CN103079041B - The method of news video automatic strip-cutting device and news video automatic strip - Google Patents
The method of news video automatic strip-cutting device and news video automatic strip Download PDFInfo
- Publication number
- CN103079041B CN103079041B CN201310032025.5A CN201310032025A CN103079041B CN 103079041 B CN103079041 B CN 103079041B CN 201310032025 A CN201310032025 A CN 201310032025A CN 103079041 B CN103079041 B CN 103079041B
- Authority
- CN
- China
- Prior art keywords
- frame
- ewscaster
- head portrait
- mute
- news video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Television Signal Processing For Recording (AREA)
- Studio Circuits (AREA)
Abstract
A kind of news video automatic strip-cutting device, comprising: extraction module, memory module, feature identification module, judge module and fractionation module.Wherein, extraction module extracts mute frames all in news video automatically; Memory module prestores sound and the head portrait of ewscaster, and stores mute frame and time point thereof; Feature identification module extracts the frame containing ewscaster's sound in news video and the frame containing ewscaster's head portrait; Judge module judges whether the mute frame in news video is the frame comprising ewscaster's head portrait, and whether the frame after judging mute frame when mute frame is the frame comprising ewscaster's head portrait comprises head portrait and the sound of ewscaster; Split when the frame of module after mute frame comprises ewscaster's head portrait and sound and mute frame is set to start frame, and from the time point that start frame is corresponding, news video is split.When utilizing the present invention to news video itemize, efficiency is high, and speed is fast, and saves human cost.
Description
Technical field
The present invention relates to Video processing, particularly relate to a kind of method of news video automatic strip-cutting device and news video automatic strip.
Background technology
Current news splits or is pure to split by hand, or is when starting according to new theme news, and incipient a few frame is general all quiet, splits.
Manual fractionation length is the news video of N minute, and often need the time being several times as much as N minute, cost of labor is too high, splits speed too slow.
Split according to " when new theme news starts; incipient a few frame is general all quiet ", often the switching cutting of camera lens in same theme news in order to different theme news, as long as because Shot change, all may cause " incipient a few frame is general all quiet ", although so this technology is better than pure manual fractionation, after carrying out first cutting by this technology, still need manually to carry out a large amount of error correction work.
Summary of the invention
In view of this, be necessary to provide a kind of to news video automatic strip, the method that saves labour turnover.
News video automatic strip-cutting device provided by the invention, comprises extraction module, feature identification module, judge module and splits module.Wherein, extraction module, for automatically extracting mute frames all in news video; Memory module, for prestoring sound and the head portrait of ewscaster, and stores mute frame and time point thereof; Feature identification module, for extracting the frame containing ewscaster's sound in news video and extracting the frame containing ewscaster's head portrait in news video according to described ewscaster's head portrait according to the sound characteristic of described ewscaster; Judge module, for judging whether the described mute frame in news video is the frame comprising ewscaster's head portrait, wherein, described judge module is also used in when described mute frame is the frame comprising ewscaster's head portrait and judges whether the frame after described mute frame comprises head portrait and the sound of ewscaster; Split module, when comprising ewscaster's head portrait and sound for the frame after described mute frame, described mute frame is set to start frame, and from the time point that described start frame is corresponding, news video is split.
The present invention also provides a kind of method of news video automatic strip, comprises the following steps: extract the mute frame in described news video; Read the sound characteristic of the ewscaster of news video; The frame containing ewscaster's sound in news video is extracted according to the sound characteristic of described ewscaster; Read the frame containing ewscaster's head portrait in news video; The frame containing ewscaster's head portrait in news video is extracted according to described ewscaster's head portrait; Judge in described mute frame whether be the frame comprising described ewscaster's head portrait; If so, then judge whether comprise ewscaster's sound and head portrait in the frame after described mute frame; If described mute frame is set to start frame, and from the time point that described start frame is corresponding, news video is split.
The method of news video automatic strip-cutting device provided by the invention and news video automatic strip by mute frame, comprise ewscaster's head portrait frame and comprise ewscaster's voiced frame comprehensive descision and come go to start and the time point of correspondence, to split news video, effectively raise the fractionation efficiency of news video, reduce artificial workload.
