be described below. At this time, the first switch 33A of the first decoder 22 connects between the hybrid motion compensation unit 34 and the entropy decoding unit 40, and the second switch 33B connects between the adder 36 and the inverse orthogonal transform unit 32. Connect.

  The first decoder 22 decodes the elementary stream ES1 for each macroblock. For example, in the case of a 4: 2: 0 signal, the macroblock has a luminance signal (Y) of 16 × 16 pixels and two kinds of color difference signals (Cb, 8 × 8 pixels) corresponding to spatial positions of the luminance signals. Cr). The entropy decoding unit 30 performs entropy decoding such as variable length decoding on the elementary stream ES1 to generate a quantized coefficient, and analyzes the elementary stream ES1 to generate syntax. Get the element. The inverse quantization unit 31 generates a transform coefficient of DCT (Discrete Cosine Transform) by inversely quantizing the quantization coefficient in accordance with a quantization parameter which is one of syntax elements, and the inverse orthogonal transform unit 32 further converts these transforms. The coefficients are subjected to inverse orthogonal transform (for example, inverse DCT transform) to generate an image signal, and this image signal is supplied to the adder 36 via the second switch 33B.

  When the output of the entropy decoding unit 30 is an intra-coded image (intra-frame coded image), the first decoder 22 operates in the intra mode. At this time, the third switch 38 is connected to the hybrid motion compensation unit 34, but the hybrid motion compensation unit 34 does not output data. Therefore, the adder 36 outputs the image signal from the second switch 33B as it is to the outside as the decoded image signal DO1. The decoded image that is the output of the adder 36 is accumulated in the frame buffer memory 35.

  On the other hand, when the output of the entropy decoding unit 30 is a prediction encoded image (inter-frame coded image) by motion compensation prediction, the first decoder 22 operates in the motion compensation prediction mode. At this time, motion vector information is input from the entropy decoding unit 30 to the first motion compensation unit 34A via the first switch 33A. The first motion compensation unit 34A reads a reference image from the frame buffer memory 35, generates a predicted image of a region specified by a motion vector with 1/2 pixel accuracy (half-pel accuracy) from the reference image, and converts the predicted image into The signal is supplied to the adder 36 through the third switch 38. In addition, as a predicted image, (1) an image predicted in the forward direction from the previous reference image in time, or (2) an image predicted (interpolated) from two reference images in both the front and rear in time. There can be. The adder 36 reproduces the decoded image by adding the predicted image to the difference image input from the inverse orthogonal transform unit 32 via the second switch 33B. The reproduced decoded image is output to the outside as the decoded image signal DO1 and is simultaneously stored in the frame buffer memory 35.

  The second decoder 23 decodes the elementary stream ES2 for each macroblock. As the size of the macroblock, for example, four types of sizes of 16 × 16 pixels, 16 × 8 pixels, 8 × 16 pixels, or 8 × 8 pixels are designated, and the size of the subblock is 8 × 8 pixels, 8 × 4. Four types of sizes of pixels, 4 × 8 pixels, and 4 × 4 pixels may be designated. The entropy decoding unit 40 performs entropy coding such as variable length decoding on the elementary stream ES2 to generate a quantized coefficient, and analyzes the elementary stream ES2 to generate a syntax element. To get. The inverse quantization unit 41 inversely quantizes the quantization coefficient according to a quantization parameter which is one of syntax elements to generate DCT transform coefficients, and the inverse orthogonal transform unit 42 further converts these transform coefficients with integer precision. The image signal is generated by inverse orthogonal transformation, and the image signal is supplied to the adder 47 and the image conversion unit 25.

  When the output of the entropy decoding unit 40 is an intra-encoded image (an intra-frame encoded image), the second decoder 23 It operates in an intra mode based on the H.264 standard. At this time, the switch 45 is connected to the motion compensation unit 43, but the motion compensation unit 43 does not output data. Therefore, the adder 47 supplies the image signal from the inverse orthogonal transform unit 42 to the deblocking filter 48 as it is. The output of the adder 47 is also given to the intra prediction unit 44. The deblocking filter 48 filters the output of the adder 47 for the purpose of suppressing the occurrence of so-called block noise (distortion that occurs near the boundary between macroblocks), and the resulting filter signal is a decoded image signal. Output to the outside as DO2. The filter signal is stored in the frame buffer memory 46 to be used as a reference image.

