JP2013030259A - Video processor and video processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a video processor and video processing method capable of separating video of a broadcast program at a suitable position.SOLUTION: The video processor of this embodiment includes reception means for receiving video data of a broadcast program, CM detection means for detecting a CM section from the received video data, and storage means for associating separation information with a first position in the vicinity of the head of a first section of the video data and with a second position in the vicinity of the rear end of the CM section to store the separation information together with the video data in a storage medium in the case that video is different between a first section before the detected CM section and a second section after the CM section, and that a portion of characters of a subtitle corresponding to the first section coincides with a portion of characters of a subtitle corresponding to the second section.

Description

  Embodiments described herein relate generally to a video processing apparatus and a video processing method.

  When recording a broadcast program video, there are a technique for detecting a CM section and adding delimiter information such as chapters at the boundary between the main part of the program and the CM section, and a technique for deleting the CM section by dividing it.

JP 2007-272155 A JP 2007-82091 A

  Depending on the broadcast program, in the main part before and after the CM, the video corresponding to the content may be broadcast before and after the CM. In such a case, the video of the broadcast program may be divided at a suitable position. preferable.

  Therefore, an object of the embodiment of the present invention is to provide a video processing apparatus and a video processing method capable of dividing a video of a broadcast program at a suitable position.

  In order to solve the above problems, the video processing apparatus according to the present embodiment includes a receiving unit that receives video data of a broadcast program, a CM detecting unit that detects a CM section from the received video data, and a detected CM. The video of the first section before the section and the second section after the section are different, and a part of the subtitle characters corresponding to the first section and a part of the subtitle characters corresponding to the second section are In the case of coincidence, there is provided storage means for associating the delimiter information with the first position in the vicinity of the head of the first section of the video data and the second position in the vicinity of the rear end of the CM section and storing them in the storage medium together with the video data.

The figure which shows the utilization example of the video processing apparatus of embodiment. The figure which shows the system configuration example of the video processing apparatus of embodiment. FIG. 3 is a diagram illustrating an example of video segmentation processing by the video processing apparatus according to the embodiment. The figure which shows the example of subtitle data which the video processing apparatus of embodiment uses in a division | segmentation process. The figure which shows the example of a processing flow which concerns on the division | segmentation process of the image | video by the video processing apparatus of embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
FIG. 1 is a diagram showing an example of how the video processing apparatus according to this embodiment is used. The video processing apparatus according to the present embodiment is realized as the recording apparatus 100, for example. The recording device 100 receives a broadcast wave received by a television broadcast receiving antenna and records (records) video data of a broadcast program included in the broadcast wave. The recording device 100 is connected to the display device 200 by wireless / wired via a digital interface, and outputs the recorded video data to the display device 200 to display the video of the video data on the display device 200. The recording apparatus 100 according to the present embodiment can divide sections before and after the CM section of the broadcast program at suitable positions when recording video data of the broadcast program. More specifically, in a case where the content broadcast before the CM section is repeatedly broadcast after the CM section, the recording apparatus 100 preferably determines the contents of the section before and after the CM, When a video is recorded and played back, it can be divided at a suitable position so that a portion broadcast repeatedly can be skipped.

FIG. 2 is a diagram illustrating a system configuration example of the recording device 100 and the display device 200.
The recording apparatus 100 includes a reception unit 110, a separation unit 120, a section detection unit 130, a caption similarity detection unit 140, a separation processing unit 150, a storage unit 160, an operation reception unit 170, a recording / playback control unit 180, a video output unit 190, and the like. Is provided.

  The receiving unit 110 is a tuner module, for example, and receives a broadcast wave of digital television broadcasting. Note that, for example, video data (main video), audio data, and subtitle data (sub-video) streams are superimposed on the broadcast wave, for example, in a TS (Transport Stream) format. The separation unit 120 separates the video data stream and the subtitle video stream superimposed on the broadcast wave, and outputs the main video to the section detection unit 130 and the subtitle video to the subtitle similarity detection unit 140.

