KR20160027651A - Method and apparatus for indexing moving picture - Google Patents

Abstract

The present invention relates to a method and apparatus for indexing a video. The method for indexing the video comprises the following steps: extracting edge component images of each image frame from a plurality of image frames included in a video; individually selecting title areas of first and second image frames from the image frames by using the edge component images; and selecting at least any one from the first and second image frames as a key frame according to a result of comparing the selected title areas. Accordingly, even if a position of the title area, which may include text information, is changed in a screen, the present invention can adaptively deal with a circumstance thereof and can index the text information.

Description

METHOD AND APPARATUS FOR INDEXING MOVING PICTURE [0001]

The present invention relates to a moving picture indexing method and apparatus, and more particularly, to a method and apparatus for selecting a key frame for indexing in moving pictures.

In general, an index of broadcast contents is obtained by acquiring a finger print in each frame, generating metadata based on the finger print, adding metadata to the database, and searching for the image based on the information.

However, in the case of using the fingerprint as described above, there is a problem that it is difficult to develop a finger print algorithm that accurately classifies similar scenes, complicated operations are performed to correctly classify the frames.

As a conventional technique, Korean Patent Laid-Open Publication No. 10-2011-0039744 (published on Apr. 20, 2011) discloses an image search method.

However, such a conventional technique can be effectively applied only when the position including the text information in the moving picture is fixed. When the text information changes with time, the user must change the default value or input the corrected value . That is, there is a problem in that it can not adaptively cope with various videos having various formats.

According to an embodiment of the present invention, there is provided a video indexing method and apparatus for automatically selecting a title region in a plurality of video frames included in a moving image and then comparing the selected title region and selecting a key frame for indexing.

The problems to be solved by the present invention are not limited to those mentioned above, and another problem to be solved can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a moving image indexing method comprising: extracting an edge component image of each image frame in a plurality of image frames included in a moving image; Selecting at least one of the first image frame and the second image frame as a key frame according to a result of comparing the selected title area with the title area of the first image frame and the second image frame, Step < / RTI >

According to another aspect of the present invention, there is provided an apparatus for indexing moving images, comprising: an edge extracting unit for extracting an edge component image of each image frame in a plurality of image frames included in a moving image; A title selection unit for selecting a title area of a first image frame and a second image frame from among the first image frame and the second image frame according to a result of comparing the selected title area, And a frame selection unit for selecting a frame.

According to an embodiment of the present invention, a title area is automatically selected from a plurality of image frames included in moving images, and then a key frame for an index is selected by comparing the title areas. Index information such as a title area image of the selected key frame, a thumbnail using the key frame, or a text recognized in the title area can be specified in correspondence with the selected key frame.

Therefore, even if the position of the title area, which can include the text information, is changed within the screen of the moving picture, it can adaptively cope with the indexing process.

1 is a block diagram of a moving image indexing apparatus according to an embodiment of the present invention.
2 is a flowchart illustrating a moving image indexing method according to an embodiment of the present invention.
FIG. 3 is an exemplary view of a video index result screen according to an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. To fully disclose the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims.

In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The following terms are defined in consideration of the functions in the embodiments of the present invention, which may vary depending on the intention of the user, the intention or the custom of the operator. Therefore, the definition should be based on the contents throughout this specification.

A moving picture file such as a broadcast content which can be indexed using the moving picture indexing method and apparatus according to the embodiment of the present invention often includes a title in the title area. For example, as illustrated in FIG. 3, the title area 307 may be disposed near the upper left of the moving picture screen 301. The title displayed in the title area 307 can be composed of, for example, a character and an image, and has a feature that the change is small and the form is maintained even when a frame change of the screen occurs. The title in the title area plays a role of knowing what content the content contains, including a broadcast program name and a corner name in the program, so that the information recognized in the title area is stored in the corresponding position And can be utilized as corresponding index information. The embodiment of the present invention provides a video indexing method and apparatus capable of efficiently performing such index processing.

1 is a block diagram of a moving image indexing apparatus according to an embodiment of the present invention.

The moving image indexing apparatus 100 according to the embodiment includes an edge extracting unit 110, a title selecting unit 120, a frame selecting unit 130, an indexing unit 140, and the like.

The edge extracting unit 110 extracts an edge component image of each image frame in a plurality of image frames included in the input moving image file 10.