  On the other hand, when the output of the entropy decoding unit 40 is a prediction encoded image (interframe encoded image) by motion compensation prediction, the second decoder 23 operates in the motion compensation prediction mode. At this time, the switch 45 is connected to the motion compensation unit 43. The motion compensation unit 43 receives motion vector information from the entropy decoding unit 40. The motion compensation unit 43 reads the reference image from the frame buffer memory 46, generates a predicted image of the area specified by the motion vector from the reference image with ¼ pixel accuracy (quarter-pel accuracy), and outputs the predicted image signal. The signal is supplied to the adder 47 through the switch 45. In addition to the reference images (1) and (2) that are the same as those of the first decoder 22, the reference image may be an image three frames before the reference image. Then, the adder 47 adds the difference image signal to the prediction image signal input from the motion compensation unit 43 via the switch 45, and gives the addition signal to the deblocking filter 48. The deblocking filter 48 outputs the filter signal to the outside as a decoded image signal DO2 and also supplies it to the frame buffer memory 35.

  On the other hand, when the output of the entropy decoding unit 40 is a prediction encoded image by intra prediction, the second decoder 23 operates in the intra prediction mode. At this time, the switch 45 is connected to the intra prediction unit 44. H. In the intra prediction of the H.264 standard, a predicted image is generated by interpolation from pixels adjacent to the current macroblock in the same frame. The intra prediction unit 44 generates a prediction image according to the prediction pattern specified by the syntax element, and provides the prediction image to the adder 47 via the switch 45. The adder 47 adds the difference image signal to the prediction image signal input from the intra prediction unit 44 via the switch 45, and gives the addition signal to the deblocking filter 48.

  Next, the operation of the signal processing circuit 3A when the control unit 27 detects an error will be described below. Referring to FIG. 3, at this time, the first switch 33A of the first decoder 22 connects between the syntax element conversion unit 26 and the second motion compensation unit 34B according to the control signal SW1, and at the same time, the second switch The switch 33B connects between the image conversion unit 25 and the adder 36 in accordance with the control signal SW1. Further, the second motion compensation unit 34B operates instead of the first motion compensation unit 34A. As a result, the second motion compensation unit 34B uses the syntax element from the syntax element conversion unit 26 instead of the syntax element from the entropy decoding unit 30 in the motion compensation prediction mode, so that the frame buffer memory A prediction image is generated from the high-quality reference image read out from 35. Furthermore, the intra prediction unit 37 generates a prediction image according to the prediction pattern specified by the syntax element supplied from the syntax element conversion unit 26 in the intra prediction mode.

  In the motion compensation prediction mode, the third switch 38 is connected to the hybrid motion compensation unit 34, and the second motion compensation unit 34B operates. Motion vector information is input to the second motion compensation unit 34B from the syntax element conversion unit 26 via the first switch 33A. The second motion compensation unit 34B reads the designated high-resolution reference image from the frame buffer memory 35, generates a predicted image of the region specified by the motion vector from the reference image, and outputs the predicted image to the third switch. The signal is supplied to the adder 36 through the control unit 38. The image conversion unit 25 supplies the difference image to be added to the predicted image to the adder 36 via the second switch 33B. The adder 36 adds the difference image signal from the second switch 33B to the prediction image signal, and outputs the addition signal to the outside as the decoded image signal DO1.

  Here, based on the difference image from the inverse orthogonal transform unit 42, the image conversion unit 25 generates a difference image that conforms to the image format of the elementary stream ES1. For example, in Japanese terrestrial digital broadcasting, when digital broadcasting for mobiles using 1 segment and high-definition broadcasting using 12 segments are simulcast, the resolution of high-definition broadcasting (1920 × 1080 pixels) is for mobiles. It is larger than the resolution of digital broadcasting (320 × 180 pixels). The macroblock size of the difference image input to the image conversion unit 25 is H.264. In the case of 4 × 4 pixels compliant with the H.264 standard, the image conversion unit 25 can generate a difference image having a macroblock size of 24 × 24 pixels according to the resolution ratio of both digital broadcasts.

