JP2005175710A - Digital recording and reproducing apparatus and digital recording and reproducing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a digital recording and reproducing apparatus which reproduces only a broadcast program by cutting off commercial contents when reproducing the broadcast program from a recording medium. <P>SOLUTION: The digital recording and reproducing apparatus includes: an image processing means 12 for receiving MPEG data and detecting a video image changing part for segmenting broadcast contents and commercial contents; a write buffer 13 for attaching identification information denoting the detected video image changing part to the MPEG data and recording the resulting data to a hard disk 14; a read buffer 15 for reading the MPEG data recorded on a hard disk 14 to reproduce the MPEG data; an operation means 40 for selecting a broadcast program reproduction mode or a CM reproduction mode; and a control means 30 for controlling the read buffer 15 to reproduce the broadcast contents or the commercial contents on the basis of a mode selected by the operation means 40 and scene change information read from the hard disk 14. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention is a digital recording / reproducing apparatus and a digital recording / reproducing apparatus suitable for being applied to a hard disk built-in digital video recorder for recording terrestrial digital broadcast programs, communication satellite digital broadcast programs, etc., a video tape recorder (VTR), and the like. Regarding the method. Specifically, when MPEG data in which commercial content is inserted between broadcast contents is reproduced from a hard disk or the like, either the mode for reproducing the broadcast content or the mode for reproducing the commercial content, and the identification read from the hard disk or the like Control means for controlling to reproduce broadcast content or commercial content based on the information is provided, and at the time of data reproduction, the commercial content is cut so that only the broadcast program can be reproduced.

  Conventionally, a recording / reproducing apparatus that records a broadcast program on a recording medium such as a video tape and then reads and reproduces the broadcast program from the recording medium is often used. In recent years, a digital recording / reproducing apparatus that records a digital broadcast program on a recording medium such as a hard disk and then reads and reproduces the broadcast program from the recording medium is also being used. When recording digital data on a recording medium such as a hard disk, an image compression method called MPEG-2 is adopted. In order to increase the compression efficiency of the recorded data, this method takes the correlation between the frames before and after the digital data, extracts the changed video, discards the video that does not change, and sends the changed video. doing.

  Some of such recording / playback apparatuses have a function of skipping commercials inserted into broadcast programs. For example, when a terrestrial analog broadcast program is being played and a commercial is reached, when the user presses the skip button, a program that skips an integral multiple of 30 seconds is typical. In general, commercials are characterized in that one program is created so as to be an integral multiple of 15 seconds, and once switched to commercials, it is continuously broadcast for several minutes.

  In relation to the commercial cut function, Patent Document 1 discloses a recording / reproducing apparatus, a recording / reproducing method, and a commercial discriminating method. According to the recording / reproducing apparatus, the recording / reproducing unit, the determining unit, and the control unit are provided, and the recording / reproducing unit records the video signal and the audio signal on the recording medium. The discriminating unit discriminates a commercial range as a range in which a plurality of portions having a predetermined time interval between the changing portions where the video signal and the audio signal change are continuous. The control unit skips the commercial range based on the determination result by the determination unit, and controls the recording / reproducing unit so as to normally reproduce the portion where the portion corresponding to the predetermined time appears alone. The recording / reproducing means reproduces the video signal and the audio signal of the part where the part corresponding to the predetermined time appears independently from the recording medium under the control of the control means. In this way, the commercial range can be determined, and a video cut from the commercial can be reproduced.

  Patent Document 2 discloses a detection circuit for a non-commercial message period of an audio multiplex broadcast receiver. According to this detection circuit, the audio multiplex decoder, the comparison unit, and the discrimination means are provided, the audio multiplex mode of the television program is detected, and the comparison unit compares the main audio signal and the sub audio signal from the audio multiplex decoder for discrimination. Means is adapted to determine that the television broadcast is not a commercial message when there is a difference between the two signals. In this way, since the audio multiplex program and the CM period can be discriminated, the program content can be recorded without being interrupted by the CM when the VTR is recorded.

Japanese Examined Patent Publication No. 12-3098170 (page 2, Fig. 1) JP 59-0064981 A (page 5, FIG. 8)

  By the way, the digital recording / reproducing apparatus having the commercial cut function according to the conventional example has the following problems.

  i. According to Patent Document 1, a range in which a plurality of continuous time intervals of a portion where the video signal and the audio signal change is a predetermined time is determined as a commercial range. Therefore, even if the commercial range can be determined in a terrestrial analog broadcast program or the like, the technique of Patent Document 1 is applied as it is when recording / reproducing digital data transmitted by the MPEG-2 image compression method. With difficulty to do.

  ii. According to Patent Document 2, the main audio signal and the sub audio signal from the audio multiplex decoder are compared, and when there is a difference between the two signals, it is determined that the message is not a commercial message. This method works only when the audio mode of the broadcast program is different from the audio mode of the commercial program, and actually only works under specific conditions. Therefore, it is difficult to apply the technique of Patent Document 2 as it is when recording / reproducing digital data transmitted by the MPEG-2 image compression method.

  Therefore, the present invention solves such a conventional problem, and after recording a digital broadcast program on a recording medium, at the time of data reproduction, the commercial content can be cut and only the broadcast program content can be reproduced. It is an object of the present invention to provide a digital recording / reproducing apparatus and a digital recording / reproducing method.

  The above-described problem is an apparatus for recording digital data in which commercial content is inserted between broadcast contents on a recording medium, and reproducing the digital data from the recording medium, and inputting the digital data to broadcast content and commercial content An image processing means for detecting a video switching portion that delimits the image, a recording means for adding identification information indicating the video switching portion detected by the image processing means to digital data, and recording it on a recording medium, and this recording means Reproducing means for reading out and reproducing digital data from the recording medium recorded by the operation means, operating means operated to select either a mode for reproducing broadcast content or a mode for reproducing commercial content, and the operation means The information read from the selected mode and recording medium It is solved by a digital recording and reproducing apparatus characterized by a control means for controlling the reproducing device to reproduce the broadcast content or the commercial content based on the information.

  According to the digital recording / reproducing apparatus of the present invention, when recording digital data in which commercial content is inserted between broadcast contents on a recording medium, the image processing means inputs the digital data to input the broadcast content and the commercial content. Detects the video switching part that separates. For example, the digital data is image data in a frame screen structure and data stream format, and the image processing means compares the image data of the previous frame and the image data of the current frame to obtain a correlation, and obtains the correlation here. Based on the obtained correlation, a video switching portion that separates the broadcast content and the commercial content is extracted. The recording means adds the identification information indicating the video switching portion detected by the image processing means to the digital data and records it on the recording medium.

  On the premise of this, at the time of data reproduction, the operation means is operated so as to select either the broadcast content reproduction mode or the commercial content reproduction mode. The control unit controls the reproduction unit to reproduce the broadcast content or the commercial content based on the mode selected by the operation unit and the identification information read from the recording medium. The reproduction means reads and reproduces the digital data from the recording medium under the mode control of the control means.

  Therefore, after the broadcast program is recorded on the recording medium, it is possible to cut the commercial content and reproduce only the broadcast program at the time of data reproduction. Conversely, only commercial content can be extracted and played back. Depending on the mode selection method, the commercial content can be reproduced before the broadcast content is reproduced, or the commercial content can be reproduced after the broadcast content is reproduced.

  A digital recording / reproducing method according to the present invention is a method of recording digital data in which commercial content is inserted between broadcast contents on a recording medium, and reproducing the digital data from the recording medium. A video switching part that divides broadcast content and commercial content is detected, identification information indicating the detected video switching part is added, digital data is recorded on a recording medium, and the digital recorded on the recording medium is recorded. The broadcast content or commercial content is reproduced based on the identification information from the data.

  According to the digital recording / reproducing method of the present invention, when digital data in which commercial content is inserted between broadcast contents is reproduced from the recording medium, the commercial content is recorded at the time of data reproduction after recording the broadcast program on the recording medium. It is possible to reproduce only the broadcast program by cutting.

  Conversely, only commercial content can be extracted and played back. Depending on the mode selection method, the commercial content can be reproduced before the broadcast content is reproduced, or the commercial content can be reproduced after the broadcast content is reproduced.

