JP2003046896A - Image processing unit, method for processing image and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow an image scene to be digested while a processing load is being reduced. SOLUTION: An image processing unit 100 for processing an image signal of a dynamic image comprises a subtitle judging unit 102 for judging whether a subtitle of a specific shape is included at a specific position of each image frame for constituting the dynamic image or not, an information extracting unit 103 for extracting predetermined information from the frame including the subtitle if it is judged that the subtitle is included by the unit 102, a storage unit for storing predetermined information extracted by the unit 103, and a subtitle relating information processing output unit 105 for outputting predetermined information stored in the storage unit in response to a user's request.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video processing device,
More specifically, the present invention relates to an image processing method and an image processing program, and more specifically, an image processing device that uses subtitles included in an image,
The present invention relates to a video processing method and a video processing program.

[0002]

2. Description of the Related Art In recent years, CS broadcasting and DVD (Digital)
Due to the widespread use of the Vertesile Dsik, etc., many video media have been flooded.
Therefore, in order to allow the viewer to effectively spend a limited viewing time, a search device having a function of efficiently searching for a program having a desired content from many videos is desired.

Conventionally, additional information (index information) about a video is added to a digital broadcast or a recorded video tape, and a user can search based on the index information to roughly grasp the content of the video. The technology to do is known. On the other hand, a technique of analyzing video and extracting index information is also known. Attempts have also been made to structure video by extracting subtitles from video and performing character recognition.

[0004]

However, in the above-mentioned prior art, since the index information depends on the taste of the person who added it, a great deal of effort is required to meet the interests of all viewers. Further, with respect to a technique of analyzing video and extracting index information, a great deal of information processing capability is conventionally required. Furthermore, regarding the technology of extracting subtitles and performing character recognition, the technology must be specialized for news scenes, and a large knowledge base is required to structure the extracted character information. It still requires a great deal of information processing capability as described above.

The present invention has been made in view of the above, and it is an object of the present invention to enable a digest of a video scene while reducing the processing load.

[0006]

In order to achieve the above object, a video processing device according to the invention of claim 1 is a video processing device for processing a video signal of a moving picture, and each video frame constituting the moving picture. In the case where it is determined that the subtitle is included by the subtitle determining unit that determines whether or not the specific position includes the subtitle of the specific shape,
Information extracting means for extracting the predetermined information from the video frame including the subtitles, storage means for storing the predetermined information extracted by the information extracting means, and the predetermined information stored by the storage means in response to a user request And a processing unit for outputting.

According to the present invention, information relating to the highlight scene of a moving image such as sports can be easily obtained.

According to a second aspect of the present invention, there is provided the video processing device according to the first aspect, wherein the information extracting means extracts the information displayed near the subtitle. And

According to the present invention, the information related to the highlight scene of the moving image can be acquired more easily.

The video processing device according to the invention of claim 3 is the video processing device according to claim 2, wherein the information displayed in the vicinity of the subtitle is score information of the sports video. To do.

According to the present invention, it becomes possible to easily construct a highlight scene having a change in score.

The video processing apparatus according to the invention of claim 4 is the video processing apparatus according to claim 1, 2 or 3, wherein when the predetermined information extracted by the information extracting means changes, the change is made. It is characterized by comprising a video frame extraction means for extracting video frames before and after.

According to the present invention, the highlight portion of the moving image can be easily extracted based on the information extracted by the information extracting unit.

A video processing method according to the invention of claim 5 is a video processing method for processing a video signal of a moving image,
A subtitle determining step of determining whether or not a subtitle having a specific shape is included in a specific position of each video frame forming the moving image,
An information extracting step of extracting predetermined information from a video frame including the subtitle when the subtitle determining step determines that the subtitle is included, and a storage storing the predetermined information extracted in the information extracting step. It is characterized by including a step and a processing step of outputting the predetermined information stored in the storage step in response to a user's request.