  H. In the H.264 standard, a prediction image can be generated with a 1/4 pixel accuracy in motion compensation prediction, but the decoded image stored in the frame buffer memory 35 of the first decoder 22 is the frame buffer memory 46 of the second decoder 23. Therefore, the second motion compensation unit 34B may execute the motion compensation prediction with the accuracy of ½ pixel or lower than that. For example, in the case of Japanese terrestrial digital broadcasting, as described above, the resolution of high-definition broadcasting (1920 × 1080 pixels) is larger than the resolution of digital broadcasting for mobiles (320 × 180 pixels). Therefore, a high-quality decoded image can be reproduced even if motion compensation prediction is executed with an accuracy lower than the 1/4 pixel accuracy.

  A motion vector that is one of the syntax elements is generated by the syntax element conversion unit 26. FIG. 4 is a diagram illustrating a method of generating the motion vector MV1 used in the second motion compensation unit 34B. The entropy decoding unit 40 of the second decoder 23 uses the syntax element of the information of the motion vector mb1 as well as the information of the macroblock type that specifies the prediction method of the encoded macroblock mb1 for the encoded macroblock mb1. This is given to the conversion unit 26. The syntax element conversion unit 26 generates the motion vector MV1 by converting the motion vector mv1 in accordance with the format of the encoded macroblock MB1 to be processed by the first decoder 22. For example, the image H1 to which the encoded macro block mb1 belongs has a resolution (horizontal resolution and vertical resolution) that is different from that of the image M1 to which the encoded macro block MB1 belongs. What is necessary is just to convert into MV1.

  H. Due to the difference in the frame rate between the H.264 standard and the MPEG-2 standard, there may be no corresponding image H1 synchronized with the image M1. More specifically, for example, as shown in FIG. 5, among the continuous images M1, M2, and M3 of the MPEG-2 standard, the corresponding H.P. H.264 standard images H1 and H2 exist, but H.264 synchronized with the image M2. This is a case where no H.264 standard image exists. In such a case, the syntax element conversion unit 26 generates a motion vector mv2 by interpolation from the motion vectors mv1 and mv3 of the previous and subsequent images H1 and H2, and further encodes the motion vector mv2 belonging to the image M2. The motion vector MV2 can be generated by conversion according to the format of the macroblock MB2. At this time, the syntax element conversion unit 26 interpolates not only the motion vector but also other syntax elements such as a macroblock type.

  Next, in the intra-frame prediction mode, the intra prediction unit 37 operates instead of the second motion compensation unit 34B, and the third switch 38 is connected to the intra prediction unit 37. A syntax element indicating the prediction pattern (prediction mode) used for encoding is given to the intra prediction unit 37 from the syntax element conversion unit 26. The intra prediction unit 37 generates a prediction image according to the prediction pattern specified by the syntax element, and provides the prediction image signal to the adder 36 via the third switch 38. Then, the adder 36 adds the difference image signal given from the image conversion unit 25 via the second switch 33B to the prediction image signal from the third switch 38, and outputs the addition signal as a decoded image signal DO1. To do. Note that the first decoder 22 may include a filter corresponding to the deblocking filter 48 of the second decoder 23.

  Here, the intra prediction unit 37 matches the image format of the elementary stream ES1 with the H.264 format. Intra prediction according to the H.264 standard is executed. H. According to the H.264 standard, a prediction image can be generated in block units of 4 × 4 pixels or 16 × 16 pixels by intra prediction. For example, in the case of Japanese terrestrial digital broadcasting, as described above, the resolution of high-definition broadcasting (1920 × 1080 pixels) is larger than the resolution of digital broadcasting for mobiles (320 × 180 pixels). Therefore, the intra prediction unit 37 can generate a prediction image having a block size of 24 × 24 pixels according to the resolution ratio between the two digital broadcasts.