  According to the digital recording / reproducing apparatus and the digital recording / reproducing method of the present invention, when digital data in which commercial content is inserted between broadcast contents is reproduced from a recording medium, the broadcast content is reproduced or the commercial content is reproduced. Control means for executing control for reproducing broadcast content or commercial content based on any one of the modes to be performed and identification information read from the recording medium.

  With this configuration, after recording a broadcast program on a recording medium, it is possible to reproduce only the broadcast program by cutting the commercial content during data reproduction. Conversely, only commercial content can be extracted and played back. Depending on the mode selection method, the commercial content can be reproduced before the broadcast content is reproduced, or the commercial content can be reproduced after the broadcast content is reproduced.

  Subsequently, an embodiment of a digital recording / reproducing apparatus and a digital recording / reproducing method according to the present invention will be described with reference to the drawings.

FIG. 1 is a conceptual diagram showing a configuration example of a digital recording / reproducing apparatus 100 as each embodiment according to the present invention.
In this embodiment, when digital data in which commercial content is inserted between broadcast contents is reproduced from a recording medium, a mode for reproducing the broadcast content or a mode for reproducing the commercial content is provided. The broadcast content or commercial content is controlled to be reproduced based on the identification information read from the recording medium, and only the broadcast program can be reproduced by cutting the commercial content during data reproduction. It is.

  FIG. 1 shows a digital recording / reproducing apparatus 100 that records digital data in which commercial content is inserted between broadcast contents on a recording medium and reproduces the digital data from the recording medium. The apparatus 100 is suitable for application to a hard disk built-in digital video recorder that records terrestrial digital broadcast programs, communication satellite digital broadcast programs, and the like.

  The digital recording / reproducing apparatus 100 has an input terminal 1. A digital interface circuit 11 is connected to the input terminal 1 to input MPEG2 standard MPEG data in a PES (Packetized Elementary Stream) format as an example of digital data. MPEG data is data including image data of a frame screen structure related to a digital broadcast program or the like, and is obtained by multiplexing video compression data and audio compression data.

  The digital interface circuit 11 is connected to an image processing means 12, a write buffer 13, and a switch 16 for switching between normal operation / recording / playback mode. The image processing means 12 is connected to the write buffer 13 and the control means 30 and operates to detect a video switching portion (hereinafter referred to as a scene change point) that inputs MPEG data and divides broadcast content and commercial content. For example, the image processing unit 12 receives the image processing control signal S1 from the control unit 30, and compares the MPEG data of the previous frame with the MPEG data of the current frame based on the image processing control signal S1 to establish a correlation. The scene change point that divides the broadcast content from the commercial content is extracted based on the correlation obtained here.

  In extracting the scene change points, the image processing means 12 first divides the data stream into video group (Group Of Pictures: GOP) units composed of I, B, and P pictures. In this example, when the period in which an I or P picture appears is M, when M = 2, one B picture is inserted between the I or P pictures, and when M = 3, A mode is adopted in which two B pictures are inserted between P pictures.

  Next, the image processing means 12 detects the first I picture in the GOP of the current frame from the MPEG data divided in GOP units. The image processing means 12 performs discrete cosine transform (DCT; hereinafter referred to as DCT) processing on the previously detected I picture of the GOP to obtain a DC value that is a luminance component, and also performs DCT processing on the I picture in the GOP of the previous frame. To obtain the DC value. The DC value is used to evaluate the luminance value at the time of scene change. Further, a difference between the DC value of the first I picture and the DC value of the I picture in the GOP of the previous frame is obtained, and a difference between the preset reference value and the calculated DC value is compared. Based on this comparison result Then, the image processing means 12 determines whether or not it is a scene change point.

  The image processing means 12 also performs scene change when the difference between the DC value of the first I picture in the GOP of the current frame and the DC value of the I picture in the GOP of the previous frame is equal to or greater than a preset reference value. Generate information. The scene change information is data indicating that the GOP of the current frame and the GOP of the previous frame are scene change points that divide broadcast content and commercial content.

  In this example, when the video display time of the commercial content is 15 seconds, which is an example of a unit second value, and becomes an integral multiple thereof, the correlation between the MPEG data of the previous frame and the MPEG data of the current frame cannot be obtained. When the point is a poor correlation point, the image processing means 12 compares the MPEG data of the previous frame with the MPEG data of the current frame and extracts the poor correlation point when extracting the scene change point of the commercial content. The time difference between the correlation deficient points is calculated, and the time difference value between the correlation deficient points calculated here is compared with an integer multiple of 15 seconds, and the time difference value and the integer multiple of 15 seconds match. Operates to search for commercial periods.

  The write buffer 13 is an example of a recording unit, and is connected to a hard disk (HDD) 14. The write buffer 13 receives the write control signal S2 from the control means 30, and adds scene change information to the MPEG data based on the write control signal S2 and records it on the hard disk 14. The hard disk 14 is an example of a recording medium and is composed of a magnetic disk, but a magneto-optical disk may be used. In addition, a DVD (Digital Video Devise), a magnetic tape, or the like may be used as the recording medium.

  A read buffer 15 is connected to the hard disk 14. The read buffer 15 constitutes an example of a reproduction unit, and receives a read control signal S3 from the control unit 30, and reads out and reproduces MPEG data from the hard disk 14 based on the read control signal S3.

  A switch 16 for switching between normal operation / recording / reproducing mode is connected to the read buffer 15. When the normal operation mode is selected, the switch 16 receives a mode switching signal S4 from the control means 30, and based on this mode switching signal S4, the contact c in the figure is connected to the a side, and the input terminal 1 MPEG data in which commercial content is inserted between the fetched broadcast content is selected. When the recording / playback mode is selected, the contact c is connected to the b side, and MPEG data relating to broadcast content, commercial (hereinafter also abbreviated as CM) content, and the like is read from the hard disk 14. .

  The normal operation / recording / reproduction mode switching switch 16 (contact c) is connected to the separation circuit 17. The separation circuit 17 receives the separation control signal S5 from the control means 30, and operates to separate the video compression data and the audio compression data from the MPEG data based on the separation control signal S5. The separation circuit 17 is connected to the video decoder 18 and the audio decoder 19.

  The video decoder 18 decodes and decodes the compressed video data, and performs digital / analog (A / D conversion) conversion on the decoded video data. The video decoder 18 outputs the analog video signal Vout after A / D conversion to a monitor or the like (not shown) via the output terminal 2. The audio decoder 19 decodes and decodes the audio compression data, A / D-converts the decoded audio data, and sends the analog audio signal Aout after A / D conversion via the output terminal 3 in the monitor. Output to the voice input unit.

  A control unit 30 is connected to the image processing unit 12, the write buffer 13, the read buffer 15, the switch 16, and the separation circuit 17 described above. An operation means 40 is connected to the control means 30. The operating means 40 is operated so as to select either the program playback mode or the CM playback mode. The program playback mode refers to an operation of playing back only broadcast content in which commercials are skipped. The CM playback mode refers to an operation of playing back only commercial content in which broadcast content is skipped.

  Further, the operation unit 40 has a function of selecting any one of a pre-CM playback mode in which commercial contents are collectively played before the broadcast content is played back, or a post-CM playback mode in which commercial contents are collectively played back after the broadcast content is played back. have. Although not shown, the operation means 40 includes a normal operation / recording playback mode selection push button, a playback instruction button, a program / CM playback mode selection push button, a pre / post playback mode selection push button, and playback. An instruction push button or the like is provided. An operation instruction signal S0 obtained by these push button operations is output from the operation means 40 to the control means 30. The operation means 40 is operated by the user.

  The control means 30 has a ROM 31, a RAM 32 and a CPU 33. The ROM 31 stores system program data for controlling the entire recording / reproducing apparatus. The system program data is not limited to the ROM 31 and may be stored in a flash memory such as an EEPROM (not shown). The RAM 32 is used as a work memory, and temporarily stores control commands such as normal operation / recording / playback mode switching, program playback mode, and CM playback mode. When the power is turned on, the CPU 33 reads the system program data from the ROM 31 or the like, starts the system, and controls the entire recording / reproducing apparatus based on the operation signal S0 from the operation means 40.