According to the present invention, it is possible to easily acquire information relating to the highlight scene of a moving image such as sports.

The video processing method according to the invention of claim 6 is the video processing method according to claim 5, wherein in the information extracting step, the information displayed in the vicinity of the caption is extracted. And

According to the present invention, the information related to the highlight scene of the moving image can be acquired more easily.

Further, in the video processing method according to the invention of claim 7, in the video processing method according to claim 6, the information displayed in the vicinity of the caption is score information of sports video. Characterize.

According to the present invention, it becomes possible to easily construct a highlight scene having a change in score.

The video processing method according to the invention of claim 8 is the video processing method according to claim 5, 6 or 7, wherein when the predetermined information extracted in the information extraction step changes, the change It is characterized by including a video frame extraction step of extracting video frames before and after.

According to the present invention, the highlight portion of the moving image can be easily extracted based on the information extracted by the information extracting unit.

A video processing program according to a ninth aspect of the present invention causes a computer to execute the video processing method according to any one of the fifth to eighth aspects.

According to the present invention, the same operational effects as those of the method according to any one of claims 5 to 8 can be obtained via a computer.

[0024]

BEST MODE FOR CARRYING OUT THE INVENTION Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that although a case where a television program is used as an input video is described here, the description here is merely an example, and the present invention is not limited to the following.

FIG. 1 is a block diagram of a video processing apparatus according to the present invention. The video processing device 100 includes a video input unit 101 for inputting a video signal of a moving image, a caption discriminating unit 102 for discriminating whether or not a specific position of each frame forming the moving image includes a specific shape subtitle, and its specification. Information Extraction Unit 10 for Extracting Predetermined Information from Video Frames Including Subtitles
3, a storage unit 104 that stores the predetermined information extracted by the information extraction unit 103, a caption-related information processing output unit 105 that outputs the predetermined information stored by the storage unit 104 in response to a user's request, and a user's Operation unit 1 for inputting requests
06 and the predetermined information extracted by the information extraction unit 103 changes, the video frame extraction unit 108 extracts the video frames before and after the change from the video storage unit 107.
And.

The video signal input by the video input unit 101 is
The video may be a digital video signal or an analog video signal. Note that the video input unit 101 is A
A mode in which a D converter is provided and D / A conversion is performed may be used. The input video signal is sent to the caption discriminating unit 102 and at the same time stored in the video storage unit 107. As the video storage unit 107, a storage medium such as a DVD or a VCR and its drive can be adopted.

The video signal input from the video input unit 101 is input to the subtitle discriminating unit 102, and it is discriminated whether or not a specific subtitle exists from each video frame. For example, in the case of a sports scene, the score status and the like are displayed on the screen. For the scoring situation, the first letter of the alphabet of the team name abbreviation or nickname (for example, "CG" in a professional baseball game, "Yoko-ura" in a professional soccer game, etc.)
Is displayed, and its display location is the same location in the frame throughout the program.

The subtitle discriminating section 102 discriminates the presence or absence of this unique subtitle from each frame. This determination can be realized by a simple pattern recognition technique because the content (type) of the subtitle and the place in the frame are the same. Therefore,
The description is omitted here. The image processing device 100 is
At the time of pattern recognition, only a subtitle having a specific shape at a specific position is discriminated, so that there is an advantage that the processing load is extremely small.

When the subtitle discriminating unit 102 discriminates that the subtitle discriminating unit 102 includes a specific subtitle, the information extracting unit 103 extracts a characteristic amount, that is, predetermined information from the video frame. For example, in the case of score display captions such as for professional baseball, the predetermined information is information related to captions such as score situations near the team name. Note that, depending on the mode of use, the information extraction unit 103 includes a very simple feature amount extraction in which the change in the score situation is extracted as information instead of recognizing the score number. In the following, a process using simple feature extraction such as the change of the score situation will be described.