  The section detection unit 130 includes a CM detection unit 131, a scene detection unit 132, and an image similarity detection unit 133. The CM detection unit 131 detects an arbitrary section (CM section) included in the input video data. Here, the CM detection unit 131 detects the CM section based on, for example, a silent section existing between the CM section and the main part. That is, since the CM section has a time length that is approximately an integral multiple of 15 seconds, the CM detection unit 131 determines whether the time between the detected silence section and the silence section is an approximate integer of 15 seconds. A CM section is detected.

  The scene detection unit 132 detects a time position (scene change position) at which the scene changes in the main section of the video data of the broadcast program, and detects a main scene between the scene change position in the main section before and after the CM section and the CM section. To do. Here, the detection of the scene change position is performed using a known technique described in Patent Document 2, for example. The scene detection unit 132 does not necessarily need to detect the scene change position for all the main sections, and may detect the scene change position for a predetermined time length at least before and after the CM section.

  The image similarity detection unit 133 determines the similarity between the video in the pre-CM scene and the video in the post-CM scene. Here, the image similarity detection unit 133 divides both the video of the scene before CM and the video of the scene after CM into a plurality of divided sections, and assigns an identifier to each divided section. And the image similarity detection part 133 discriminate | determines the similarity of the image | video of the division area before CM and the division area after CM for every area. Then, when all of the video in the segment before the CM and the image in the segment after the CM match (when the similarity is equal to or greater than a predetermined threshold), the image similarity detection unit 133 causes the segment processor 150 to The scene change position and the end position of the CM section are notified, and an instruction is given to divide the video at these positions.

  On the other hand, the image similarity detection unit 133 detects the similarity of subtitles in the non-matching divided section when a part of the video in the divided section before the CM and a part of the video in the divided section after the CM match. To the subtitle similarity detection unit 140. Note that the image similarity detection unit 133 separates the CM start position (start time) and the CM end position (end time) when the section in which the video in the scene differs before and after the CM is a predetermined time length or more. The processing unit 150 may be notified and instructed to divide at these positions. Detailed processing of the section detection unit 130 will be described later with reference to FIG.

  A caption data stream is input to the caption similarity detection unit 140. Then, the caption similarity detection unit 140 detects the similarity of the caption data displayed in the section received the instruction from the image similarity detection unit 133. That is, the caption data in the divided section before the CM section specified by the image similarity detection unit 133 is compared with the caption data in the divided section after the CM section specified by the image similarity detection unit 133, and these caption data are compared. Whether or not are similar to each other is determined. Then, the subtitle similarity detection unit 140 instructs the delimiter processing unit 150 to delimit the scene change position before the CM section and the end position of the CM section when these subtitles are similar. On the other hand, when the subtitles are not similar, the subtitle similarity detection unit 140 instructs the separation processing unit 150 to divide the video at the CM start position and the CM end position.

  Note that caption data does not necessarily have to be input as a stream to the caption similarity detection unit 140, and video data in which captions are embedded may be input. In this case, the subtitle similarity detection unit 140 detects the subtitle characters in the video and determines the similarity between the subtitle characters detected before and after the CM.

  The delimiter processing unit 150 adds delimiter information to positions of the video data stream and the subtitle data stream according to instructions from the section detection unit 130 and the subtitle similarity detection unit 140, and generates a playlist. Then, the separation processing unit 150 outputs the created playlist to the storage unit 160. The delimiter processing unit 150 may delete a part of the stream at a position corresponding to the input instruction and combine the parts that are not deleted. The storage unit 160 is, for example, a storage medium such as an HDD or a removable medium, and the video data, audio data, and subtitle data streams received by the receiving unit 110 are input in, for example, the TS format (the stream path is not shown). . The storage unit 160 stores the input playlist and the input data in the TS format in association with each other. That is, the storage unit 160 stores the video data of the broadcast program and the delimiter information of the video data in association with each other.

  The operation reception unit 170 receives an operation input from the user. Here, the operation accepting unit 170 is, for example, an operation input for instructing recording of any channel of television broadcasting, an operation input for outputting an image of any channel received by the receiving unit 110 to an external device, and the storage unit 160 An operation input for instructing reproduction of recorded video data is received. Then, the operation accepting unit 170 notifies the recording / playback control unit 180 according to the accepted operation input.

