KR20010037151A - System and Method for Making Brief Video Using Key Frame Images - Google Patents

Abstract

PURPOSE: A system and a method for generating a digest video using a logo image are provided to generate a digest video and enable programs for generating the digest video to be read through a computer by extracting logo image information among cuts constituting a video and using the entire or a part of the logo image information to produce the digest video. CONSTITUTION: A logo image information extracting unit(12) extracts information concerning a logo image representing each cut of video data inputted from the outside. A storing unit(13) stores the logo image information extracted by the logo image information extracting unit(12). A digest video selecting unit(14) selects a digest video through the logo image information stored by the storing unit(13). A digest video generating unit(15) generates a digest video(16) by extracting an image from the video data through the digest video selected by the digest video selecting unit(14). The present invention has advantages of digesting a video and automatically producing a digest video of less size in a rapid time by extracting logo images among cuts constituting the video and selecting the entire or a part of the extracted logo images.

Description

System and Method for Making Brief Video Using Key Frame Images}

The present invention relates to a system for generating a summary video and a method thereof and a computer-readable recording medium recording a program for implementing the method. In particular, the summary video shows a summary of the video by summarizing the video into several representative images. A summary video generation system for making a set of images displayed at a specific time interval, and a method thereof and a computer-readable recording medium having recorded thereon a program for realizing the method.

In the prior art, there is a method of artificially editing and producing and displaying a video. As described above, in order to produce a video by artificially editing, a separate device for editing a video is required, and a lot of time and effort are required for selecting or extracting a screen to be edited.

In addition, the conventional method has a problem in that, in most cases, the capacity of the edited video is large enough to be unsuitable for services through the Internet.

The present invention has been made to solve the above problems, and extracts the representative image information of the cut (part of the video that is considered to be a continuous scene, most of the video consists of several to several thousand to tens of thousands of cuts) constituting the video It is an object of the present invention to provide a summary video generating system for producing a summary video using all or part thereof, and a method and a computer readable recording medium recording a program for realizing the method.

The present invention also provides a summary video generation system for extracting representative image information of a cut constituting a video and the corresponding representative image and using a part thereof to create a summary video, and a method and a program for realizing the method. Another object is to provide a computer readable recording medium.

The present invention also provides a computer-readable summary recording system for extracting a representative image of a cut constituting a video and creating a summary video using all of them, and a method and a program for realizing the method. Another purpose is to provide a recording medium.

1 is a configuration diagram of an embodiment of a summary video generation system using a representative image according to the present invention;

2 is another embodiment of the summary video generation system using a representative image in accordance with the present invention.

3 is a flowchart illustrating an embodiment of a method of operating a summary video generating system according to the present invention;

4 is an overall flowchart of another embodiment of a method of operation of a summary video generation system according to the present invention;

FIG. 5 is a detailed flowchart of one embodiment of a process 33 and 43 for extracting information of a current frame shown in FIGS. 3 and 4.

6 is a detailed flowchart of one embodiment of a process 34 and 44 for determining whether a current frame shown in FIGS. 3 and 4 is a representative image.

FIG. 7 is a detailed flowchart of an embodiment of a process (37,48) of selecting the summary image shown in FIGS. 3 and 4;

8 is a detailed flowchart of an exemplary embodiment of a representative image removing process 73 using similarity between representative images shown in FIG. 7.

9 is a detailed flowchart of an embodiment of the similarity calculation and storage process 81 of the representative image shown in FIG. 8;

FIG. 10 is a detailed flowchart of an embodiment of a representative image removal process 75 using the time interval shown in FIG. 7.

FIG. 11 is a detailed flowchart of an embodiment of the summary video generation process 38, 49 shown in FIGS. 3 and 4.

