JP2019169768A - Broadcast editing device, broadcasting system and image processing method - Google Patents

Abstract

Provided is a technique for preventing a broadcasted program from being in a so-called sponsored state. When a reproduced video is acquired (S1), a video character check unit performs text extraction processing by image analysis, detects characters in the image, converts the characters into text, and outputs a detection result (S2). ). Next, the sponsor management server 20 provides the sponsor information list X1 of the program (S3). The video character check unit 13 compares the sponsor information list X1 with the detected character list Y1 to determine whether or not a text to be highlighted is included (S4). A display process for highlighting the corresponding text in the list Y1 and surrounding the corresponding area of the video screen A1 with the marker M1 is performed (S5). [Selection] Fig. 3

Description

  The present invention relates to a broadcast editing apparatus, a broadcast system, and an image processing method. For example, the broadcast editing apparatus, broadcast system, and image processing method are provided with a function of detecting characters included in a video that is subjected to image processing and transmitted as a broadcast. About.

  The broadcast video (video signal) is generally sent after editing the recorded material. When editing video, create original material information data, which is information extracted from the relationship between the original material for editing and the edited material, and when editing again, the edited material, project data, and original material information data There is known a technique of editing by using (see, for example, Patent Document 1).

JP 2012-34218 A

  By the way, commercial broadcasters have sponsors for each program, but the characters reflected in the programs such as produced dramas and variety (for example, characters written on the performers' T-shirts, can juice, background signs, etc.) ) Is not covered by the sponsor (that is, whether or not a manufacturer competing with the sponsor is displayed on the screen) is visually checked by the editor and the director. Specifically, in the case of a program with “A car company” as a sponsor, a signboard of “B car company” is prevented from being reflected in the background. Normally, as a pre-preview work before broadcasting, editors and directors replay the video every time it is edited and visually check it. If the video taken by the sponsor is reflected, edit the scene again. We do work such as cutting with or mosaic blurring the signboard. However, because of the work by human power, it is difficult to eliminate the oversight, and a countermeasure technique has been demanded.

  The present invention has been made in view of such a situation, and an object thereof is to solve the above problems.

The broadcast editing apparatus according to the present invention compares a text extraction unit that detects characters included in a reproduced video and converts them into text, and sponsor information associated with the video and the text. And a matching text display unit for displaying that the text matches.
The match text display unit may superimpose a mark indicating the match on characters included in the video.
The match text display unit may display and output characters included in the video as a list.
The broadcast system of the present invention includes the above-described broadcast editing apparatus, a transmission server that transmits video edited by the broadcast editing apparatus, and a sponsor information server that holds the sponsor information.
The image processing method for a video according to the present invention compares a text extraction step of detecting characters contained in a reproduced video and converting it into text, and sponsor information associated with the video and the text, and they match. A matching text display step for displaying the text so that it can be recognized that the text matches.

  ADVANTAGE OF THE INVENTION According to this invention, the program (namely, video signal) to broadcast can be prevented from becoming a so-called sponsor cover state, and the technique which can reduce the work load of an editor or a director can be provided.

1 is a block diagram illustrating a schematic configuration of a video editing system according to an embodiment. It is a block diagram which shows schematic structure of the editing apparatus based on embodiment. It is a figure explaining the text extraction and emphasis processing in the video based on the sponsor information by the video editing system according to the embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a block diagram showing a schematic configuration of a video editing system 1 according to the present embodiment. The video editing system 1 includes a camera 10, a recording device 11, a recording device 12 (video server), an editing device 14, a transmission server 18, a media playback device 19, a sponsor management server 20, and a system control unit 15. Are connected via a network 2 such as a LAN line or a predetermined communication line. The system control unit 15 controls the entire video editing system 1 in an integrated manner, and may be configured independently or included in the same manner as other devices (such as the recording device 12 and the editing device 14). May be.

  The camera 10 performs a digital conversion process on a video image captured by a CCD (Charge Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor) device, etc., and converts video data (for example, HD-SDI standard material video data) Using the recording device 11, the data is output to the recording device 12 via the network 2. The recording device 12 stores this.

  When the media playback device 19 receives a playback instruction from the editing device 14 or the transmission server 18, the media playback device 19 plays back and outputs a video file in the media 5 (optical media 5 a, magnetic media 5 b, VTR tape 5 c) according to the instructions.

  As a data storage function, the recording device 12 includes a material video data unit 121 that records material video data, an edited video data unit 122 that records edited video data, and a final editing information unit 123 that records final editing information. A low-resolution server 124 for recording low-resolution files.