  The recording / playback control unit 180 executes a recording process and a playback process in response to the operation input received by the operation receiving unit 170. When receiving an operation input instructing recording of any channel, the recording / playback control unit 180 causes the receiving unit 110 to receive a channel corresponding to the operation input, and stores the TS data of the received channel in the storage unit 160. Remember. Further, when there is an operation for outputting video data of a program being broadcast to an external device, the recording / playback control unit 180 causes the receiving unit 110 to receive a channel corresponding to the operation, and the received video data is output to the video output unit. Output from 190 to an external device. The recording / playback control unit 180 causes the video output unit 190 to output the video stored in the storage unit 160 to an external device when a recorded video playback operation is performed. Note that when video is output from the video output unit 190 to an external device, video and audio may be decoded before output.

  Next, the display device 200 will be described. The display device 200 includes a video input unit 210, a video processing unit 220, and the like. The video input unit 210 receives video data transmitted from the recording device 100. The video processing unit 220 converts the video data received by the video input unit 210 into a video signal in a format that can be displayed by the display unit 230, and outputs the video signal to the display unit 230. If the video data received by the video input unit 210 has not been decoded, the video processing unit 220 may decode the video data and convert it into a video signal. The display unit 230 displays an image using the input image signal.

FIG. 3 is a diagram illustrating an example of video segmentation processing by the recording apparatus 100.
First, the CM detection unit 131 detects a CM section from the video data, and detects boundaries P1 and P2 between the CM section and the main part as shown in FIG. Next, the scene detection unit 132 detects the scene change position P3 from a section within a predetermined section length from the boundary P1 in the section before the CM section. Similarly, the scene detection unit 132 detects the scene change position P4 from a section within a predetermined section length from the boundary P2 among sections after the CM section.

  Then, the image similarity detection unit 133 divides the scene n2 immediately before the CM section and the scene n3 immediately after the CM section into a plurality of shots (divided sections), and the divided sections at the same position in the scene n2 and the scene n3, respectively. The similarity of is detected. That is, the similarities between the shots m1 and m4, the shots m2 and m5, and the shots m3 and m6 are detected. Here, as shown in FIG. 3C, when the shot m1 and the shot m4, the shot m2 and the shot m5 are similar, and the shot m3 and the shot m6 are dissimilar, the recording apparatus 100 is shown in FIG. Separation positions are selected according to the similarity of subtitles to be described later.

  That is, even if the video of shot m3 and the video of shot m6 are dissimilar, if the subtitle of shot m3 and the subtitle of shot m6 are similar, recording apparatus 100 makes scene n2 and scene n3 similar. It is determined that it is a section, and the position P9 and the position P10 are set as separation positions. Then, the delimiter processing unit 150 assigns delimiter information such as chapters to the positions P9 and P10, and stores them in the storage unit 160 in association with the video data. The chapter is delimiter information for jumping the video playback position from the chapter position to the next chapter position during video playback.

  Further, when the delimiter processing unit 150 assigns the delimiter information to the video data and associates them with each other, the delimiter 150 does not necessarily have to strictly apply the positions P9 and P10. For example, the delimiter 150 may be shifted by several hundred msec. However, by matching the position of the delimiter information with the position P9 of the scene change point, it is possible to suppress unnatural interruption of the video when jumping using chapters during video playback.

  In FIG. 3B, the rear end of the scene n2 is cut by CM, but since the scene n4 is connected to the rear end of the scene n3, the scene change point may be originally set between the scene n3 and the scene n4. In some cases, P4 is not detected. In this case, the scene detection unit 132 first detects the scene change point P3 from the section before the CM, detects the section length (time length) of the scene n2, and sets the position after the section length from the position P2 as a temporary scene. It may be detected as a change point. The subsequent processing in this case is performed in the same manner as in FIGS.

  As a result, in variety programs and quiz programs, before and after the CM section, even if the video is slightly different, the content is similar. For example, before the CM section, a mosaic or a strip-shaped image is displayed on a part of the video Even in the case where these are not present after the CM section, the video can be divided at a suitable position based on the similarity of the subtitles.

  If the shot m3 video and the shot m6 video, and the subtitle of the shot m3 and the subtitle of the shot m6 are not similar, the delimiter 150 determines that the scene n2 and the scene n3 are dissimilar sections and sets the delimiter information as CM You may give to a start position and an end position.