* Explanation of symbols for the main parts of the drawings

12: representative image information extractor 13: representative image information table

14: summary image selector 15: summary video generator

A first apparatus of the present invention for achieving the above object is a summary video generation system, comprising: representative image information extraction means for extracting information on a representative image representing each cut of video data input from the outside; Storage means for storing representative image information extracted by said representative image information extracting means; Summary image selecting means for selecting a summary image using representative image information stored in said storage means; And summary video generating means for generating a summary video by extracting an image from the video data using the summary image selected by the summary image selecting means.

In addition, the second apparatus of the present invention, in the summary video generation system, the information on the representative image representing each cut of the video data input from the outside, the representative image information and representative image extraction means for extracting the representative image; ; Storage means for storing the representative image information and the representative image information extracted by the representative image extraction means and the corresponding representative image; Summary image selection means for selecting a summary image from among the representative images stored using the representative image information stored in the storage means; And a summary video generating means for generating a summary video using the summary image selected by the summary image selecting means.

The third apparatus of the present invention also provides a summary video generating system comprising: representative image information extracting means for extracting information on a representative image representing each cut of video data input from the outside; Storage means for storing representative image information extracted by said representative image information extracting means; And summary video generating means for generating a summary video by extracting an image from the video data using representative image information stored in the storage means.

The fourth apparatus of the present invention also provides a summary video generating system comprising: representative image extracting means for extracting a representative image representing each cut of video data input from the outside; Storage means for storing the representative image extracted by said representative image extracting means; And a summary video generating means for generating a summary video using the representative image stored in the storage means.

On the other hand, the first method of the present invention, a summary video generating method applied to the summary video generating system, the first step of extracting and storing information about the representative image representing each cut of the input video data; A second step of selecting a summary image by using the stored representative image information; And extracting an image from the video data using the selected summary image to generate a summary video.

In addition, the second method of the present invention, in the summary video generation method applied to the summary video generation system, a first step of extracting and storing information about the representative image representing each cut of the input video data and the representative image; ; A second step of selecting a summary image from among the stored representative images by using the stored representative image information; And a third step of generating a summary video using the selected summary image.

According to another aspect of the present invention, there is provided a summary video generation method applied to a summary video generation system, comprising: a first step of extracting information about a representative image representing each cut of input video data; A second step of storing the extracted representative image information; And generating a summary video by extracting an image from the video data by using the stored representative image information.

In addition, the fourth method of the present invention includes a summary video generation method applied to a summary video generation system, comprising: a first step of extracting a representative image representing each cut of input video data; A second step of storing the extracted representative image; And generating a summary video using the stored representative image.

On the other hand, the present invention, the summary video generation system to summarize the entire video, the first function for extracting and storing information about the representative image representing each cut of the input video data; A second function of selecting a summary image by using the stored representative image information; And a computer readable recording medium having recorded thereon a program for realizing a third function of generating a summary video by extracting an image from the video data using the selected summary image.

The present invention also provides a summary video generation system for summarizing and displaying a video, comprising: a first function of extracting and storing information about a representative image representing each cut of input video data and a corresponding representative image; A second function of selecting a summary image from among the stored representative images by using the stored representative image information; And a computer readable recording medium having recorded thereon a program for realizing a third function of generating a summary video using the selected summary image.

The present invention also provides a summary video generation system for summarizing and displaying a video, comprising: a first function of extracting information about a representative image representing each cut of input video data; A second function of storing the extracted representative image information; And a computer-readable recording medium having recorded thereon a program for realizing a third function of generating a summary video by extracting an image from the video data using the stored representative image information.

The present invention also provides a summary video generation system for summarizing and displaying a video, comprising: a first function of extracting a representative image representing each cut of input video data; A second function of storing the extracted representative image; And a computer readable recording medium having recorded thereon a program for realizing a third function of generating a summary video using the stored representative image.

As described above, the present invention selects a small number of images which are smaller in capacity than the methods provided by the prior art for video-related services using a network or the like, by quickly / automatically selecting a small number of similar images so that the selected image is fixed. It creates a summary video showing the time interval.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram of an embodiment of a summary video generation system using a representative image according to the present invention.