  The control unit 30 receives the operation instruction signal S0 from the operation unit 40, and outputs an image processing control signal S1 to the image processing unit 12 based on the operation instruction signal S0, thereby changing the scene change between the broadcast content and the commercial content. Control to detect points. Further, the control means 30 outputs a write control signal S2 to the write buffer 13 and controls the scene change information to be added to the MPEG data and recorded on the hard disk 14.

  Further, the control means 30 outputs a read control signal S3 to the read buffer 15, and reproduces broadcast content or commercial content based on various modes selected by the operation means 40 and scene change information read from the hard disk 14. (Hereinafter also referred to as mode selection control). Furthermore, the control means 30 outputs a mode switching signal S4 to the switch 16 to execute switching control between normal operation / recording / reproducing mode. The control means 30 outputs a separation control signal S5 to the separation circuit 17, and controls to separate the video compression data and the audio compression data from the MPEG data.

  2A to 2C are conceptual diagrams showing examples of compression of image data in the MPEG-2 format. One image shown in FIG. 2A is compressed in MPEG2 format, for example, and is converted to MPEG data in PES format and input to the digital interface circuit 11 through the input terminal 1 shown in FIG. According to the MPEG2 format, image compression is performed by removing redundancy in the space and time domains.

  In FIG. 2A, the image 41 is divided into a plurality of blocks 42. The block 42 has x pixels in the horizontal direction and y pixels in the vertical direction at the display position shown in FIG. 2B (x = y = 8). According to the MPEG2 format, image data is compressed by converting it into a spatial frequency (time domain) in units of blocks in the spatial domain. When the block 42 in the spatial domain is converted into a spatial frequency, a discrete cosine transform process (hereinafter referred to as DCT) is performed. The DCT process is performed, for example, in units of 8 × 8 pixel blocks. The block 43 after the DCT processing shown in FIG. 2C is expressed by a spatial frequency. The spatial frequency becomes higher as the DCT transform system moves in the horizontal direction and the vertical direction.

[For spatial domain]
In general, in general, when the spatial region is narrowed, the luminance level values of pixels are often close to each other. In such a case, the degree of change (spatial frequency) of the pixel value in a certain spatial region becomes relatively small, and when the image is converted into data in the spatial frequency region, the image data is biased toward the low frequency side. As a result, by assigning a smaller number of bits to the image data on the high frequency side, the image can be encoded with a smaller number of bits as compared to before the image compression as a whole.

[In the time domain]
FIG. 3 is a block diagram illustrating a configuration example of an information encoding device 200 with a motion compensation function. In general, a moving image is composed of images called a plurality of frames. In many cases, each image is similar between a certain frame and the immediately preceding frame which is an image 1/30 second before. According to the encoding example by the inter-frame prediction of MPEG data, the image data of the current frame (hereinafter referred to as the current frame) is encoded based on the immediately preceding frame (hereinafter referred to as the previous frame). According to this encoding example, only the difference between the image data of the previous frame and the image data of the current frame is extracted and encoded, and the number of bits can be greatly reduced.

  In addition, if the motion compensation function is used, the number of bits can be further reduced. 3 includes memories 44 and 45, a motion vector detection unit 46, a motion compensation unit 47, a difference extraction unit 48, and a prediction error output unit 49. According to the information encoding apparatus 200, the motion vector detection unit 46 receives the current frame image data from the memory 44 and the previous frame image data from the memory 45, and the motion vector detection unit 46 It works to detect moving objects. Furthermore, the motion compensation unit 47 receives the motion detection data output from the motion vector detection unit 46 and the image data of the previous frame, and the motion compensation unit 47 operates to predict the motion.

  The difference extraction unit 48 extracts a difference between the motion prediction result data output from the motion compensation unit 47 and the image data of the current frame. The prediction error output unit 49 operates to output a prediction error from the difference extraction unit 48 based on the difference. Thereby, the accuracy of prediction can be improved.

  4A and 4B are configuration diagrams showing an example of frame rearrangement of MPEG data. In general, a frame screen structure of MPEG data (hereinafter referred to as an MPEG frame) includes three types of pictures (I, B, and P) having different data amounts, and an image is configured by combining these. In the example shown in FIG. 4A, when M = 3, the I, B, B, P, B, B, and P pictures of frame numbers # 0 to # 6 are arranged in order in the direction of motion compensation prediction. MPEG data is configured.

  The I picture is a screen that can be generated without the image data of the previous and subsequent frames. A P picture is a screen obtained by performing inter-frame prediction in one direction from a past frame and encoding a difference. The amount of data is intermediate between the I picture and the B picture. The B picture is a screen obtained by performing inter-frame prediction in two directions from two past and future frames and encoding the difference. The amount of data is the smallest compared to other pictures. Since a B picture uses a future frame for prediction, the future frame used for prediction must be encoded prior to the encoding, and the rearrangement of the frames is necessary at the time of encoding. .

  According to the MPEG data frame rearrangement example shown in FIG. 4B, the P picture with the frame number # 3 is replaced behind the I picture with the frame number # 0, and the frame numbers # 1 and # 2 are placed behind the P picture. The B picture is moved. The P picture of frame number # 6 is moved behind the B picture of # 2, and the B pictures of frame numbers # 4 and # 5 are moved behind the P picture. Thus, MPEG data having a frame screen structure input to the digital interface circuit 11 through the input terminal 1 shown in FIG.

  5A to 5F are conceptual diagrams showing examples of the hierarchical structure of MPEG data. In general, the MPEG-2 format used in a TV broadcast system, a digital video disc (DVD), or the like forms an image by the three types of pictures I, P, and B described with reference to FIG. Of these, only I pictures can be decoded without image data of other pictures. Therefore, an operation that breaks the relationship between the previous and next frames, such as random access, is realized by accessing the I picture. Also, the MPEG data has a hierarchical structure as shown in FIGS. 5A to 5F by a plurality of pictures such as an I picture as a decoding reference, a P picture, and a B picture.

  FIG. 5A shows the sequence layer. In the sequence layer, n programs are described. The program is accessed randomly. FIG. 5B shows GOP layers constituting the lower layer of program n. In the GOP layer, three types of pictures I, P, and B shown in FIG. 4 are described. FIG. 5C shows a picture layer constituting a lower layer of the GOP layer. In the picture layer, image data related to the I, P, or B picture shown in FIG. 4 is described. Image data is described in units of frames or fields.

  FIG. 5D shows slice layers constituting the lower hierarchy of the picture layer. In the slice layer, image data obtained by slicing (dividing) the screen related to the I, P, or B picture shown in FIG. 4 in the horizontal direction is described. FIG. 5E shows a macroblock layer constituting a lower layer of the slice layer. In the macroblock layer, image data obtained by dividing the screen constituting the slice layer in the vertical direction is described. Image data is described in prediction units. This image data includes a luminance signal Y and color difference signals (Cr, Cb).

  FIG. 5F shows a block layer constituting a lower hierarchy of the macroblock layer. In the block layer, the luminance signal Y of each block is described in units of DCT. In this example, the I picture in the GOP of the previous frame is DCT processed to obtain a DC value from the luminance signal Y constituting the block layer, and the DC value is similarly obtained by performing DCT processing on the I picture in the GOP of the current frame. The difference between the two I pictures and the DC value of the I picture is obtained, the preset reference value is compared with the calculated DC value, and whether or not the scene change point is determined based on the comparison result. It is made to decide.

  In general, in the MPEG-2 format, a field or a frame can be selected as an encoding unit. When the image motion is small, the correlation between the two fields is strong, so it is advantageous to use a frame structure that uses the entire frame. When the image motion is intense, the correlation between the fields is weakened. It is advantageous to process the field structure separately.

  When MPEG data has a frame structure, a field / frame adaptive prediction method is adopted. According to this method, frame prediction and field prediction are switched and applied in motion prediction. This is because the encoding / compression process can be effectively executed by using frame prediction for a macroblock with little motion and using field prediction for a macroblock with fast motion.