FIG. 4 is a diagram showing an example of subtitles displayed in each shot section shown in FIG.
In the example shown in FIG. 4A, the subtitles of shot m1 and shot m4, shot m2 and shot m5, and shot m3 and shot m6 match. Therefore, the caption similarity detection unit 140 determines that the shot m1 and the shot m4, the shot m2 and the shot m5, and the shot m3 and the shot m6 are similar sections, and the scenes n2 and n3 that include these shots. Are determined to be similar sections.

  In the example shown in FIG. 4B, “Goal! Second time ...” is displayed in the subtitle of the shot m3 before the CM section, and “Goal! 77 points in the second time of the shot m6 after the CM section. ! "Is displayed. Here, in the subtitles before CM, in order to prevent the contents of important scenes after CM from being known to the user in advance, for example, quiz answers and game scores are indicated with “...” and the like in the CM In some cases, a method of entering is used. Therefore, when the subtitle similarity detection unit 140 has a notation such as “...” In the subtitle before the CM, the subtitle similarity detection unit 140 applies the post-CM post-CM to the characters before the “. Compare with subtitle characters.

  Then, the subtitle similarity detection unit 140 matches the subtitles of the shot m2 before “...” And the portions of the shot m6 before “77 points!”. Determine. In other words, the subtitle similarity detection unit 140 determines that the subtitles are similar if the characters for a predetermined number of characters from the head of the subtitles before the CM match a part of the subtitle characters after the CM. .

  In the example shown in FIG. 4C, “That's why, start first” is displayed on the caption of the shot m3 before the CM section, and “First time, OK” is displayed on the caption of the shot m6 after the CM section. “Start” is displayed. Here, when the content of the subtitle before the CM is long, the subtitle after the CM may be displayed by deleting the head portion of the character string of the subtitle before the CM. For this reason, the subtitle similarity detection unit 140 determines that these subtitles are similar when a part of the character behind the subtitle before the CM matches the subtitle character after the CM.

  As shown in FIGS. 4C and 4D, when a part of subtitles before CM and a part of subtitles after CM match, if it is determined that these subtitles are similar, If the character string length of the subtitle is remarkably short or the character string is not meaningful, it may be determined to be similar. In such a case, it may not be determined to be similar.

  In the example shown in FIG. 4D, the subtitles of the shot m3 before the CM section are displayed with “<Surely Saito ...>”, “Please do!” And “(Staff) Ikimasu”. Subtitles of shot m6 after the section are displayed as “<Saily Saito ...>”, “(Staff) Ikimasu”, and “(Saito) Yoshi!”. As described above, some shots may include a plurality of subtitles, and the contents may be slightly different before and after the CM.

  In the example of FIG. 4D, the first, fourth, third, and fifth character strings match, but the second subtitle before the CM does not exist after the CM. In such a case, a match is determined for each of a plurality of subtitles included in the shot before the CM and a plurality of subtitles included in the shot after the CM, and the shots before and after the CM are similar according to the ratio of the number of matches. It may be determined whether or not. That is, in the example of FIG. 4D, there are three subtitles in each of the shots m3 and m6, and the two match each other, so the number of matches is 4, the number of mismatches is 2, and the number of subtitle matches / total As the number of subtitles = 4/6 = 0.67, it may be determined that the shots are similar if this numerical value is equal to or greater than a predetermined threshold. Also in the example of FIG. 4D, as shown in FIGS. 4C and 4D, it is determined that the subtitles are similar by matching a part of the subtitles. Also good.

FIG. 5 is an example of a processing flow related to the separation processing by the recording apparatus 100.
First, when the receiving unit 110 receives video data of a television broadcast program (S501), the CM detecting unit 131 detects a CM section in the video data (S502). Next, the scene detection unit 132 detects a scene change point from a certain length section before and after the CM section (S503). Then, the image similarity detection unit 133 divides the previous section and the subsequent section of the CM section into a plurality of divided sections, and whether or not the video matches before and after the CM section for each divided section. It is determined (whether or not it is equal to or higher than a predetermined similarity) (S504).