The summary video generation system using the representative image according to the present invention can be divided into one or more cuts constituting the video by selecting or inputting a video signal or video file 11 from the outside and representing each cut. The representative image information extractor 12 which determines the representative image and extracts the information of the representative image, and stores the information of the extracted representative image in the form of a data structure existing in a database (DB), a file system, or a memory or a hard disk of a computer. A summary image selector 14 for selecting a summary image capable of summarizing the entire video among the representative images using the representative image information table 13 and the representative image information in the representative image information table 13; Video signal input from the outside using the summary image selected by the summary image selector 14 Extracting an image from a video file and a summary of the video summary video generator 15 to create (16).

As shown in FIG. 1, in extracting the representative image, only the image related information is extracted without extracting the image itself, and the image is extracted from the video signal or the video file input from the outside using the summary image when creating the summary video. Alternatively, in another embodiment, in extracting the representative image, the image itself may be extracted together with the image related information. Another embodiment of the present invention will be described in detail with reference to FIG. 2.

2 is a configuration diagram of another embodiment of a summary video generation system using a representative image according to the present invention.

A summary video generation system according to another embodiment of the present invention is a representation that can be represented by representing each cut as one or more cuts constituting a video by selecting or inputting a video signal or video file 21 from the outside. Representative image information and representative image extractor 22 which determines an image and extracts the representative image together with the representative image information, and extracts the extracted representative image information and the corresponding representative image from a database or a file system or a memory or hard disk of a computer. The entire video from among the representative images stored using the representative image information and the representative image table 23 stored in the form of a data structure existing on the disk, and the representative image information existing in the representative image information and the representative image table 23. And a summary image selector 24 to select a summary image to summarize the image. Creating a summary of an image selected by the image selector in the Summary 24. Summary video 26 and a summary of the video generator (25).

In the above two embodiments, a system for generating a summary video by extracting a representative image of a cut constituting a video and then selecting a portion of the representative image as a summary image has been described. The system may be implemented to generate a summary video by using a modified example of FIGS. 1 and 2, which are not separately attached. Looking at that another embodiment in more detail as follows.

Summary video generation system according to another embodiment according to the present invention, selecting a video signal or video file from the outside or receives as an input representative image that can be represented by representing one or more cuts each of which constitutes a video Representative image information extractor that determines and extracts the information of the representative image, and the representative image information table that stores the extracted representative image information in the form of a database or file system or a data structure existing in the memory or hard disk of a computer. And a summary video generator for extracting an image from a video signal or a video file input from the outside using information of the representative image existing in the representative image information table to generate a summary video.

Summary video generation system according to another embodiment according to the present invention, selecting a video signal or video file from the outside or receives as an input representative image that can be represented by representing one or more cuts each of which constitutes a video Representative image extractor to extract, representative image table that stores the extracted representative image in the form of data structure existing in database (DB), file system or computer memory or hard disk, and the representative image present in the representative image table And a summary video generator for creating a summary video.

In addition, the above embodiments may be combined and implemented. That is, the two embodiments may be combined by selecting whether to use all or part of the representative image and determine whether to store the representative image.

A detailed operation of the summary video generation system according to each embodiment will be described in detail with reference to the drawings below with reference to FIG. 3.

3 is a flowchart illustrating an embodiment of a method of operating a summary video generating system according to the present invention.

First, a video signal or a video file to be made of a summary video is selected or received as an input (31), and an input video is selected (31) and the input of the selected video signal or video file is checked (32). If the input of the video has not been terminated, the current frame information of the video is extracted (33) and the current frame is checked to be the representative image of the new cut constituting the video (34). If the current frame is a representative image, the extracted representative image information is stored in the representative image information table (35), and the current frame is designated as a reference frame for selecting a new representative image. Later 36, the process proceeds to step 32, where the video input is terminated.

On the other hand, if the video input is completed, after selecting the summary image from the representative image by the number of the summary images necessary for the summary video configuration using the information of the stored representative image information table (37), the input video signal or video using the selected summary image The image is extracted from the file to generate a summary video (38), and then ends.