  As for the encoding / compression process, a field / frame adaptive DCT processing system is adopted in addition to the field / frame adaptive prediction system. According to this DCT processing method, a method of switching between a first mode for executing DCT processing for each field on image data in a macroblock and a second mode for performing DCT processing as a frame structure by alternately mixing fields. Is taken. In the case of a frame structure, the second mode can be selected on a macroblock basis. This digital recording / reproducing apparatus 100 supports a field / frame adaptive prediction method and a field / frame adaptive DCT processing method set independently.

  On the other hand, when the MPEG data has a field structure, since the image data is encoded separately for each field, the field prediction method and the field DCT processing method are always used, but the size of the prediction block is 16 × 16. You can choose from two types: pixels and 16x8 pixels. This corresponds to selecting a balance between the information amount of the motion vector and the prediction efficiency. The digital recording / reproducing apparatus 100 supports a field prediction method and a field DCT processing method set independently.

  As described above, according to the MPEG-2 format, it is possible to detect a correlation between two fields in an MPEG frame. This correlation detection result is not data compression, but MPEG data is divided into frame units, and the correlation detection result is used as a marker at the time of content reproduction, so that the commercial content is inserted between broadcast contents. When data is played back from the hard disk 14, broadcast content or commercial content can be played back based on either the CM playback mode or the broadcast program playback mode and the scene change information read from the hard disk 14.

  In this digital recording / reproducing apparatus 100, when MPEG data is digitally recorded, frame correlation detection is performed in the MPEG data, and points where no correlation can be obtained are listed, and a time difference between these points is calculated for 15 seconds. This part is made to be selectable when a search is made for a thing that matches an integer multiple of and is played back.

  FIG. 6 is a block diagram illustrating an internal configuration example of the image processing unit 12. The image processing means 12 shown in FIG. 6 includes a separation unit 21, a GOP analysis unit 22, an I picture extraction unit 23, a file A creation unit 24, a DCT analysis unit 25, a video change point detection unit 26, a file B creation unit 27, a file The C creation unit 28, the comparison unit 29, and the final file creation unit 210 are configured.

  The separator 21 operates to input MPEG data in PES format and separate (filter) it into a packet element stream (PES) and a macroblock (MB). PES format MPEG data includes a sequence layer, a GOP layer, a picture layer, a slice layer, a macroblock layer, and a block layer. A GOP analysis unit 22 and a DCT analysis unit 25 are connected to the separation unit 21.

  The GOP analysis unit 22 receives the PES from the separation unit 21 and performs GOP analysis processing. For example, the GOP analysis unit 22 divides the PES into GOP units and detects the presence or absence of the GOP layer in the current frame. Based on this detection presence / absence result, it is determined whether or not a GOP layer exists in the current frame. The GOP layer is composed of I, B, and P pictures. An I picture extraction unit 23 is connected to the GOP analysis unit 22. The I picture extraction unit 23 extracts the first I picture from the GOP layer immediately after the GOP sequence header (SH) of the current frame.

  A file A creation unit 24 is connected to the I picture extraction unit 23. The file A creating unit 24 creates a file A by analyzing a PTS (Presentation Time Stamp). For example, the file A creation unit 24 performs a discrete cosine transform process on the previously detected I picture of the GOP to obtain a DC value, compares the value with the previous value, and compares the PTS or the point that exceeds a certain value. The file number is created by checking the packet number. In other words, when recording MPEG data in a data stream format such as digital broadcasting on the hard disk 14, the file A compares the DC value of the I picture of the current frame with the DC value of the I picture of the previous frame, and is constant. The PTS or packet number at the point where the value is exceeded is examined and filed as a list.

  This DC value is checked every time a DCT process of an I picture of each GOP layer is performed by TS or PS filtering processing simultaneously with the recording of MPEG data. This corresponds to the case where the analog video signal is encoded into MPEG data, and the DC value is checked for each DCT process of each I picture. The file A creation unit 24 is connected to the comparison unit 29.

  The DCT analysis unit 25 connected to the above-described separation unit 21 inputs the separated macroblock, performs a discrete cosine transform process, and performs a DCT analysis process on the macroblock. A video change point detector 26 is connected to the DCT analyzer 25. The video change point detection unit 26 operates to detect a video change point of the video content and find a scene change point at which the video is switched from the broadcast content to the commercial content.

  A file B creation unit 27 is connected to the video change point detection unit 26. The file B creation unit 27 creates a file B by analyzing the packet time stamp (PTS). File B is a case where the GOP layer exists in the PTS, and when the PES is divided into GOP units, each I picture existing at the beginning of the GOP layer is checked, its PTS or packet number is checked, It is filed as a list.

  The file B creation unit 27 determines whether an I picture existing at the very beginning of the GOP layer is detected. Based on the determination result, DCT processing is performed to detect the DC value of the I picture. Then, the file B creation unit 27 compares the DC value obtained by DCT processing of the I picture with a preset reference value, and determines whether or not the DC value of the I picture is greater than or equal to the reference value. When the DC value of this I picture is greater than or equal to the reference value, control is performed so as not to create a file. If the DC value of the I picture is less than the reference value, the PTS or packet number is checked and described in file B as a list of correlation NG (change) points.

  This correlation NG (change) point is a candidate for a scene change point, and a scene change point is obtained based on this correlation NG (change) point. The calculation is performed by DCT processing the I picture in the GOP of the previous frame to obtain a DC value, DCT processing of the I picture in the GOP of the current frame to obtain the DC value, and the difference between the DC values of the two I pictures. And a difference between the reference value set in advance and the calculated DC value is compared, and based on the comparison result, it is determined whether or not it is a scene change point.

  If the difference between the DC value of the first I picture in the GOP of the current frame and the DC value of the I picture in the GOP of the previous frame is equal to or greater than a preset reference value, the file B creation unit 27 Scene change information indicating that the GOP and the GOP of the previous frame are scene change points that divide broadcast content and commercial content is generated.

  A file C creation unit 28 is connected to the file B creation unit 27. The file C creation unit 28 assumes a commercial section (period) and creates a file C describing scene change information for specifying the section. File C is used to estimate the commercial period based on the difference between the DC value of the I picture of the current frame and the DC value of the I picture of the previous frame. According to the current commercial content, a display time that is an integral multiple of 15 seconds is allocated.

  For example, the file C creation unit 28 reads a list of correlation NG (change) points from the file B, and calculates a time difference TW from one correlation NG point (correlation point) to the other continuous correlation NG point. The file C creation unit 28 determines whether the time difference TW is an integral multiple of 15 seconds and is, for example, 60 seconds or less. If the time difference TW is not an integral multiple of 15 seconds and less than 60 seconds, the calculation interval for calculating the time difference TW is advanced by one. When the time difference TW is an integer multiple of 15 seconds and 60 seconds or less, the PTS or packet number at the start and end of the calculation is checked, and is described in the file C as a list of scene change points related to the display period of the commercial content. .

  A comparison unit 29 is connected to the file C creation unit 28. The comparison unit 29 compares the list contents (file B) of the correlated NG points that are candidates for scene change points with the checked contents of the correlated NG points checked by the file C creating unit 28, and determines the commercial period. The detection accuracy is increased. For example, the file A created by the file B creation unit 27 by switching to the GOP unit and the file C listed by the file C creation unit 28 are input, and the calculation start and calculation end extracted by the file C creation unit 28 The PTS or packet number is verified.

  Since the commercial content is different from the broadcast content, when the commercial content is inserted into the broadcast content, it is always divided into GOP units. Utilizing this property, for example, an AND logic operation is applied to the description content of the file A and the description content of the file C, so that the detection accuracy of the commercial period can be improved.

  The comparison unit 29 is connected to the final file creation unit 210. The final file creation unit 210 creates a final file for executing the program playback mode, the CM playback mode, the pre-CM playback mode, or the post-CM playback mode. In this file, data for reading scene change information (ID signal) from the file C is described together with the file C, and these are stored in the hard disk 14 together with the MPEG data.

  7A and 7B are time charts showing an example of scene change point extraction in the image processing means 12. In this example, scene change points between broadcast content and commercial content are extracted from MPEG data in which commercial content is inserted between broadcast content.