Accompanying drawing explanation
Fig. 1 is the module map of news video automatic strip-cutting device in an embodiment of the present invention;
Fig. 2 is the flow chart of news video automatic strip method in an embodiment of the present invention.
Embodiment
Be described below in detail embodiments of the invention, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has element that is identical or similar functions from start to finish.Being exemplary below by the embodiment be described with reference to the drawings, only for explaining the present invention, and can not limitation of the present invention being interpreted as.
In describing the invention, term " interior ", " outward ", " longitudinal direction ", " transverse direction ", " on ", D score, " top ", the orientation of the instruction such as " end " or position relationship be based on orientation shown in the drawings or position relationship, be only the present invention for convenience of description instead of require that the present invention with specific azimuth configuration and operation, therefore must can not be interpreted as limitation of the present invention.
Refer to Fig. 1, Figure 1 shows that the module map of news video automatic strip-cutting device 10 in an embodiment of the present invention.
In the present embodiment, news video automatic strip-cutting device 10 comprises: acquisition module 102, feature identification module 104, judge module 106, fractionation module 108, processor 110 and memory module 112, acquisition module 102, feature identification module 104, judge module 106 and fractionation module 108 are stored in memory module 112, and processor 110 is for performing the modules in memory module 112.
In the present embodiment, extraction module 102 is for automatically extracting mute frames all in news video.
Memory module 112 for prestoring sound and the head portrait of ewscaster, and stores mute frame and time point thereof.
Feature identification module 104 is for extracting the frame containing ewscaster's sound in news video and extracting the frame containing ewscaster's head portrait in news video according to described ewscaster's head portrait according to the sound characteristic of described ewscaster.
In the present embodiment, described memory module 112 time point containing mute frame is stored as quiet set B=B1, B2 ... Bn}, the time point of the frame of the sound of ewscaster is stored as sound set C={C1, C2 ... the time point of the frame of Cm} and the head portrait containing ewscaster is head portrait set D={D1, D2 ..., Dk}.
In the present embodiment, the sound of described ewscaster and described ewscaster is one_to_one corresponding.
Judge module 106 is for judging whether the described mute frame in news video is the frame comprising ewscaster's head portrait.In the present embodiment, described judge module 106 is also used in when described mute frame is the frame comprising ewscaster's head portrait and judges whether the frame after described mute frame comprises head portrait and the sound of ewscaster.
In the present embodiment, judge module 106 is also for judging that when described mute frame is not the frame comprising ewscaster's head portrait whether described mute frame for being last mute frame, if not, then read next mute frame and continue to judge whether described mute frame is the frame comprising ewscaster's head portrait.
In the present embodiment, when judge module 106 does not also comprise ewscaster's head portrait and sound for the frame after described mute frame, then represent that the frame after described mute frame is still mute frame, continue to judge whether described mute frame is the frame comprising ewscaster's head portrait.
In the present embodiment, described quiet set B and described head portrait set D are carried out intersection operation and are obtained the first common factor E={E1 by judge module 106, E2, Ej}, also obtain the second common factor F={F1 for described sound set C and described head portrait set D is carried out intersection operation, F2, Fi}, wherein, described judge module 106 also obtains postponing set G={F1 for all elements in described second common factor F being subtracted one, F2, Fi-1}, and described delay set G and described first common factor E is sought common ground obtain the 3rd common factor H={H1, H2, Hp}, frame wherein in the 3rd common factor H corresponding to time point is " before there is the head portrait of ewscaster and the frame of sound and that frame of the head portrait of ewscaster appears in mute frame " simultaneously.
In the present embodiment, judge module 106 also just the first common factor E={E1, E2,, in Ej}, continuous time point puts into same subset, then gather E '={ S1, S2,, Sq}, wherein S1, S2,, Sq is the subset containing continuous time point, and the time point in each subset is according to time order arrangement from small to large.
In the present embodiment, judge module 106 is analyzed for each subset in E ' again, such as, subset Si is analyzed, if judge module 106 finds containing any one element in the 3rd common factor H in subset Si, then first time point that judge module 106 is determined in subset Si is the time point of a new theme news start frame.