4 is an overall flowchart of another embodiment of a method of operating a summary video generating system according to the present invention.

First, a video signal or video file to be made of a summary video is selected or received as an input, and an input video is selected (41), and it is checked whether the input of the selected video signal or video file is terminated (42). If the input of the video is not terminated, the current frame information of the video is extracted (43) and the current frame is checked to be the representative image of the new cut constituting the video (44). If the current frame is not the representative image, the video input is terminated. If the current frame is the representative image, the extracted representative image information is stored in the representative image information and the representative image table (45), and the current frame is selected for selecting a new representative image. After designating the frame (46), the representative image is extracted and stored in the representative image information and the representative image table (47), and the process proceeds to step 42 of checking whether the video input is terminated.

On the other hand, if the video input is completed, after selecting a summary image capable of summarizing the entire video from the stored representative images using the representative image information and the representative image information present in the representative image table (48), the selected summary image is selected. After the summary video is generated (49), the process ends.

In the two exemplary embodiments of the method, a method of generating a summary video by extracting a representative image of a cut constituting a video and then selecting a portion of the representative image as a summary image has been described. It is also possible to generate a summary video using all of them (the modifications of FIGS. 3 and 4 are not attached to the drawings). Looking at that another embodiment in more detail as follows.

According to another embodiment of the present invention, a method for generating a summary video includes first selecting a video signal or a video file to be produced as a summary video or accepting as an input to select an input video, and checking whether the input of the selected video signal or video file is finished. Check it. If the input of the video is not terminated, the process of extracting the current frame information of the video and checking whether the current frame is the representative image of the new cut constituting the video and checking whether the input of the video is terminated if the current frame is not the representative image If the current frame is the representative image, the extracted representative image information is stored in the representative image information table, the current frame is designated as a reference frame for selecting a new representative image, and the process of checking whether the video input is terminated. To the next. On the other hand, when the video input is completed, the image is extracted from the input video signal or the video file by using the representative image information of the stored representative image information table, and the summary video is generated.

According to another embodiment of the present invention, a method for generating a summary video includes first selecting a video signal or a video file to be produced as a summary video or accepting as an input to select an input video, and checking whether the input of the selected video signal or video file is finished. Check it. If the input of the video is not terminated, the process of extracting the current frame information of the video and checking whether the current frame is the representative image of the new cut constituting the video and checking whether the input of the video is terminated if the current frame is not the representative image If the current frame is the representative image, the current frame is designated as a reference frame for selecting a new representative image, the representative image is extracted and stored in the representative image table, and the process of checking whether the video input is terminated ( Proceed to 42). On the other hand, if the video input is terminated after generating a summary video by using the representative image existing in the representative image table.

In addition, the embodiments of the method may be combined. That is, the two embodiments may be combined by selecting whether to use all or part of the representative image and determine whether to store the representative image.

Meanwhile, in another variation of the system and method of the present invention, instead of designating a frame determined as the representative image as the reference frame (36,46), the current frame is unconditionally determined regardless of whether the representative image is 34 or 44. You can also specify That is, the previous method is a method of determining whether the current frame becomes the representative image as the previous representative image, and the latter method is a method of determining whether the current frame is the representative image as the previous frame. .

FIG. 5 is a detailed flowchart of an embodiment 33 and 43 of extracting information of a current frame illustrated in FIGS. 3 and 4.

First, by reading new frame information (51) and checking whether the representative image targets only a specific frame, such as an I frame of a moving pictures expert group (MPEG) image (52), the representative image is extracted from only a specific frame. In this case, it checks whether the frame currently being input is a specific frame (53), and if it is not a specific frame, ignores the currently input frame and waits for the next frame information to be input or stops reading the current frame and reads the next frame. After proceeding (54), the process proceeds to step 53 of checking whether the current frame is a specific frame.