  The horizontal axis shown in FIGS. 7A and 7B is PTS or packet number (time t). In the drawing, block data Din indicated by “IBBP” in □ is obtained by dividing MPEG data into the frames (GOP units). “I” is an I picture, “B” is a B picture, and “P” is a P picture. In FIG. 7A, an upward arrow is a correlation NG point, and an upward arrow between broadcast content and commercial content indicates a scene change point.

  In this example, when a TV program or the like is recorded, a commercial period is found. First, the correlation between MPEG frames is taken and all correlation NG points are detected. Thereafter, a time difference TW (interval) between correlated NG points as described in FIG. 8 is examined, and a unit display time that is an integral multiple of 15 seconds, which is the current unit second value, is extracted.

  In FIG. 7A, when the frame correlation of MPEG data is detected, since there is no correlation between the commercial image and the TV image, it is always possible to detect no front-rear correlation at this image switching point. Can be defined as a point. If such correlation detection is continued, a correlation NG point can be detected again at the time of switching to the next commercial. This interval is always an integral multiple of 15 seconds for one program because of commercial transmission.

  In this example, the image processing means 12 first divides MPEG data into GOP units composed of I, B, and P pictures, and then from the MPEG data divided into GOP units, the first I in the GOP of the current frame. Detect pictures. The image processing means 12 calculates the DC value by DCT processing the I picture of the GOP detected here, and calculates the DC value by DCT processing the I picture in the GOP of the previous frame. Then, as shown in FIG. 7B, a difference ΔDC between the DC value of the first I picture in the GOP of the previous frame and the DC value of the first I picture in the GOP of the current frame is calculated, and a preset reference value Vref is calculated. The calculated DC value difference ΔDC is compared, and the image processing means 12 determines whether or not the scene change point is based on the comparison result.

  In this example, when the difference ΔDC between the DC value of the first I picture in the GOP of the current frame and the DC value of the I picture in the GOP of the previous frame is equal to or greater than a preset reference value Vref, the image processing means 12 Then, it is found (determined) that the GOP of the current frame and the GOP of the previous frame are scene change points that divide broadcast content and commercial content. The scene change information indicating the scene change point is recorded on the hard disk 14 together with the MPEG data.

  In this example, the extraction of scene change information is not limited to the determination method based on the difference ΔDC between both DC values of the I picture, and may be the following determination method. That is, the MPEG data is divided into GOP units composed of I, B, and P pictures, and the first I picture in the GOP of the current frame is detected from the MPEG data divided into the GOP units, and the GOP detected here is detected. The DC value is calculated by DCT processing the I picture. The DC value of the first I picture in the GOP of the current frame may be compared with a preset reference value Vref ′, and it may be determined whether or not it is a scene change point based on the comparison result. .

  In this example, when the DC value of the first I picture in the GOP of the current frame is equal to or greater than a preset reference value Vref ′, the GOP of the current frame and the GOP of the previous frame are used as broadcast content and commercial content. It is found to be a scene change point that separates. The scene change information indicating the scene change point is recorded on the hard disk 14 together with the MPEG data in the same manner as in the previous example.

  FIG. 8 is a time chart showing an example of commercial period extraction in the image processing means 12. In this example, when a scene change point between broadcast content and commercial content is extracted in MPEG data in which commercial content is inserted between broadcast content, then a time difference between correlated NG points is examined in DCT processing. The unit display time in which the time difference is an integral multiple of 15 seconds is detected, and the commercial period is extracted based on the unit display time. The commercial content is composed of commercial A (CMA) and commercial B (CMB). The display time of commercial A is 30 seconds, and the display time of commercial B is 15 seconds.

  The horizontal axis shown in FIG. 8 is the PTS or packet number (time t). In the figure, block data Din indicated by “I” in □ is obtained by dividing MPEG data into GOP units. “I” is an I picture. A triangle is a correlation NG point. In this example, the calculation process proceeds such as stroke P1 to stroke P5... Shown in the calculation direction of the white arrow. In this calculation process, the calculation operation at the correlation NG point is repeatedly executed up to an integral multiple of 15 seconds and about 30 seconds (60 seconds).

  In FIG. 8, first, in the process P1, the #A section between the first correlation NG points in the figure is calculated, and it is determined whether or not the time difference of the #A section is an integral multiple of 15 seconds. If it is not an integral multiple of 15 seconds, the time difference in the #B section is calculated. Similarly, the #C section is calculated. Until such a section of an integral multiple of 15 seconds is found, the section is expanded and the search is continued. When the difference of 30 seconds or more cannot be obtained, the process proceeds to the process P2, and the same process is repeated for the processes P3, P4, and P5 by advancing the detection start part by one.

  In this example, when the #A section is calculated in the process P5, the time difference is 15 seconds, and when the #B section is calculated, it is 15 seconds × 2 times. The result of this calculation process coincides with the display time of commercial A (CMA). Thus, when a commercial period that is an integral multiple of 15 seconds is found, the search (investigation) is terminated for the #C interval. Accordingly, the calculation processes i, ii, iii and the like indicated by the crosses in the processes P6 and P7 are omitted. Of course, when a commercial period that is an integral multiple of 15 seconds is not found, the calculation process in the processes P6 and P7 is continued.

  9A to 9C are time charts showing commercial insertion sections and search examples thereof. FIG. 9A shows the position of a commercial (content) sandwiched between TV programs (broadcast content). According to the example of commercial insertion shown in FIG. 9A, four commercials A to D are inserted between TV program contents.

  In FIG. 9B, the commercial A has a display time of 15 seconds. Commercial B is displayed following commercial A, and the display time is 30 seconds. Commercial C is displayed following commercial B, and the display time is 15 seconds. The commercial D is displayed following the commercial C, and the display time is 15 seconds. In this example, the commercial period is the total display time of the four commercials A to D, and commercial A + B + C + D = 75 seconds.

  In this example, in order to extract the scene change point of the TV program and commercial A, in the DCT processing, the time difference TW between the correlated NG points is examined, and the portion that becomes an integral multiple of 15 seconds is detected. Estimated (principal processing). Naturally, it is assumed that there is a correlation NG point where MPEG data cannot be correlated even within one commercial program, and therefore, from one correlation NG point to the next correlation NG point is less than an integral multiple of 15 seconds. The time difference is continuously added to calculate a point that is an integral multiple of 15 seconds (exception processing).

  That is, as a special case, the section of adjacent blocks in one commercial may be less than an integral multiple of 15 seconds. For example, in the three blocks extracted in the commercial B having a display time of 30 seconds shown in FIG. 9C, the time difference TW between two correlated NG points is 7 seconds, TW ′ = 13 seconds, TW ″ = 7 seconds, etc. In this case, the calculation between correlated NG points is advanced by one, and in this example, an integer multiple of 15 seconds and 30 seconds (60 The calculation operation at the correlation NG point is repeatedly executed up to about (second).

  For example, in the commercial B shown in FIG. 9C, when the section (interval) detected as the marshalling period is less than an integral multiple of 15 seconds, the time difference TW ′ or TW ″ up to the next correlated NG point, etc. 30 seconds (= 7 seconds + 13 seconds + 7 seconds) is added to a value that is an integer multiple of 15 seconds, and the addition is about 2 to 3 minutes at maximum from the commercial content transmission standard. And

  Thus, since the commercial (content) continues for several minutes once playback starts, it is confirmed whether or not the extracted block is applicable to this period, and if it is OK, from the MPEG data recorded on the hard disk 14, this Scene change information (ID signal) for extracting only a part is created and recorded on the hard disk 14 simultaneously with the MPEG data.

  Next, the digital recording / reproducing method according to the present invention will be described. FIG. 10 is a flowchart showing an example of recording / reproducing in the digital recording / reproducing apparatus 100 according to the first embodiment.

  In this embodiment, MPEG data is MPEG data in a frame screen structure and data stream format, and when broadcast content such as a TV program is recorded on the hard disk 14 or the like, commercial content is inserted between the broadcast content. Using the correlation between the frames before and after the MPEG data, a commercial period in which the correlation between NG points is an integral multiple of 15 seconds is extracted, and when the MPEG data is reproduced from the hard disk 14, the block extracted at the time of recording is recorded. The user can freely select (designate) broadcast content or commercial content.