  Next, in the intra mode, the hybrid motion compensation unit 34 and the intra prediction unit 37 do not operate, and the hybrid motion compensation unit 34 does not output data. At this time, the image conversion unit 25 converts the resolution of the image signal supplied from the inverse orthogonal transform unit 42 of the second decoder 23 in accordance with the image format of the elementary stream ES1, and the converted image signal is converted to the second switch. 33B and the adder 36 can be output to the outside.

  As described above, the digital broadcast receiving apparatus 1A according to the first embodiment is capable of providing an elementary stream ES2 of the other digital broadcast even if an error occurs with respect to the elementary stream ES1 of the two digital broadcasts that have been simulcast. Is generated in accordance with the image format of the one elementary stream ES1, and a predicted image is generated by performing motion compensation prediction or intra prediction using the converted syntax element. it can. Therefore, it is possible to provide a visually natural image as much as possible.

  In addition, when the high-quality digital broadcast and the low-quality digital broadcast are simulcast, even if an error occurs in the high-quality digital broadcast, the second motion compensation unit 34B stores it in the frame buffer memory 35. A high-quality predicted image can be generated by referring to the high-quality decoded image and executing motion compensation prediction using the converted syntax element. In parallel with this, the image conversion unit 25 generates a difference image to be added to the predicted image based on the difference image supplied from the inverse orthogonal transform unit 42 of the second decoder 23. Therefore, it is possible to provide a high-quality and visually natural image as much as possible.

  For example, in mobile objects such as vehicles and PDAs (Personal Digital Assistants), digital broadcasts for mobile objects can often be received even if the reception status of digital broadcasts for fixed receivers deteriorates. Is expected. Even in such a case, the digital broadcast receiving apparatus 1A according to the present embodiment can provide a high-quality and visually natural video as much as possible.

Second embodiment

  Next, a second embodiment according to the present invention will be described. FIG. 6 is a functional block diagram showing a schematic configuration of the digital broadcast receiving apparatus 1B of the second embodiment. FIG. 7 is a functional block diagram showing a schematic configuration of the decoder block 29 of the digital broadcast receiving apparatus 1B. It should be noted that the components denoted by the same reference numerals between FIG. 2 and FIG. 6 and between FIG. 3 and FIG. 7 have substantially the same function and the same configuration and will not be described in detail.

  Referring to FIG. 6, the digital broadcast receiving apparatus 1B includes an antenna 10, a receiving circuit (front end) 2, and a signal processing circuit 3B. The reception circuit 2 includes a tuner 11 and a demodulation circuit 12. The signal processing circuit 3B includes an error monitoring unit 20, a demultiplexer (DMUX) 21, a decoder block 29, a control unit (output control unit) 27, and an input unit 28. The control unit 27 can individually control the operations of the error monitoring unit 20 and the decoder block 29. Similarly to the first embodiment, in the second embodiment, one elementary stream ES1 output from the demultiplexer 21 is a high-quality compressed code bit string of digital broadcasting for a fixed receiver, and the other elementary stream ES1. The stream ES2 is a compression code bit string of relatively low quality for mobile digital broadcasting, but is not limited to this.

  Referring to FIG. 7, the decoder block 29 includes a first decoder 22B, a second decoder 23B, a frame buffer memory 46, and a signal switching unit 50. The first decoder 22B includes, as a general MPEG-2 decoder, an entropy decoding unit (ED) 30, an inverse quantization unit (IQ) 31, an inverse orthogonal transform unit (IOT) 32, a first motion compensation unit ( MC) 34A and an adder 36. The operations of these components 30 to 32, 36 are the same as those of the corresponding components shown in FIG. The first motion compensation unit 34A illustrated in FIG. 7 performs the same operation as the first motion compensation unit 34A illustrated in FIG. 3, and performs motion compensation prediction using the syntax element acquired by the entropy decoding unit 30. By doing so, a predicted image is generated. The entropy decoding unit 30, the inverse quantization unit 31, and the inverse orthogonal transform unit 32 may constitute the “first inverse transform processing unit” of the present invention.