On the other hand, if the representative image is not extracted only for a specific frame or if the current frame is a specific frame, the current frame information is read (55) and information for determining the representative image is extracted (56) from the read frame information.

Meanwhile, in the process of reading new frame information 51, when the input is a video signal input in real time, the current input signal may not always be the start of a new frame. In this case, the signal input until the new frame starts is You can ignore it. This is because determining the representative image of the cut constituting the video does not matter even if a few frames are lost, unlike determining the cut boundary.

FIG. 6 is a detailed flowchart of an embodiment 34 and 44 for determining whether a current frame illustrated in FIGS. 3 and 4 is a representative image.

First, it is checked whether a reference frame exists (61), and if the reference frame does not exist, the current frame is determined as a representative image (65), and if the reference frame exists, the similarity between the reference frame and the current frame is calculated (62). In step 63, the calculated similarity is greater than a reference value to be a representative image. As a result of the inspection, if the similarity is greater than the reference value (if similar), it is determined that the current frame is not the representative image (64), and if the similarity is less than the reference value (if not similar), the current frame is determined as the representative image (65). Returns.

In step 63 of checking the similarity and reference value of FIG. Can be used. By using this, a representative image can be extracted very quickly when using a video file such as MPEG having DCT component information of each frame constituting the video as an input.

FIG. 7 is a detailed flowchart of an embodiment of selecting the summary images 37 and 48 shown in FIGS. 3 and 4.

First, it is checked whether the number of the representative images is larger than the number of the summary images necessary for producing the summary video (71). If the number of the representative images is not large, it is returned. If the number of the representative images is large, the number of images to be removed is determined (72). Using similarity between them, we remove as many representative images as we can remove (73) and check if the number of removed images is less than the number of images to be removed (74). After the representative image is further removed using the time interval (75), the process returns.

In another modification of FIG. 7, only one of the removal using the similarity or the removal using the time interval may be adopted, or the removal order using the similarity may be adopted after the first removal using the time interval.

In another variation of FIG. 7, a compound similarity may be calculated using a time interval and similarity together, and a summary image may be selected using the compound similarity. By doing so, the representative image having similarity or compound similarity below a certain reference value is not excluded from the selection of the summary image only because the similarity is large, and the summary image only because the time interval is small for the representative image having a time interval above a certain reference value. Can be excluded from selection.

FIG. 8 is a detailed flowchart of a representative image removing process 73 using similarity between representative images shown in FIG. 7.

First, the similarity of each representative image is calculated and stored (81), and then the similarity threshold for preventing the removal of the representative image with less similarity is calculated (82), and the minimum similarity for removing the given number of representative images ( The similarity of the image that is finally removed when the representative image is removed in a given number, starting from the one with the highest similarity) (83), and after specifying a larger value among the similarity threshold and the minimum similarity as the minimum similarity (84) If the representative image to be removed remains (85) and if there are no representative images to be removed any more, the number of the removed representative images is returned (90). If there is a representative image to be removed, a single representative image is selected (86) and the similarity of this image is greater than the minimum similarity (87). ), If greater than the minimum similarity, the representative image is removed (88), and it is checked whether the representative image has been removed by the given number (89). If the number of the representative images has been removed, the number of removed images is returned (90), and if the number of the representative images has not been removed, the process proceeds to a process 85 of checking whether there are any remaining representative images to be removed.

In calculating the similarity threshold (82), a threshold value used to extract the representative image may be used or a value may be specified that is not related to the number of images to be removed, such as a separate specific threshold, or may be removed based on a specific threshold. The threshold may be calculated and specified to have a displacement depending on the number of images. An example of a threshold with displacement is Tnew = Tbasic × (1 + Threshold can be determined. Where R is the number of representative images to be removed, T is the number of representative images, α is 0.5+ (Tavg-Tbasic), Tavg is the mean value of the similarity of the representative images, and Tbasic is an arbitrary reference value.