The title selecting unit 120 selects a title area in each of a plurality of image frames using a plurality of edge component images extracted by the edge extracting unit 110. [ The title selecting unit 120 selects a title area by comparing edge component images of a plurality of image frames adjacent to the first image frame according to a time order, It is possible to select the title area by comparing the edge component images with respect to the image frames.

The title selection unit 120 may select a title area using pixel weight map filtering, cumulative image filtering, and connected component labeling.

The frame selecting unit 130 selects a key frame for an index from a plurality of image frames according to the result of comparing the title areas selected by the title selecting unit 120. [ At least one of the first and second image frames may be selected as a key frame.

The frame selection unit 130 can compare the aspect ratio of the title area or the Euclidean distance to select a key frame.

The index unit 140 designates the index information recognized in the key frame selected by the frame selection unit 130 at the corresponding position of the progress bar.

2 is a flowchart illustrating a moving image indexing method according to an embodiment of the present invention.

As described above, the moving image indexing method according to the embodiment includes steps (S201 to S205) of extracting an edge component image of each image frame from a plurality of image frames included in the input moving image file.

The method further includes the step (S207) of selecting a title area in each of the plurality of image frames using the extracted plurality of edge component images. Here, the title area can be selected for each of the first image frame and the second image frame.

Next, a step (S209) of selecting a key frame for an index from a plurality of image frames according to a result of comparing the selected title areas. Here, at least one of the first image frame and the second image frame may be selected as a key frame.

Next, the method further includes steps (S211 and S213) of designating the index information recognized in the selected key frame to the corresponding position of the progress bar.

FIG. 3 is an exemplary view of a video index result screen according to an embodiment of the present invention.

The index information 305 can be designated and displayed at the corresponding position of the progress bar 303 arranged corresponding to the moving picture screen 301 as shown in FIG.

Hereinafter, a video indexing method performed by the video indexing apparatus according to an embodiment of the present invention will be described in detail with reference to FIG. 1 to FIG.

First, when the moving picture file 10 is input to the moving picture indexing apparatus 100 (S201), the edge extracting unit 110 extracts a plurality of picture frames included in the moving picture file 10 (S203) The edge component image of the frame is extracted (S205).

Here, extracting the edge component image of the image frame is for selecting the title area 307 in the image frame, because the title included in the image has a feature rich in edge components. For example, a Difference of Gaussian (DoG) edge detection method, a Sobel edge detection method, a prewitt edge detection method, a Harris edge detection method, and the like can be used to extract an edge component image.

Next, the title selecting unit 120 selects the title area 307 in each of the plurality of image frames using the plurality of edge component images extracted by the edge extracting unit 110. [ At this time, the title area 307 can be selected for each of the first image frame and the second image frame.

Here, the title selecting unit 120 may compare the edge component images of a plurality of image frames adjacent to the first image frame according to the time order to select the title area 307, The title region 307 can be selected by comparing the edge component images with respect to a plurality of image frames adjacent to the edge component image (S207).

The title selection unit 120 may select the title area 307 using pixel weighted map filtering, cumulative image filtering, or connected component labeling operations. For example, edge component images obtained through the DoG edge detection method include not only titles required in the present invention, but also general subtitles and backgrounds. In order to remove them, pixel weight map filtering and cumulative image filtering are utilized. The pixel weight map uses the position information of the title. The title is exposed at a fixed position for a relatively long period of time, generating a weight map using the difference between the current frame DoG and the pixel by pixel of the previous frame DoG value. Cumulative image filtering is based on temporal information of an image. In the case of a title, the value is kept constant even if a time change occurs. On the other hand, since the background is changed, it is possible to efficiently remove the background by applying the cumulative image filtering. The connected component labeling operation performs an operation of filtering the components whose sizes are too small or too large using 8-neighbor information, for example.

The frame selecting unit 130 selects a key frame for an index from a plurality of image frames according to a result of comparing the title area 307 selected by the title selecting unit 120. [ At this time, at least one of the first image frame and the second image frame may be selected as a key frame.

In order to select a key frame in the frame selection unit 130, it is necessary to recognize that a change occurs by comparing the selected title regions 307 in a plurality of image frames. Here, the frame selecting unit 130 may compare the aspect ratio of the title area 307 or the Euclidean distance to select a key frame.