  In this example, it is assumed that a recording mode for recording MPEG data related to a broadcast program on the hard disk 14 is set. The control means 30 that has input the operation instruction signal S0 by the recording mode setting from the operation means 40 outputs the image processing control signal S1 to the image processing means 12 based on the operation instruction signal S0, and the broadcast content, the commercial content, Controls to detect scene change points that delimit

  Under these MPEG data recording / playback conditions, first, the MPEG data is input to the digital interface circuit 11 in step E1 of the flowchart shown in FIG. In step E2, the image processing unit 12 detects a scene change point that divides the broadcast content and the commercial content based on the image processing control signal S1, and extracts a commercial period. At this time, the image processing means 12 compares the MPEG data of the previous frame with the MPEG data of the current frame to obtain a correlation, and a scene change that divides the broadcast content and the commercial content based on the obtained correlation. Extract points.

  Specifically, the flowchart (subroutine) shown in FIG. 11 is called to create file A, the flowchart (subroutine) shown in FIG. 12 is called to create file B, and the flowchart (subroutine) shown in FIG. To create file C. Then, scene change information indicating a scene change point based on the previously created files A to C is added to the MPEG data in step E3 and recorded on the hard disk 14. At this time, the control means 30 outputs a write control signal S2 to the write buffer 13. The write buffer 13 receives the write control signal S2 from the control means 30, and adds scene change information to the MPEG data based on the write control signal S2 and records it on the hard disk 14.

  Thereafter, in step E4, the control unit 30 inputs a normal operation / recording reproduction instruction, a reproduction instruction, and the like through the operation unit 40. The normal operation / recording / playback instruction or the like is input by the user operating the operation means 40. Here, when the recording / playback mode is selected, the contact c is connected to the b side, and MPEG data relating to broadcast content, commercial content, and the like is read from the hard disk 14.

  When the recording / playback mode is selected, the operation unit 40 is operated so as to select either the program playback mode or the CM playback mode. When the program playback mode is selected, only the broadcast content that skips the commercial is played. When the CM playback mode is selected, only commercial content in which broadcast content is skipped is played back. When the normal operation mode is selected, the switch 16 for switching the normal operation / recording / playback mode receives the mode switching signal S4 from the control means 30, and the contact c shown in FIG. The MPEG is selected in which commercial contents are inserted between the broadcast contents fetched from the input terminal 1 and connected. An operation instruction signal S0 obtained by operating the operation means 40 is output from the operation means 40 to the control means 30. If there is no reproduction instruction, the process proceeds to step E8.

  If there is a reproduction instruction in step E4 described above, the control unit 30 proceeds to step E5 and accepts the selection of the broadcast program reproduction mode or the CM reproduction mode. At this time, when the broadcast program playback mode is selected, the process proceeds to step E6 to play back the broadcast content based on the scene change information. For example, the read buffer 15 receives the read control signal S3 from the control means 30, extracts the MPEG data related to the broadcast content from the hard disk 14, reads it, and reproduces it (mode selection control).

  For example, the separation circuit 17 receives the separation control signal S5 from the control means 30, and operates to separate video compression data and audio compression data from MPEG data based on the separation control signal S5. The video decoder 18 decodes and decodes the compressed video data, and A / D converts the decoded video data. The video decoder 18 outputs the analog video signal Vout after A / D conversion to a monitor or the like via the output terminal 2. The audio decoder 19 decodes and decodes the audio compression data, A / D-converts the decoded audio data, and sends the analog audio signal Aout after A / D conversion via the output terminal 3 in the monitor. Output to the voice input unit. Then, the process proceeds to step E8.

  If the CM playback mode is selected in step E5, the process proceeds to step E7, where commercial content is played based on the scene change information. For example, the read buffer 15 receives the read control signal S3 from the control means 30, extracts the MPEG data related to the commercial content from the hard disk 14, reads it, and reproduces it. Then, the process proceeds to step E8, and the control means 30 determines the end. For example, the control unit 30 detects the power-off information and ends the recording / reproducing process. If the power-off information is not detected, the process returns to step E1 to repeat the above process.

  In this way, based on the scene change information, only the extracted broadcast content block is identified and reproduced, or the extracted commercial content block is deleted and the remaining broadcast content blocks are reproduced. The last detected block period is compared with the currently performed audio mode at the time of transmission, and an operation such as taking an AND logic operation is added to further improve the reproduction accuracy of audio and video information. .

  FIG. 11 is a flowchart showing a file A creation example (subroutine) related to commercial period extraction. In this example, the separation unit 21 of the image processing means 12 shown in FIG. 6 receives MPEG data in the PES format and separates it into a packet element stream (PES) and a macroblock. Thereafter, the PES is output from the separation unit 21 to the GOP analysis unit 22.

  On the premise of this, in step F1 of the flowchart shown in FIG. 11, the GOP analysis unit 22 inputs PES from the separation unit 21 and performs GOP analysis processing. For example, the GOP analysis unit 22 divides the PES into GOP units and detects the presence or absence of the GOP layer in the current frame. Thereafter, in step F2, it is determined whether or not there is a GOP layer in the current frame based on the GOP layer detection result. If there is a GOP layer in the current frame in step F2, the process proceeds to step F3, where the I picture extraction unit 23 extracts the first I picture from the GOP layer immediately after the GOP sequence header of the current frame.

  In step F4, the file A creation unit 24 performs DCT processing on the previously detected I picture of the GOP to obtain a DC value in order to analyze the PTS. The DC value is compared with the previous value, and the PTS or packet number at a point exceeding a certain value is checked. Then, the process proceeds to step F5, where the file A is created by describing the PTS or packet number of the point that has been checked and exceeds a certain value. If the GOP layer does not exist in step F2, the process is terminated without creating the file A. Thereafter, the process returns to step E2 of the flowchart shown in FIG.

  FIG. 12 is a flowchart showing a file B creation example (subroutine) related to commercial period extraction. In this example, when the video display time of the commercial content is an integral multiple of 15 seconds and the point at which the correlation between the MPEG data of the previous frame and the MPEG data of the current frame cannot be obtained is set as the correlation NG point. When extracting the scene change point of the commercial content, the MPEG data of the previous frame and the MPEG data of the current frame are compared to extract the correlation NG point, and the time difference TW between the correlation NG points is calculated. The time difference value between the correlated NG points calculated in step 1 is compared with a value that is an integer multiple of 15 seconds, and a commercial period in which the time difference value and the value that is an integer multiple of 15 seconds coincide is searched.

  In this processing routine, MPEG data is supplied from the separation unit 21 of the image processing means 12 shown in FIG. 6 to the DCT analysis unit 25 in units of macroblocks (MB). The separated macroblock is input to the DCT analysis unit 25, and the video change point detection unit 26 detects the video change point of the video content and finds a scene change point at which the video is switched from the broadcast content to the commercial content. Operate.

  In addition, since the commercial period is a content different from the broadcast content and the commercial content as shown in FIGS. 7 to 9, the video is always divided in GOP units at this division, and the top I in the GOP The DC value of a picture changes greatly because there is no frame correlation before and after the video. The DC component is larger than other pictures.

  Based on this assumption, the file B creation unit 27 detects an I picture in the GOP in step G1 of the flowchart shown in FIG. In step G2, the file B creation unit 27 determines whether or not an I picture existing at the very beginning of the GOP layer is detected. Based on the determination result, DCT processing is performed in step G3 to detect the DC value of the I picture.

  Then, the file B creation unit 27 compares the DC value obtained by DCT processing of the I picture with a preset reference value Vref ′ in step G4, and determines whether or not the DC value of the I picture is greater than or equal to the reference value Vref ′. Is determined. When the DC value of this I picture is less than the reference value Vref ', control is performed so as not to create a file. If the DC value of the I picture is greater than or equal to the reference value Vref ′, the process proceeds to step G6, where the PTS or packet number is checked, and this is described in file B in step G7 as a list of correlation NG (change) points. The

  This correlation NG point is a candidate for a scene change point, and a scene change point is obtained from this correlation NG point. The calculation is performed by DCT processing the I picture in the GOP of the previous frame to obtain a DC value, DCT processing of the I picture in the GOP of the current frame to obtain the DC value, and the difference between the DC values of the two I pictures. ΔDC is obtained, a preset reference value Vref is compared with the calculated DC value difference ΔDC, and based on the comparison result, it is determined whether or not it is a scene change point.