  On the other hand, the second decoder 23B is an H.264 decoder. As a configuration conforming to the H.264 standard, an entropy decoding unit (ED) 40, an inverse quantization unit (IQ) 41, an inverse orthogonal transform unit (IOT) 42, a second motion compensation unit (MC) 43B, an intra prediction unit (IP ) 44B, switch 45B, adder 47B, and deblocking filter 48. The entropy decoding unit 40, the inverse quantization unit 41, and the inverse orthogonal transform unit 42 can constitute the “second inverse transform processing unit” of the present invention.

  The second motion compensation unit 43B has the same configuration and the same function as the second motion compensation unit 34B of the first embodiment (FIG. 3), and the intra prediction unit 44B is the same as that of the first embodiment (FIG. 3). It has the same configuration and the same function as the intra prediction unit 37. The second decoder 23B further includes an image conversion unit 25 that converts the resolution of the output image of the inverse orthogonal transform unit 42 so as to match the resolution of the elementary stream ES1, and the second decoder 23B includes the thin decoder acquired by the entropy decoding unit 40. It has a syntax element conversion unit 26 that converts the tax elements so as to conform to the resolution of the elementary stream ES1. The image conversion unit 25 and the syntax element conversion unit 26 shown in FIG. 7 are also substantially the same in configuration and function as the image conversion unit 25 and the syntax element conversion unit 26 in the first embodiment (FIG. 3), respectively. It is what has.

  The first motion compensation unit 34A of the first decoder 22B and the motion compensation unit 43B of the second decoder 23B each refer to the same frame buffer memory 46, and read the reference image from the frame buffer memory 46 at the time of motion compensation prediction. .

  The signal switching unit 50 selects and selects one of the decoded image signal DO1 supplied from the first decoder 22B and the decoded image signal DO2 supplied from the second decoder 23B according to the switching control signal SW. The signal is supplied as an output image signal CS to an external device (not shown). As will be described later, when the control unit 27 does not detect an error, the signal switching unit 50 selects the decoded image signal DO1, and when the control unit 27 detects an error, the signal switching unit 50 selects the decoded image signal DO2. .

  The operation of the digital broadcast receiving apparatus 1B having the above configuration will be described below. The operation described below is based on the premise that the digital broadcast receiving apparatus 1B is receiving simultaneous broadcast digital broadcasts for fixed receivers and mobile broadcasts simultaneously.

  First, the operation of the decoder block 29 when the control unit 27 does not detect an error will be outlined below. At this time, in the first decoder 22B, the entropy decoding unit 30 analyzes the elementary stream ES1 to acquire syntax elements, and supplies the syntax elements to the motion compensation unit 34 and the inverse quantization unit 31. . At the same time, the entropy decoding unit 30 performs entropy decoding on the elementary stream ES1 to generate a quantized coefficient. The inverse quantization unit 31 inversely quantizes the quantization coefficient in accordance with a quantization parameter that is one of syntax elements to generate DCT transform coefficients, and the inverse orthogonal transform unit 32 performs inverse orthogonal transform on these transform coefficients. An image signal is generated by (for example, inverse DCT conversion), and this image signal is supplied to the adder 36.

  When the first decoder 22B operates in an intra mode for decoding an intra-coded image, the first motion compensation unit 34A does not output data. Therefore, the adder 36 gives the image signal from the inverse orthogonal transform unit 32 to the signal switching unit 50 as it is as the decoded image signal DO1. The signal switching unit 50 selects the decoded image signal DO1 as the output image signal CS according to the switching control signal SW. The decoded image signal DO1 is supplied from the signal switching unit 50 to the external device and stored in the frame buffer memory 46 at the same time.

  On the other hand, when the first decoder 22B operates in the motion compensated prediction mode for decoding the prediction encoded image, the first motion compensation unit 34A refers to the decoded image stored in the frame buffer memory 46, and the reference image Then, a predicted image of the area specified by the motion vector is generated with a 1/2 pixel accuracy, and the predicted image is given to the adder 36. The adder 36 reproduces the decoded image by adding the predicted image to the difference image from the inverse orthogonal transform unit 32. The reproduced decoded image is given to the signal switching unit 50 as a decoded image signal DO1. The decoded image signal DO1 is supplied from the signal switching unit 50 to an external device (not shown) and simultaneously stored in the frame buffer memory 46.