Here, when the title area 307 is changed in the continuous frame of the moving picture file 10, the image frame located at the rear side will be selected as the key frame. Therefore, in the case of the forward moving picture, The second image frame is selected as the key frame. However, in the backward moving image, the first image frame is selected as a key frame. In a case where indexing is not performed for the entire moving image but only a few indexes are intermittently indexed, the key frame may be selected as a key frame when the title area 307 of the first image frame is different from that of the second image frame (S209 ).

For example, the title of the broadcast content has an aspect ratio of about 5: 2, which is two times longer than the width. The width is variable depending on the length of the subtitles and the width is affected by it. If the aspect ratio difference is more than a certain level by using this feature, it is treated as a change occurrence. Otherwise, it is treated as no change occurrence. In addition, the title has a constant brightness value, but the background has a characteristic that changes depending on the scene change. The euclidean distance between the candidate regions is calculated using the average value, variance value, and histogram value of the title regions 307, and if the number is greater than or equal to a certain level, the change occurs. Otherwise, no change occurs Lt; / RTI >

Next, the indexing unit 140 extracts information such as an image of the title area 307, a thumbnail using the key frame, and a text recognized in the title area 307 in the key frame selected by the frame selecting unit 130 And uses the recognized and extracted information as index information of the moving picture file 10 (S211). Then, the index information thus recognized is designated corresponding to the key frame. Thus, the index information 305 can be designated and displayed at the corresponding position of the progress bar 303 arranged corresponding to the moving picture screen 310 as illustrated in FIG. 3 (S213).

As described above, according to the embodiment of the present invention, a title area is automatically selected from a plurality of image frames included in a moving image, and then a key frame for an index is selected by comparing the title area. Then, And designates corresponding to the key frame. Accordingly, even if the position of the title area, which can include the text information, is changed within the screen of the moving picture, it can adaptively cope with the indexing process.

Combinations of the steps of the flowcharts appended to the present invention may be performed by computer program instructions. These computer program instructions may be loaded into a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus so that the instructions, which are executed via a processor of a computer or other programmable data processing apparatus, Lt; / RTI > These computer program instructions may also be stored in a computer usable or computer readable memory capable of directing a computer or other programmable data processing apparatus to implement the functionality in a particular manner so that the computer usable or computer readable memory It is also possible to produce manufacturing items that contain instruction means for performing the functions described in each step of the flowchart. Computer program instructions may also be stored on a computer or other programmable data processing equipment so that a series of operating steps may be performed on a computer or other programmable data processing equipment to create a computer- It is also possible for the instructions to perform the processing equipment to provide steps for executing the functions described in each step of the flowchart.

In addition, each step may represent a module, segment, or portion of code that includes one or more executable instructions for executing the specified logical function (s). It should also be noted that in some alternative embodiments, the functions mentioned in the steps may occur out of order. For example, the two steps shown in succession may in fact be performed substantially concurrently, or the steps may sometimes be performed in reverse order according to the corresponding function.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

According to an embodiment of the present invention, a title area is automatically selected from a plurality of image frames included in moving images, and then a key frame for indexing is selected by comparing the title areas. Therefore, even if the position of the title area, which can include the text information, is changed within the screen, it can adaptively cope with the index process.

The present invention can be widely used in an industrial technology field that requires index processing for various moving pictures including broadcast contents and the like.

100: Moving picture indexing device 110:
120: Title selection part 130: Frame selection part
140: index part

Claims (6)

Extracting an edge component image of each image frame from a plurality of image frames included in the moving image,
Selecting a title region of a first image frame and a title region of a second image frame among the plurality of image frames using a plurality of the edge component images,
And selecting at least one of the first image frame and the second image frame as a key frame according to a result of comparing the selected title area.
The method according to claim 1,
And specifying index information recognized in the key frame corresponding to the key frame.
3. The method of claim 2,
Wherein the index information is at least one of a title area image of the key frame, a thumbnail using the key frame, or text recognized in the title area.
An edge extracting unit for extracting an edge component image of each image frame from a plurality of image frames included in the moving image;
A title selecting unit for selecting a title area of a first image frame and a second image frame from among the plurality of image frames using the plurality of edge component images,
And a frame selecting unit for selecting at least one of the first image frame and the second image frame as a key frame according to a result of comparing the selected title area.
5. The method of claim 4,
And an indexing unit for specifying index information recognized in the key frame corresponding to the key frame.
6. The method of claim 5,
Wherein the index information is at least one of a title area image of the key frame, a thumbnail using the key frame, or a text recognized in the title area.

Images