  When the videos in all the divided sections are the same before and after the CM section (No in S505), the separation processing unit 150 determines the positions of the positions P9 and P10 in FIG. 3D of the video data. The delimiter information is associated with and stored in the storage unit 160 together with the video data (S510). On the other hand, when there is a divided section in which the video does not match before and after the CM section as in shot m3 and shot m6 in FIG. 3C (Yes in S504), the caption similarity detection unit 140 The similarity of the caption displayed at the time corresponding to the divided section where the videos do not match is detected (S506).

  Here, when the subtitle similarity detection unit 140 matches part of the subtitle characters in the segment before the CM and part of the subtitle characters in the segment after the CM (Yes in S507), the scene before and after the CM Are similar to each other (S509). Note that, when the subtitle similarity detection unit 140 matches all the subtitle characters in the shot m3 and all the subtitle characters in the shot m6 as shown in FIG. ), When a part of the front / back of the subtitles matches, a predetermined number or more of the subtitles corresponding to the shot before the CM are included in the subtitles as shown in FIG. 4D. If all of the characters match the characters included in the caption of the shot after the CM section, the process proceeds to the Yes flow in S507.

  The delimiter processing unit 150 adds delimiter information (such as chapters) to the scene change point before the CM and the end position of the CM section, and stores them in the storage unit 160 together with the video data (S510). On the other hand, in S507, when the subtitles in the division section before CM and the subtitles in the division section after CM do not match (No in S507), the delimiter processing unit 150 displays the CM section as in positions P1 and P2 in FIG. Separation information (such as chapters) is assigned to the vicinity of the start position and the vicinity of the end position (S508).

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

  DESCRIPTION OF SYMBOLS 100 ... Recording apparatus 110 ... Receiving part 120 ... Separation part 130 ... Section detection part 131 ... CM detection part 132 ... Scene detection part 133 ... Image similarity detection part 140 ... Subtitle similarity detection part 150 ... Delimiter processing unit, 160 ... Storage unit, 170 ... Operation receiving unit, 180 ... Recording / playback control unit, 190 ... Video output unit, 200 ... Display device, 210 ... Video input unit, 220 ... Video processing unit, 230 ... Display unit

Claims (8)

Receiving means for receiving video data of a broadcast program;
CM detecting means for detecting a CM section from received video data;
The detected video of the first section before the CM section and the second section after the detected CM section are different, and a part of the subtitle character corresponding to the first section and the subtitle corresponding to the second section When a part of the character matches, the delimiter information is stored in association with the video data in association with the first position near the beginning of the first section of the video data and the second position near the rear end of the CM section. A video processing apparatus comprising storage means for storing in a medium.   The video processing apparatus according to claim 1, wherein the delimiter information includes setting information for causing a jump from the first position to the second position when the video data is reproduced. A scene detecting means for detecting a first scene change point in a section before the detected CM section;
The video processing apparatus according to claim 1, wherein the first section is a section between the first scene change point and the CM section.   The first section according to claim 3, wherein the first section is a section that does not include an image that matches a section after the CM section among sections between the first scene change point and the CM section. Video processing device.   The storage means associates the delimiter information with the video data when all of the subtitle characters corresponding to the first section and all of the subtitle characters corresponding to the second section match. The video processing apparatus according to claim 1, wherein the video processing apparatus is stored in a storage medium.   In the case where a plurality of subtitles correspond to each of the first section and the second section, the storage means includes subtitles in a predetermined number or more of subtitles corresponding to the first section. 2. The video processing device according to claim 1, wherein when a part of characters included in the text matches a character included in the subtitles of the second section, the video data is stored in the storage medium in association with the video data. .   In the case where a plurality of subtitles correspond to each of the first section and the second section, the storage means includes subtitles in a predetermined number or more of subtitles corresponding to the first section. The video processing apparatus according to claim 1, wherein when all of the characters included in the text match the characters included in the subtitles of the second section, the video data is stored in the storage medium in association with the delimiter information. A video processing method in an apparatus for processing video,
Receiving broadcast program video data;
Detecting a CM section from received video data;
The detected video of the first section before the CM section and the second section after the detected CM section are different, and a part of the subtitle character corresponding to the first section and the subtitle corresponding to the second section When a part of the character matches, the delimiter information is stored in association with the video data in association with the first position near the beginning of the first section of the video data and the second position near the rear end of the CM section. A video processing method comprising: storing the information in a medium.

Images