  Next, the operation of the decoder block 29 when the control unit 27 detects an error will be described below. At this time, in the second decoder 23B, the entropy decoding unit 40 performs entropy coding such as variable length decoding on the elementary stream ES2 to generate a quantized coefficient and simultaneously the elementary stream ES2. ES2 is analyzed to obtain syntax elements. These syntax elements are supplied to the syntax element conversion unit 26 and the inverse quantization unit 41. The inverse quantization unit 41 inversely quantizes the quantization coefficient according to a quantization parameter that is one of syntax elements to generate DCT transform coefficients, and the inverse orthogonal transform unit 42 further converts these transform coefficients with integer precision. The decoded image or difference image is reproduced by inverse orthogonal transformation, and the decoded image or difference image is given to the image conversion unit 25.

  Based on the image from the inverse orthogonal transform unit 32, the image transform unit 25 generates an image that conforms to the image format of the elementary stream ES1. For example, in the case of Japanese terrestrial digital broadcasting, as described above, the resolution of high-definition broadcasting (1920 × 1080 pixels) is larger than the resolution of digital broadcasting for mobiles (320 × 180 pixels). In such a case, the image conversion unit 25 can convert, for example, an image having a block size of 4 × 4 pixels into an image having a block size of 24 × 24 pixels in accordance with the resolution of high-definition broadcasting.

  When the second decoder 23B operates in an intra mode for decoding an intra-coded image, the adder 47B provides the decoded image from the image conversion unit 25 to the deblocking filter 48 as it is. The deblocking filter 48 performs filtering on the decoded image signal from the adder 47B, and supplies the filtered signal to the signal switching unit 50 as a decoded image signal DO2. The signal switching unit 50 selects the decoded image signal DO2 as the output image signal CS. The output image signal CS is supplied from the signal switching unit 50 to an external device (not shown) and simultaneously stored in the frame buffer memory 46.

  On the other hand, when the second decoder 23B operates in the motion compensation prediction mode, the switch 45B is connected to the second motion compensation unit 43B. At this time, the second motion compensation unit 43B uses the syntax element (motion vector information) converted by the syntax element conversion unit 26 to refer to the frame buffer memory 46 and the reference decoded by the first decoder 22B. An image is read out, a predicted image of a region specified by a motion vector is generated from the reference image, and the predicted image is given to the adder 47B via the switch 45B. The adder 47B adds the predicted image to the difference image from the image conversion unit 25 to generate a decoded image. This decoded image is provided to the signal switching unit 50 after being filtered by the deblocking filter 48. The signal switching unit 50 selects the decoded image signal DO2 from the deblocking filter 48 as the output image signal CS, and the output image signal CS is supplied to an external device (not shown) and simultaneously sent to the frame buffer memory 46. Accumulated.

  On the other hand, when the second decoder 23B operates in the intra prediction mode, the switch 45B is connected to the intra prediction unit 44B. At this time, the intra prediction unit 44B performs intra prediction using the syntax element (information indicating the prediction mode) converted by the syntax element conversion unit 26 to generate a prediction image, and the prediction image is switched to the switch 45B. To the adder 47B. The adder 47B adds the predicted image to the difference image from the image conversion unit 25 to generate a decoded image. This decoded image is provided to the signal switching unit 50 after being filtered by the deblocking filter 48. The signal switching unit 50 selects the decoded image signal DO2 from the deblocking filter 48 as the output image signal CS, and the output image signal CS is supplied to an external device (not shown) and simultaneously sent to the frame buffer memory 46. Accumulated.

  As described above, the digital broadcast receiving device 1B of the second embodiment, like the first embodiment, even if an error occurs with respect to one elementary stream ES1 of two simulcast digital broadcasts, The syntax element acquired from the elementary stream ES2 of the other digital broadcasting is converted in accordance with the image format of the one elementary stream ES1, and motion compensation prediction or intra prediction is performed using the converted syntax element. By executing this, a predicted image can be generated. Therefore, it is possible to provide a visually natural image as much as possible.