If the removal using the time interval, which is a variation of FIG. 7, is performed first and the representative image is removed later using the similarity, the removal using the time interval may be similar to that of FIG. 8.

FIG. 9 is a detailed flowchart of an embodiment of the similarity calculation and storage process 81 of the representative image illustrated in FIG. 8.

First, it reads the representative image set that collects the representative images for similarity calculation (91) and checks whether the representative image exists in the set (92) and returns if the representative image does not exist, and returns one representative from the set if the representative image exists. After selecting the image (93) and removing the selected representative image from the set (94), it calculates the similarity with the remaining representative images in the set, sets the largest value among the calculated similarities as the similarity of the selected representative image, and saves it. (95) Proceed to step 92 of checking whether a representative image exists in the set.

FIG. 10 is a detailed flowchart of an exemplary embodiment of a representative image removing process 75 using the time interval shown in FIG. 7.

First, the time interval of each representative image is calculated (101), and the maximum time interval for removing a given number of representative images (starting from the smallest time interval, and sequentially removing the given number of representative images in turn) After extraction (102), it is checked whether the representative image is removed by the given number (102). If the given number of the representative images is removed (103), the number of the removed representative images is returned (107). If it is not possible to remove the representative image by a given number, select one representative image (104) and check whether the time interval of this image is smaller than the maximum time interval (105). If it is smaller than the maximum time interval, the process removes the representative image (106) and checks whether the representative image has been removed by the given number (103).

Meanwhile, in calculating the time interval, the difference between the time (or frame number) of the previous representative image and its own time (or frame number) becomes the time interval, but the time interval calculated to consider the similarity of the image. You can also specify the time divided by the similarity.

On the other hand, if the removal using a time interval, which is a variation of FIG. 7, is performed first and then the representative image is removed using similarity, or only by the similarity, the removal using similarity may be performed by taking a form similar to that of FIG. 10.

FIG. 11 is a detailed flowchart illustrating an embodiment of the summary video generation process 38 and 49 illustrated in FIGS. 3 and 4. The process of generating a summary video when the method illustrated in FIGS. 3 and 4 is combined is shown. Explaining.

First, by reading the summary image information from the information table (111) and checking whether the image itself is included in the summary image information (112), if it is included, select the image (115) to generate a summary video (116), the image itself If is not included, after reading the information necessary for image extraction (113) to extract a summary image (114) to generate a summary video using the extracted summary image (116) and the end of the operation.

On the other hand, if the control of the video signal is difficult to include the representative image itself in the table by the method as shown in Figure 4 and to generate a summary video through the process of 111,115,116, but the representative image is not all the summary image Because it wastes time and space for creating a representative image that is not needed, if the video signal or information such as MPEG file can be controlled, only the representative image information should be included in the table as shown in FIG. 3 and summarized through the process of 111, 113, 114, and 116. It is a good idea to create an image. There may be a number of ways to produce a summary video, but it is preferable to display the summary image at regular intervals, such as a moving GIF.

The present invention described above is not limited to the above-described embodiments and the accompanying drawings, and various substitutions, modifications, and changes can be made in the art without departing from the technical spirit of the present invention. It will be apparent to those of ordinary knowledge.

As described above, the present invention extracts a representative image of a cut constituting a video, selects all or a part of the extracted representative image to summarize the entire video, and automatically displays a summary video having a small size in a short time. There is an effect that can be made.

Claims (40)