  When the difference ΔDC between the DC value of the first I picture in the GOP of the current frame and the DC value of the I picture in the GOP of the previous frame is equal to or greater than a preset reference value Vref, the file B creation unit 27 Scene change information indicating that the GOP of the frame and the GOP of the previous frame are scene change points that divide the broadcast content and the commercial content is generated. Thereafter, the process returns to step E2 of the flowchart shown in FIG.

  FIG. 13 is a flowchart showing an example of file C creation related to commercial period extraction. In this example, a file that describes a commercial section (period) starting from a correlation NG point that is a candidate for a scene change point, finds a PTS or packet number for specifying the section, and describes this as scene change information Create C.

  Based on this assumption, the file C creation unit 28 reads a list of correlation NG points that are candidates for scene change points from the file B in step H1 of the flowchart shown in FIG. Thereafter, the process proceeds to step H2, and the file C creation unit 28, as shown in FIG. 9, the time differences TW, TW ′, TW ″, etc. between the correlated NG points that are consecutive from the correlated NG point to other correlated NG points. Calculate

  In step H3, the file C creation unit 28 determines whether the time difference TW between correlated NG points is an integral multiple of 15 seconds and 60 seconds or less. In this determination, when the time difference TW between the correlated NG points is not an integral multiple of 15 seconds and not more than 60 seconds, the process proceeds to step H4 to advance one calculation section and return to step H1. When the time difference TW between the correlated NG points is an integral multiple of 15 seconds and 60 seconds or less, the process proceeds to step H5 to check the PTS or packet number related to the calculation start and calculation end.

  Next, the file C creation unit 28 proceeds to step H6 in order to increase the detection accuracy, and determines whether or not a GOP layer exists in the current frame. If there is a GOP layer, the process proceeds to step H7, where the comparison unit 29 compares the list contents of the correlated NG points (file B) that are candidates for the scene change point and the correlated NG points checked in step H5 described above. Compare the check contents. By this comparison process, the detection accuracy of scene change information at the scene change point can be improved.

  Thereafter, in step H8, the commercial period from the correlated NG point that is regarded as a scene change point is extracted, the PTS or packet number related to the calculation start and calculation end is checked, and a file C describing these is created. This file C is scene change information (list) of scene change points related to a commercial content display period. The final file creation unit 210 creates a final file for executing the program playback mode, the CM playback mode, the pre-CM playback mode, or the post-CM playback mode based on the scene change information. These files A to C and the final file are stored in the hard disk 14 together with the MPEG data. If there is no GOP layer in step H6, the process ends. Thereafter, the process returns to step E2 of the flowchart shown in FIG.

  As described above, according to the digital recording / reproducing apparatus 100 as the first embodiment of the present invention, when MPEG data in which commercial content is inserted between broadcast contents is recorded on the hard disk 14, the image processing means 12 is used. Calculates the correlation by comparing the MPEG data of the previous frame and the MPEG data of the current frame, and extracts a scene change point that divides the broadcast content and the commercial content based on the correlation obtained here. The write buffer 13 adds the scene change information indicating the scene change point detected by the image processing means 12 and records the MPEG data on the hard disk 14.

  On the premise of this, at the time of data reproduction, the operation means 40 is operated so as to select either the broadcast program reproduction mode or the CM reproduction mode. The control unit 30 controls the read buffer 15 so as to reproduce broadcast content or commercial content based on the broadcast program playback mode or CM playback mode selected by the operation unit 40 and scene change information read from the hard disk 14. . The read buffer 15 receives the mode selection control by the control means 30 and reads the MPEG data from the hard disk 14 and reproduces it.

  Therefore, after the broadcast program is recorded on the hard disk 14, it is possible to cut the commercial content and reproduce only the broadcast program at the time of data reproduction. Conversely, only commercial content can be extracted and played back. Thereby, the viewer can enjoy the broadcast program at once without a commercial.

  FIG. 14 is a flowchart showing an example of reproduction processing in the digital recording / reproducing apparatus 100 as the second embodiment according to the present invention. In this embodiment, at the time of MPEG data playback, a mode in which commercial content is collectively played back before playback of broadcast content (hereinafter referred to as pre-CM playback mode), or a mode in which commercial content is played back collectively after playback of broadcast content ( Any one of the “post-CM playback mode” is accepted. Therefore, the operation means 40 shown in FIG. 1 has a function of selecting either the pre-CM playback mode or the post-CM playback mode.

  Using this as a processing condition, first, MPEG data is input to the digital interface circuit 11 in step L1 of the flowchart shown in FIG. In step L2, the image processing means 12 detects a scene change point that divides broadcast content and commercial content based on the image processing control signal S1, and extracts a commercial period. At this time, the image processing means 12 compares the MPEG data of the previous frame and the MPEG data of the current frame to obtain a correlation, and a scene change that separates the broadcast content and the commercial content based on the obtained correlation. Extract points.

  Specifically, the flowchart (subroutine) shown in FIG. 11 is called to create file A, the flowchart (subroutine) shown in FIG. 12 is called to create file B, and the flowchart shown in FIG. (Subroutine) is called to create file C. Then, scene change information indicating scene change points based on the previously created files A to C is added to the MPEG data in step L3 and recorded on the hard disk 14. At this time, the control means 30 outputs a write control signal S2 to the write buffer 13. The write buffer 13 receives the write control signal S2 from the control means 30, and adds scene change information to the MPEG data based on the write control signal S2 and records it on the hard disk 14.

  Thereafter, in step L4, the control means 30 inputs a normal operation / recording reproduction instruction, a reproduction instruction, and the like through the operation means 40. The normal operation / recording / playback instruction or the like is input by the user operating the operation means 40. Here, when the recording / playback mode is selected, the contact c is connected to the b side, and MPEG data relating to broadcast content, commercial content, and the like is read from the hard disk 14.

  If the recording / playback mode is selected, the process proceeds to step L5, and the operation means 40 is operated to select either the pre-CM playback mode or the post-CM playback mode. When the pre-CM playback mode is selected, the process proceeds to step L6, and the control unit 30 controls the read buffer so that the commercial content is collectively played back before the broadcast content is played based on the scene change information. At this time, the read buffer 15 receives the read control signal S3 from the control means 30, extracts only the MPEG data related to the commercial content, extracts it from the hard disk 14, reads it, and reproduces it. Thereafter, the process proceeds to step L7, and the control means 30 controls the read buffer so that broadcast contents are collectively reproduced based on the scene change information.

  If the post-CM playback mode is selected in step L5, the control unit 30 proceeds to step L8, and the control means 30 reads the broadcast content so that the broadcast contents are collectively played back before the commercial playback based on the scene change information. To control. At this time, the read buffer 15 receives the read control signal S3 from the control means 30, extracts only the MPEG data related to the broadcast content, extracts it from the hard disk 14, reads it, and reproduces it.

  Thereafter, the process proceeds to step L9, and the control means 30 controls the read buffer so that the commercial contents are collectively reproduced based on the scene change information. When there is no reproduction instruction, the process proceeds to step L10, and the control means 30 determines the end. For example, the control unit 30 detects the power-off information and ends the recording / reproducing process. If the power-off information is not detected, the process returns to step L1 to repeat the above process.

  As described above, the digital recording / reproducing apparatus 100 according to the second embodiment accepts selection of either the pre-CM reproduction mode or the post-CM reproduction mode when reproducing MPEG data.

  Therefore, like a screening mode in a movie theater, commercial contents can be reproduced together before reproduction of broadcast contents, or commercial contents can be reproduced together after reproduction of broadcast contents.

  The present invention is extremely suitable when applied to a hard disk built-in digital video recorder for recording terrestrial digital broadcast programs, communication satellite digital broadcast programs, and the like.