  In addition, when the high-quality digital broadcast and the low-quality digital broadcast are simulcast, even if an error occurs with respect to the high-quality digital broadcast, the second motion compensation unit 43B decodes with the first decoder 22B. A high-quality prediction image can be generated by referring to the high-quality decoded image stored in the frame buffer memory 46 and executing motion compensation prediction using the converted syntax element. In parallel, the image conversion unit 25 generates a difference image to be added to the predicted image based on the difference image supplied from the inverse orthogonal transform unit 42. Therefore, it is possible to provide a high-quality and visually natural image as much as possible.

  The various embodiments according to the present invention have been described above. All or part of the configuration of the signal processing circuits 3A and 3B in the first and second embodiments may be realized by hardware, or may be recorded on a recording medium such as a nonvolatile memory or an optical disk. It may be realized by another program (or program code). Such a program (or program code) can cause a processor such as a CPU to execute all or part of the processing of the signal processing circuits 3A and 3B.

Claims (14)

A signal processing apparatus for processing first and second encoded bit streams respectively generated from reception signals of both a first digital broadcast and a second digital broadcast that are simulcast,
An error monitoring unit for detecting an error in the first encoded bitstream;
The first encoded bitstream is analyzed to obtain a first syntax element, an intra-frame encoded image of the first encoded bitstream is decoded to reproduce a decoded image, and the first A first inverse transform processing unit that decodes a predictive encoded image of the encoded bitstream and reproduces a first differential image;
A first motion compensation unit that generates a first predicted image to be added to the first difference image by executing a first motion compensated prediction using the first syntax element with reference to a decoded image When,
The second encoded bitstream is analyzed to obtain a second syntax element, and an intra-frame encoded image of the second encoded bitstream is decoded to reproduce a decoded image, and the second A second inverse transform processing unit that decodes a predictive encoded image of the encoded bitstream and reproduces a second differential image;
A syntax element conversion unit that converts the second syntax element into a syntax element suitable for an image format of the first encoded bitstream;
A second motion compensation unit that generates a second predicted image to be added to the second difference image by executing a second motion compensated prediction using the converted syntax element with reference to the decoded image When,
When the error is not detected, the first decoded image that is reproduced by adding the first prediction image to the first difference image is output, and when the error is detected, the second difference is output. An output control unit that outputs a second decoded image reproduced by adding the second predicted image to an image instead of the first decoded image;
A signal processing apparatus comprising: The signal processing apparatus according to claim 1, further comprising a buffer memory that stores the decoded image reproduced from the intra-frame encoded image by the first inverse transform processing unit and the first decoded image,
The signal processing apparatus, wherein the second motion compensation unit performs the second motion compensation prediction with reference to a decoded image stored in the buffer memory. The signal processing device according to claim 1, further comprising an image conversion unit that converts the second difference image into a difference image that conforms to an image format of the first encoded bitstream,
The signal processing apparatus, wherein the second reproduced image is generated by adding the second predicted image to the difference image converted by the image conversion unit.   4. The signal processing apparatus according to claim 1, wherein the syntax element conversion unit is used for the second motion compensation prediction using the second syntax element. 5. A signal processing apparatus characterized by interpolating syntax elements to be performed.   5. The signal processing apparatus according to claim 1, wherein the syntax element conversion unit converts at least motion vector information as one of the syntax elements. 6. Signal processing device. The signal processing device according to any one of claims 1 to 5,
The first inverse transform processing unit includes: a first entropy decoding unit that performs entropy decoding on the first encoded bit stream; and a first inverse quantum that performs inverse quantization on an output of the first entropy decoding unit. And a first inverse orthogonal transform unit that performs inverse orthogonal transform on the output of the first inverse quantization unit,
The second inverse transform processing unit includes a second entropy decoding unit that performs entropy decoding on the second encoded bitstream, and a second inverse quantum that performs inverse quantization on an output of the second entropy decoding unit. And a second inverse orthogonal transform unit that performs inverse orthogonal transform on the output of the second inverse quantization unit,