In some sports, scoring scenes can be considered an important attraction in the match. Therefore, by extracting the score transition, it is possible to edit between the transitions and easily edit only the score scene. Note that subtitles are not limited to characters, but include fixed images common to videos, for example, in the case of a baseball relay, a diamond-shaped image showing how many runners are on the base. By extracting the information related to such subtitles, similarly, in the case of a baseball relay example, when there are many runners, since it is an opportunity for the attacking side, it can be digested as a climax scene of the game.

The storage unit 104 stores the feature quantities of the scene extracted by the information extraction unit 103 in time series.
The information extraction unit 103 extracts "presence or absence of score change"
Since it is a simple feature quantity such as "the number of runners", it does not require a large amount of memory. Also, the score and runner do not change in seconds or less, and there is no need to extract features for each frame.
Therefore, the storage unit 104 can be constructed by a high-speed and small-capacity storage means such as a memory buffer, and the configuration of the video processing device 100 can be simplified.

The operation unit 106 makes it possible to present the extracted feature amount to the user, and to give a command such that the user reproduces a video from the beginning of a specific scene according to the feature amount. Even in the case of a medium that cannot be randomly accessed, such as a VCR, it becomes possible to reproduce from a scene at a specific time by using the information in the storage unit 104.

The video frame extraction unit 108 is the operation unit 10.
The image is output in accordance with the command input from 6. Depending on the mode of use, it may be configured to display an image. In the configuration of FIG. 1, the video frame extraction unit 10
8 is a moving image (video signal) stored in the video storage unit 107
Although the frame extraction is performed from the above, the present invention is not limited to this, and a mode in which the frame is directly input from the video input unit 101 may be used.

[0034]

As described above, the image processing apparatus of the present invention can reduce the processing load and enable the digesting of a video scene. In other words, by using the video analysis device of the present invention, the highlight scene of a sports video can be easily extracted without using a high-level video processing device, that is, a computer having a large processing capacity, and the viewer can easily extract the highlight scene. Will be able to select and view the scene.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a video processing device of the present invention.

[Explanation of symbols]

100 video processing device 101 Video input section 102 Subtitle discrimination unit 103 information extraction unit 104 memory 105 Subtitle-related information processing output unit 106 Operation part 107 video storage 108 Video frame extraction unit

Claims (9)

[Claims] 1. A video processing device for processing a video signal of a moving image, wherein the subtitle determining means determines whether or not a specific position of each video frame forming the moving image includes a specific shaped subtitle. An information extracting unit that extracts predetermined information from a video frame including the subtitle when the subtitle determining unit determines that the subtitle is included, and a storage unit that stores the predetermined information extracted by the information extracting unit. And a processing unit for outputting the predetermined information stored by the storage unit in response to a user's request. 2. The video processing apparatus according to claim 1, wherein the information extracting means extracts information displayed near the subtitles. 3. The video processing device according to claim 2, wherein the information displayed near the subtitles is score information of sports video. 4. A video frame extracting means for extracting video frames before and after the change when the predetermined information extracted by the information extracting means changes. The video processing device according to. 5. A video processing method for processing a video signal of a video, comprising a caption determination step of determining whether a specific position of each video frame forming the video includes a subtitle of a specific shape. An information extraction step of extracting predetermined information from a video frame including the subtitle when the subtitle determination step determines that the subtitle is included, and a storage step of storing the predetermined information extracted in the information extraction step And a processing step of outputting the predetermined information stored in the storage step in response to a user's request. 6. The image processing method according to claim 5, wherein in the information extracting step, information displayed near the caption is extracted. 7. The image processing method according to claim 6, wherein the information displayed in the vicinity of the subtitle is score information of a sports image. 8. The method according to claim 5, further comprising a video frame extracting step of extracting video frames before and after the change when the predetermined information extracted in the information extracting step changes. Image processing method described in. 9. A video processing program, which causes a computer to execute the video processing method according to claim 5.