It is a conceptual diagram which shows the structural example of the digital recording / reproducing apparatus 100 as each Example which concerns on this invention. (A)-(C) are the conceptual diagrams which show the example of compression of the image data by MPEG-2 format. It is a block diagram which shows the structural example of the information coding apparatus 200 with a motion compensation function. (A) And (B) is a block diagram which shows the example of a frame rearrangement of MPEG data. (A)-(F) are the conceptual diagrams which show the hierarchical structure example of MPEG data. 3 is a block diagram illustrating an example of an internal configuration of an image processing unit 12. FIG. (A) And (B) is a time chart which shows the example of extraction of the scene change point in the image processing means 12. FIG. It is a time chart which shows the example of extraction of the commercial period in the image processing means 12. (A)-(C) are time charts showing commercial content insertion sections and search examples thereof. It is a flowchart which shows the example of recording / reproducing in the digital recording / reproducing apparatus 100 which concerns on a 1st Example. It is a flowchart which shows the file A preparation example (subroutine) which concerns on commercial period extraction. It is a flowchart which shows the file B preparation example (subroutine) which concerns on commercial period extraction. It is a flowchart which shows the example of file C preparation which concerns on commercial period extraction. It is a flowchart which shows the example of a reproduction | regeneration processing in the digital recording / reproducing apparatus 100 as 2nd Example based on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 12 ... Image processing means, 13 ... Write buffer (recording means), 14 ... Hard disk (recording medium), 15 ... Read buffer (reproducing means), 30 ... Control means, 40 ... .Operating means, 100 ... digital recording / reproducing apparatus

Claims (17)

An apparatus for recording digital data in which commercial content is inserted between broadcast contents on a recording medium and reproducing the digital data from the recording medium,
Image processing means for detecting a video switching part that separates the broadcast content and the commercial content by inputting the digital data;
Recording means for adding the identification information indicating the video switching portion detected by the image processing means to the digital data and recording it on the recording medium;
Reproduction means for reading out and reproducing digital data from the recording medium recorded by the recording means;
Operation means operated to select either the mode for reproducing the broadcast content or the mode for reproducing the commercial content;
Digital recording / reproduction comprising: control means for controlling the reproduction means to reproduce broadcast content or commercial content based on the mode selected by the operation means and identification information read from the recording medium apparatus. The digital data is a frame screen structure and image data in a data stream format,
The image processing means includes
Comparing the image data of the previous frame with the image data of the current frame to obtain a correlation,
2. The digital recording / reproducing apparatus according to claim 1, wherein a video switching part that separates the broadcast content and the commercial content is extracted based on the obtained correlation. The image processing means includes
The data stream is divided into video group units composed of I, B, and P pictures,
Detecting the leading I picture in the video group of the current frame from the image data divided in units of the video group;
A DC value that is a luminance component is obtained by performing discrete cosine transform processing on the detected I picture of the video group,
Comparing the DC value of the leading I picture of the video group of the current frame with a preset reference value;
3. The digital recording / reproducing apparatus according to claim 2, wherein it is determined whether or not the video switching portion is based on a comparison result. The image processing means includes
When the DC value of the first I picture in the video group of the current frame is equal to or greater than a preset reference value, the video group of the current frame and the video group of the previous frame separate broadcast content and commercial content. 4. The digital recording / reproducing apparatus according to claim 3, wherein identification information indicating that it is a video switching portion is generated. The image processing means includes
The data stream is divided into video group units composed of I, B, and P pictures,
Detecting the leading I picture in the video group of the current frame from the image data divided in units of the video group;
The detected I picture of the video group of the current frame is subjected to discrete cosine transform processing to obtain a DC value that is a luminance component, and the I picture in the video group of the previous frame is subjected to DCT processing to obtain a DC value that is a luminance component. ,
A difference between a DC value of a leading I picture in the video group of the current frame and a DC value of a leading I picture in the video group of the previous frame;
Compare the preset reference value with the calculated difference between the DC values,
3. The digital recording / reproducing apparatus according to claim 2, wherein it is determined whether or not the video switching portion is based on a comparison result. The image processing means includes
When the difference between the DC value of the first I picture in the video group of the current frame and the DC value of the I picture in the video group of the previous frame is greater than or equal to a preset reference value, the video group of the current frame 6. The digital recording / reproducing apparatus according to claim 5, wherein identification information indicating that the video group of the previous frame and the video group of the previous frame is a video switching part that divides broadcast content and commercial content is generated. When the video display time of the commercial content is an integer multiple of a unit second value,
When the point at which the correlation between the image data of the previous frame and the image data of the current frame cannot be obtained is a poor correlation point,
The image processing means includes
When extracting the video switching part of the commercial content,
Comparing the image data of the previous frame and the image data of the current frame to extract the poor correlation point, and calculate a time difference between the poor correlation points,
Compare the calculated time difference value between the poor correlation points and the integral multiple of the unit second value,
The digital recording / reproducing apparatus according to claim 1, wherein a commercial period in which the time difference value and a value that is an integral multiple of the unit second value match is searched. The operation means includes
2. The function of selecting one of a mode in which the commercial content is collectively played back before the broadcast content is played back, or a mode in which the commercial content is played back together after the broadcast content is played back. The digital recording / reproducing apparatus of description. A method of recording digital data in which commercial content is inserted between broadcast contents on a recording medium and reproducing the digital data from the recording medium,
Detecting a video switching part that separates the broadcast content and commercial content by inputting the digital data,
Adding identification information indicating the detected video switching part and recording the digital data on a recording medium;
A digital recording / reproducing method, wherein broadcast content or commercial content is reproduced based on the identification information from digital data recorded on the recording medium. When playing back the digital data,
10. The digital recording / reproducing method according to claim 9, wherein selection of reproduction of the broadcast content or commercial content is accepted. When playing back the digital data,
The selection of either a mode in which the commercial content is collectively reproduced before reproduction of the broadcast content or a mode in which commercial content is collectively reproduced after reproduction of the broadcast content is accepted. Digital recording and playback method. The digital data is a frame screen structure and image data in a data stream format,
Comparing the image data of the previous frame with the image data of the current frame to obtain a correlation,
The digital recording / reproducing method according to claim 9, further comprising: extracting a video switching portion that divides the broadcast content and the commercial content based on the obtained correlation. The data stream is divided into video group units composed of I, B, and P pictures,
Detecting the leading I picture in the video group of the current frame from the image data divided in units of the video group;
A DC value that is a luminance component is calculated by performing discrete cosine processing on the detected I picture of the video group,
Comparing the calculated DC value of the first I picture with a preset reference value;
13. The digital recording / reproducing method according to claim 12, wherein it is determined whether or not the video switching portion is based on a comparison result.   When the DC value of the first I picture in the video group of the current frame is equal to or greater than a preset reference value, the video group of the current frame and the video group of the previous frame are combined with the broadcast content and commercial content. 14. The digital recording / reproducing method according to claim 13, wherein identification information indicating that the video switching part is divided is recorded. The data stream is divided into video group units composed of I, B, and P pictures,
Detecting the leading I picture in the video group of the current frame from the image data divided in units of the video group;
A DC value as a luminance component is calculated by performing discrete cosine processing on the detected I picture of the video group of the current frame, and a DC value as a luminance component is calculated by performing DCT processing on the I picture in the video group of the previous frame. And
Calculating the difference between the DC value of the first I picture in the video group of the current frame and the DC value of the first I picture in the video group of the previous frame;
Compare the preset reference value with the calculated difference between the DC values,
13. The digital recording / reproducing method according to claim 12, wherein it is determined whether or not the video switching portion is based on a comparison result.   When the difference between the DC value of the first I picture in the video group of the current frame and the DC value of the first I picture in the video group of the previous frame is equal to or greater than a preset reference value, 16. The digital recording / reproducing method according to claim 15, wherein identification information indicating that the video group and the video group of the previous frame are video switching portions that divide the broadcast content and commercial content is recorded. When the video display time of the commercial content is an integer multiple of a unit second value,
When the point at which the correlation between the image data of the previous frame and the image data of the current frame cannot be obtained is a poor correlation point,
When extracting the video switching part of the commercial content,
Comparing the image data of the previous frame and the image data of the current frame to extract the poor correlation point, and calculate a time difference between the poor correlation points,
Compare the calculated time difference value between the poor correlation points and the integral multiple of the unit second value,
10. The digital recording / reproducing method according to claim 9, wherein a commercial period in which the time difference value and a value that is an integral multiple of the unit second value are matched is searched.

Images