The signal processing apparatus, wherein the first syntax element is acquired by the first entropy decoding unit, and the second syntax element is acquired by the second entropy decoding unit. 7. The signal processing apparatus according to claim 1, wherein the signal processing apparatus adds to the second difference image by performing intra-frame prediction using the converted syntax element. 8. An intra prediction unit that generates a third predicted image;
When the error is detected, the output control unit outputs a decoded image reproduced by adding the third predicted image to the second difference image instead of the first decoded image. A characteristic signal processing apparatus.   8. The signal processing apparatus according to claim 1, wherein the error monitoring unit measures a bit error rate of the first encoded bit stream, and the measured value is an allowable value. 9. A signal processing device that detects the error when the value exceeds the threshold.   9. The signal processing apparatus according to claim 1, wherein each of the first inverse transform processing unit and the first motion compensation unit is a process based on a first coding standard. And the second inverse transform processing unit and the second motion compensation unit each perform processing based on a second coding standard different from the first coding standard. Signal processing device.   10. The signal processing apparatus according to claim 9, wherein the first encoding standard is the MPEG-2 standard, and the second encoding standard is H.264. H.264 standard signal processing apparatus.   9. The signal processing apparatus according to claim 1, wherein the quality of the first digital broadcast is higher than that of the second digital broadcast. apparatus. A digital broadcast receiving apparatus that processes reception signals of both the first digital broadcast and the second digital broadcast that are simulcast,
Demodulating circuits for generating first and second encoded bit streams from the received signals of the first and second digital broadcasts, respectively;
A digital broadcast receiving apparatus comprising: the signal processing apparatus according to claim 1. A signal processing method for processing first and second encoded bit streams respectively generated from reception signals of both a first digital broadcast and a second digital broadcast that are simulcast,
(A) detecting an error in the first encoded bitstream;
(B) analyzing the first encoded bitstream to obtain a first syntax element;
(C) Decoding an intra-frame encoded image of the first encoded bitstream to reproduce a decoded image, and decoding a predictive encoded image of the first encoded bitstream to generate a first difference image Step to play,
(D) generating a first predicted image to be added to the first difference image by executing a first motion compensated prediction using the first syntax element with reference to a decoded image; ,
(E) analyzing the second encoded bitstream to obtain a second syntax element;
(F) Decoding an intra-frame encoded image of the second encoded bit stream to reproduce a decoded image, and decoding a predictive encoded image of the second encoded bit stream to generate a second difference image Step to play,
(G) generating a second predicted image to be added to the second difference image by executing a second motion compensated prediction using the converted syntax element with reference to the decoded image; ,
(H) When the error is not detected, the first decoded image reproduced by adding the first predicted image to the first difference image is output. On the other hand, when the error is detected, the first difference image is output. A second decoded image that is reproduced by adding the second predicted image to the difference image of 2 and is output instead of the first decoded image;
A signal processing method comprising: A signal for causing the processor to process the first and second encoded bit streams respectively generated from the received signals of the simulcast high-quality first digital broadcast and the low-quality second digital broadcast. A processing program, wherein the processing is
An error detection process for detecting an error in the first encoded bitstream;
A first analysis process of analyzing the first encoded bitstream to obtain a first syntax element;
Decode the intra-frame encoded image of the first encoded bitstream to reproduce the decoded image, and decode the predictive encoded image of the first encoded bitstream to reproduce the first difference image A first inverse transformation process;
First motion compensation processing for generating a first predicted image to be added to the first difference image by executing first motion compensated prediction using the first syntax element with reference to a decoded image When,
A second analysis process of analyzing the second encoded bitstream to obtain a second syntax element;
Decode the intra-frame encoded image of the second encoded bitstream to reproduce the decoded image, and decode the predictive encoded image of the second encoded bitstream to reproduce the second difference image A second inverse transformation process;
Second motion compensation processing for generating a second predicted image to be added to the second difference image by executing second motion compensated prediction using the converted syntax element with reference to the decoded image When,
When the error is not detected, the first decoded image that is reproduced by adding the first predicted image to the first difference image is output, and when the error is detected, the second difference is output. An output control process for outputting a second decoded image reproduced by adding the second predicted image to an image instead of the first decoded image;
A signal processing program comprising:

Images