In the summary video generation system, Representative image information extracting means for extracting information about a representative image representing each cut of video data input from the outside; Storage means for storing representative image information extracted by said representative image information extracting means; Summary image selecting means for selecting a summary image using representative image information stored in said storage means; And Summary video generating means for generating a summary video by extracting an image from the video data using the summary image selected by the summary image selecting means. Summary video generation system comprising a. In the summary video generation system, Representative image information and representative image extracting means for extracting information on a representative image representing each cut of video data input from the outside, and extracting the representative image; Storage means for storing the representative image information and the representative image information extracted by the representative image extraction means and the corresponding representative image; Summary image selection means for selecting a summary image from among the representative images stored using the representative image information stored in the storage means; And Summary video generating means for generating a summary video using the summary image selected by the summary image selecting means. Summary video generation system comprising a. In the summary video generation system, Representative image information extracting means for extracting information about a representative image representing each cut of video data input from the outside; Storage means for storing representative image information extracted by said representative image information extracting means; And Summary video generating means for generating a summary video by extracting an image from the video data using representative image information stored in the storage means. Summary video generation system comprising a. In the summary video generation system, Representative image extracting means for extracting a representative image representing each cut of video data input from the outside; Storage means for storing the representative image extracted by said representative image extracting means; And Summary video generation means for generating a summary video using the representative image stored in the storage means Summary video generation system comprising a. The method according to any one of claims 1 to 4, The summary video generating means, A summary video generation system, characterized in that for generating a summary video showing the extracted summary image at a specific time interval. The method according to any one of claims 1 to 4, Each said extraction means, And a criterion for determining whether the current frame becomes a representative image is a previous representative image. The method according to any one of claims 1 to 4, Each said extraction means, And a criterion for determining whether the current frame becomes a representative image is a previous frame. The method according to any one of claims 1 to 4, Each said extraction means, And determining the representative image that the similarity with the reference image is less than a predetermined threshold. The method of claim 8, The similarity checking process, A summary video generation system characterized by checking the similarity using a direct current (DC) component histogram, a DC component difference, or both in the DCT constant of the current frame. The method according to any one of claims 1 to 4, Each said extraction means, A summary video generation system characterized by targeting only a specific frame of input video data. The method according to claim 1 or 2, The summary image selection means, Summary video generation system, characterized in that for selecting the summary image using the similarity between the representative image. The method according to claim 1 or 2, The summary image selection means, Summary video generation system, characterized in that for selecting the summary image using the time interval. The method according to claim 1 or 2, The summary image selection means, And a summary image is selected by removing the unnecessary representative image using a similarity and then removing the remaining unnecessary representative image using a time interval. The method according to claim 1 or 2, The summary image selection means, And a summary image is selected by removing the unnecessary representative image using a time interval and then removing the remaining unnecessary representative image using similarity. The method according to claim 1 or 2, The summary image selection means, Computing a compound similarity using a time interval and similarity together, and then selecting a summary image using the compound similarity. The method according to claim 1 or 2, The summary image selection means, Representative images with similarity or complex similarity below a certain reference value should not be excluded from the selection of the summary image only because the similarity is large, and representative images with time intervals above a certain reference value only need to be selected in the summary image only because the time interval is small. And a reference value in consideration of the number of representative images and the number of similar images so as not to be excluded. The method of claim 16, The reference value is, Tnew = Tbasic × (1 + A summary video generation system, characterized in that). Where R is the number of representative images to be removed, T is the number of representative images, α is 0.5+ (Tavg-Tbasic), Tavg is the mean value of the similarity of the representative images, and Tbasic is any reference value. In the summary video generation method applied to the summary video generation system, A first step of extracting and storing information about a representative image representing each cut of the input video data; A second step of selecting a summary image by using the stored representative image information; And A third step of extracting an image from the video data using the selected summary image to generate a summary video; Summary video generation method comprising a. In the summary video generation method applied to the summary video generation system, Extracting and storing information about a representative image representing each cut of the input video data and the corresponding representative image; A second step of selecting a summary image from among the stored representative images by using the stored representative image information; And A third step of generating a summary video using the selected summary image; Summary video generation method comprising a. In the summary