  The sponsor management server 20 records sponsor information for each program. Sponsor information includes, for example, a list of sponsor names, product names / service names provided by sponsors (hereinafter collectively referred to as “product names”), company names that compete with sponsors, and product names associated with programs. It is recorded. When there is a request for sponsor information from the editing device 14, the sponsor management server 20 transmits sponsor information of the designated program.

  FIG. 2 is a functional block diagram of the editing device 14. The editing device 14 includes an editing control unit 141, a display unit 142, an operation panel 143, and a video character check unit 13.

  The editing control unit 141 actually performs editing work and controls the editing device 14 in an integrated manner. The display unit 142 is a display (display unit) that displays material video data and a video based on the video data after editing. The operation panel 143 is an operation means for selecting each part in the video and audio or inputting an instruction. The display unit 142 and the operation panel 143 may be configured as an integrated touch panel display 144.

  The video character check unit 13 includes a processing target recognition unit 131 and an information storage unit 132. Note that the video character check unit 13 may be provided as a video character check device independent of the editing device 14.

  The processing target recognition unit 131 includes a text extraction unit 133, a text enhancement unit 134, and a text list output unit 135.

  The text extraction unit 133 reads material video data (video), and detects and extracts characters contained in the video using a known character recognition / document image analysis technique. Specifically, a character area in the video is extracted, and after the character area is specified, character recognition is performed to extract text (that is, a character string).

  Character area extraction techniques include binarization techniques (local binarization techniques, image decomposition techniques based on multiple divisions of gray value histograms, etc.), and character / text enhancement techniques with higher resolution, contrast correction, and focus Techniques such as deviation and camera shake correction are applied as necessary.

  When character recognition is performed after the character region is specified, various recognition methods in consideration of the characteristics of the character image acquired by the camera can be applied. For example, a method of constructing a subspace using images of multiple frames that make up a video (moving image), or a Gabor wavelet transform after normalizing the luminance level to remove the influence of non-uniformity of character images There is a method to extract.

  In recent years, with the development of image analysis technology, mobile terminals such as smartphones have been equipped with character recognition technology in video, and while correcting geometric distortion and shading in video, The feature amount of a character can be extracted from the frame (still image) and recognized as a text (character string). Such technology has greatly improved recognition accuracy, particularly due to recent advances in artificial intelligence technology and deep learning technology.

  The text extraction unit 133 collates the extracted text with the sponsor information recorded in the sponsor management server 20, and determines whether there is a matching text (character string). If there is a matching text, the text extraction unit 133 associates the position (time code and coordinates) where the character string exists with the matched text, sponsor information, program information (ID of the material video data, etc.) Records in the storage unit 132.

  The text emphasizing unit 134 highlights the extracted text portion by surrounding it with a mark in the video. Other highlighting methods are: That is, it is possible to perform brightness level adjustment, color correction, contrast adjustment, sharpness (outline emphasis) adjustment, etc. of the target portion (the character area in the video and its surroundings). In addition, generation of enlarged image that enlarges a part of the image, generation of reduced image, soft focus, mosaic, luminance inversion, highlight display (highlighting) of a part of the image, change of the atmosphere of the whole color, etc. Image effect processing, separation and synthesis of images for divided display of captured images, generation of character images and image images, processing of combining generated images with captured images, and the like can also be performed.

  By such display processing, an operator such as an editor or a director can reliably recognize characters included in the video. In the present embodiment, highlighting is performed when sponsor information matches, but different sponsors may be easily confirmed by highlighting when they do not match.

  The operator can set what kind of processing is performed on the extracted text. For example, if the extracted text is a sponsor's competitor, a red marker is displayed, if the competitor's product name is a blue marker, and if the sponsor is a sponsor's name, there is processing such as character outline highlighting. Further, it may be possible to set whether or not to perform enhancement processing according to the size of characters in the video. In other words, in the case of small characters, if the user who viewed the transmitted broadcast cannot recognize it immediately, it may not be necessary to edit that part. , The burden on the operator can be reduced.

  The text list output unit 135 lists the extracted character strings as a list, and displays them, for example, in the side of a video or in a pop-up frame. Further, the text list output unit 135 emphasizes that the text list output unit 135 can be distinguished from other text if it matches the sponsor information. In addition to the text, it is also possible to display the relationship with the sponsor, that is, whether it is a sponsor name, a product name, a competitor name, or the like.

  The text extraction / display processing in the video reflecting the sponsor information by the video editing system 1 described above will be briefly described with reference to FIG.