video generation method applied to the summary video generation system, Extracting information about a representative image representing each cut of the input video data; A second step of storing the extracted representative image information; And A third step of generating a summary video by extracting an image from the video data by using the stored representative image information; Summary video generation method comprising a. In the summary video generation method applied to the summary video generation system, Extracting a representative image representing each cut of the input video data; A second step of storing the extracted representative image; And A third step of generating a summary video by using the stored representative image Summary video generation method comprising a. The method according to any one of claims 18 to 21, The summary video generation process, Summary video generation method characterized in that for generating a summary video showing the extracted summary image at a specific time interval. The method according to any one of claims 18 to 21, Each extraction process of the first step, And determining whether the current frame becomes the representative image based on the previous representative image. The method of claim 23, wherein The determination process, And determining that the representative image has a similarity with the reference image less than a predetermined threshold as the representative image. The method of claim 24, And a similarity is checked using a DC (current current) histogram, DC component difference, or both in the DCT constant of the current frame at the predetermined threshold. The method according to any one of claims 18 to 21, Each extraction process of the first step, And determining whether the current frame becomes a representative image based on the previous frame. The method of claim 26, The determination process, And determining that the representative image has a similarity with the reference image less than a predetermined threshold as the representative image. The method of claim 27, And a similarity is checked using a DC (current current) histogram, DC component difference, or both in the DCT constant of the current frame at the predetermined threshold. The method according to any one of claims 18 to 21, Each extraction process of the first step, And a specific frame of the input video data. The method of claim 18 or 19, The process of selecting a summary image of the second step, A summary video generation method comprising selecting a summary image using similarity between representative images. The method of claim 18 or 19, The process of selecting a summary image of the second step, The summary video generation method of selecting a summary image using a time interval. The method of claim 18 or 19, The process of selecting a summary image of the second step, A fourth step of removing unnecessary representative images using similarity; And The fifth step of selecting summary images by removing remaining unnecessary representative images using time intervals Summary video generation method comprising a. The method of claim 19 or 19, The process of selecting a summary image of the second step, A fourth step of removing unnecessary representative images using a time interval; And The fifth step of selecting summary images by removing remaining unnecessary representative images using similarity Summary video generation method comprising a. The method of claim 18 or 19, The process of selecting a summary image of the second step, A fourth step of calculating a compound similarity using the time interval and the similarity together; And A fifth step of selecting a summary image using the calculated complex similarity diagram Summary video generation method comprising a. The method of claim 18 or 19, The process of selecting a summary image of the second step, Representative images with similarity or complex similarity below a certain reference value should not be excluded from the selection of the summary image only because the similarity is large, and representative images with time intervals above a certain reference value only need to be selected in the summary image only because the time interval is small. In order not to be excluded, the summary video generation method characterized in that the reference value is set in consideration of the number of the representative image and the number of similar images. 36. The method of claim 35 wherein The setting reference value is, Tnew = Tbasic × (1 + Method for generating a summary video, characterized in that). Where R is the number of representative images to be removed, T is the number of representative images, α is 0.5+ (Tavg-Tbasic), Tavg is the mean value of the similarity of the representative images, and Tbasic is any reference value. In order to summarize the entire video, the summary video generation system, A first function of extracting and storing information about a representative image representing each cut of the input video data; A second function of selecting a summary image by using the stored representative image information; And A third function of generating a summary video by extracting an image from the video data using the selected summary image A computer-readable recording medium having recorded thereon a program for realizing this. In order to summarize the entire video, the summary video generation system, A first function of extracting and storing information about a representative image representing each cut of the input video data and the corresponding representative image; A second function of selecting a summary image from among the stored representative images by using the stored representative image information; And A third function of generating a summary video using the selected summary image A computer-readable recording medium having recorded thereon a program for realizing this. In order to summarize the entire video, the summary video generation system, A first function of extracting information about a representative image representing each cut of the input video data; A second function of storing the extracted representative image information; And A third function of generating a summary video by extracting an image from the video data by using the stored representative image information A computer-readable recording medium having recorded thereon a program for realizing this. In order to summarize the entire video, the summary video generation system, A first function of extracting a representative image representing each cut of the input video data; A second function of storing the extracted representative image; And A third function of generating a summary video by using the stored representative image A computer-readable recording medium having recorded thereon a program for realizing this.