  As shown in the figure, the video image A1 includes human objects (person 1) A2 and object (person 2) A3, signboard object (signboard 1) A4, and object (signboard 2) A5. ing.

  For example, the object (person 1) A2 is wearing a T-shirt on which “Ah” is written. Object (person 2) A3 wears a T-shirt labeled “Ii” and has a can labeled “AA”. The object (signboard 1) A4 has “UUU”, and the object (signboard 2) A5 has “YES”.

  When the video character check unit 13 (text extraction unit 133) of the editing device 14 acquires the reproduced video (S1), the video character check unit 13 performs text extraction processing by image analysis on the video, detects characters in the video, and converts them into text. The detection result is output (S2). For example, as shown in the detected character list Y1, “Ah”, “No”, “Wow”, “Yes, yes” and “AA” are obtained as a list.

  Next, when the video character check unit 13 (text extraction unit 133) requests the sponsor information about the program to be processed from the sponsor management server 20, the sponsor management server 20 sponsor information as sponsor information associated with the program. A list X1 is provided (S3).

  The video character check unit 13 (text extraction unit 133) compares the sponsor information list X1 and the detected character list Y1 to determine whether or not text to be highlighted is included (S4). When the text to be highlighted is included, that is, when characters such as the name or product of the sponsor's competing competitor are included, the video character check unit 13 (text list output unit 135) detects the detected character. The corresponding text in the list Y1 is highlighted. Here, “UUUU” is detected and highlighted by black and white inversion.

  Further, the video character check unit 13 (text emphasizing unit 134) performs a display process of surrounding the area including “UUUU” to be highlighted on the video screen A1 with the marker M1, and emphasizes the portion (S5). .

  In this way, by emphasizing the names of competitors, products, etc. on the video, editing cuts and blurring can be done before broadcasting, preventing the broadcast program from becoming a so-called sponsored state Can do. Therefore, it is possible to reduce the work load on the editor and the director.

The features of the above-described embodiment can be summarized as follows.
In the video editing system 1 (broadcast system), the video character check unit 13 (video character check device) of the editing device 14 detects a character included in the reproduced video and converts it into text, A text emphasis unit 134 (matching text display unit) is provided that compares the sponsor information associated with the text with each other and displays them so that the texts match when they match.

  The text emphasizing unit 134 (matched text display unit) may superimpose a mark indicating the match on the characters included in the video.

  The text emphasis unit 134 (match text display unit) may display and output characters included in the video as a list. That is, after the characters in the video (screen) are extracted as text, they are displayed in a frame as a list, for example.

  The video editing system 1 (broadcast system) includes a video character check unit 13 (video character check device) of the editing device 14, a transmission server 18 that transmits the video edited by the video character check unit 13, and sponsor information. Holding sponsor management server 20.

  The image processing method for the video in the video editing system 1 compares the text extraction step of detecting characters contained in the reproduced video and converting it into text, the sponsor information associated with the video, and the text, and matches them. A matching text display step for displaying the text so that it can be recognized that the text matches.

  The present invention has been described based on the embodiments. This embodiment is an exemplification, and it will be understood by those skilled in the art that various modifications can be made to combinations of these components, and such modifications are also within the scope of the present invention.

1 Video Editing System 2 Network 5 Media 10 Camera 11 Recording Device 12 Recording Device 13 Video Character Checking Unit 14 Editing Device 15 System Control Unit 18 Transmission Server 19 Media Playback Device 20 Sponsor Management Server 131 Processing Object Recognition Unit 132 Information Storage Unit 133 Text Extraction unit 134 Text enhancement unit 135 Text list output unit 141 Edit control unit 142 Display unit 143 Operation panel 144 Touch panel display

Claims (5)

A text extraction unit that detects characters contained in the reproduced video and converts them into text;
A match text display unit that compares sponsor information associated with the video and the text, and displays the text so that it can be recognized that the text matches if they match;
A broadcast editing apparatus comprising:   The broadcast editing apparatus according to claim 1, wherein the match text display unit superimposes a mark indicating the match on characters included in the video.   The broadcast editing apparatus according to claim 1, wherein the match text display unit displays and outputs characters included in the video as a list. The broadcast editing apparatus according to any one of claims 1 to 3,
A transmission server for transmitting video edited by the broadcast editing device;
A sponsor information server that holds the sponsor information;
A broadcasting system comprising: A text extraction step of detecting characters contained in the reproduced video and converting them to text (character strings);
A match text display step for comparing the sponsor information associated with the video with the text and displaying the text so that it can be recognized that the text matches if they match;
An image processing method for video, comprising:

Images