Split when module 108 comprises ewscaster's head portrait and sound for the frame after described mute frame and described mute frame is set to start frame, and from the time point that described start frame is corresponding, news video is split.
Refer to Fig. 2, Figure 2 shows that the flow chart of news video automatic strip in an embodiment of the present invention.
In the present embodiment, the method for news video automatic strip comprises the following steps:
In step s 200, extraction module 102 extracts the mute frame in described news video.In the present embodiment, memory module 112 prestores sound and the head portrait of ewscaster, and stores mute frame and time point thereof.
In the present embodiment, described memory module 112 time point containing mute frame is stored as quiet set B=B1, B2 ..., Bn}.
In step S202, feature identification module 104 reads the sound characteristic of the ewscaster of news video.
In step S204, feature identification module 104 extracts the frame containing ewscaster's sound in news video according to the sound characteristic of described ewscaster.In the present embodiment, the time point of the frame of the sound of ewscaster is stored as sound set C={C1 by described memory module 112, C2 ..., Cm}.
In step S206, feature identification module 104 reads the frame containing ewscaster's head portrait in news video.
In step S208, feature identification module 104 extracts the frame containing ewscaster's head portrait in news video according to described ewscaster's head portrait.In the present embodiment, the time point of the frame of the head portrait of described memory module 112 containing ewscaster is head portrait set D={D1, D2 ..., Dk}
In step S210, whether judge module 106 judges in described mute frame is the frame comprising described ewscaster's head portrait.In the present embodiment, described quiet set B and described head portrait set D are carried out intersection operation and are obtained the first common factor E={E1, E2 by judge module 106 ... Ej}, the frame that the time point in the first common factor E is corresponding is " quiet and occur the frame of newsreader's head portrait ".
If for comprising the frame of described ewscaster's head portrait in mute frame, then in step S212, judge module 106 judges whether comprise ewscaster's sound and head portrait in the frame after described mute frame.In the present embodiment, described sound set C and described head portrait set D is carried out intersection operation and obtains the second common factor F={F1 by judge module 106, F2, Fi}, wherein, described judge module 106 also obtains postponing set G={F1 for all elements in described second common factor F being subtracted one, F2, Fi-1}, and described delay set G and described first common factor E is sought common ground obtain the 3rd common factor H={H1, H2, Hp}, frame wherein in the 3rd common factor H corresponding to time point is " before there is the head portrait of ewscaster and the frame of sound and that frame of the head portrait of ewscaster appears in mute frame " simultaneously.
In the present embodiment, judge module 106 also just the first common factor E={E1, E2,, in Ej}, continuous time point puts into same subset, then gather E '={ S1, S2,, Sq}, wherein S1, S2,, Sq is the subset containing continuous time point, and the time point in each subset is according to time order arrangement from small to large.
In the present embodiment, judge module 106 is analyzed for each subset in E ' again, such as, subset Si is analyzed, if judge module 106 finds containing any one element in the 3rd common factor H in subset Si, then first time point that judge module 106 is determined in subset Si is the time point of a new theme news start frame.
If comprise ewscaster's sound and head portrait in the frame after mute frame, then in step S214, split module 108 and mute frame is set to start frame.
In step S216, split module 108 and from the time point that described start frame is corresponding, news video is split.
In the present embodiment, if when not comprising ewscaster's sound and head portrait in the frame after mute frame, then represent that the frame after described mute frame is still mute frame, need get back to step S210, judge module 106 continues to judge in mute frame whether be the frame comprising described ewscaster's head portrait.
If described mute frame is not the frame comprising described ewscaster's head portrait, then in step S218, judge module 106 judges whether mute frame is last mute frame.
If mute frame is not last mute frame, then in step S220, judge module 206 reads next mute frame.
In step S210, judge module 206 carries out judging whether described mute frame is the frame comprising described ewscaster's head portrait.
News video automatic strip-cutting device 10 in embodiment of the present invention and the method for news video automatic strip by mute frame, comprise ewscaster's head portrait frame and comprise ewscaster's voiced frame comprehensive descision and come go to start and the time point of correspondence, to split news video, effectively raise the fractionation efficiency of news video, reduce artificial workload.
Although the present invention is described with reference to current better embodiment; but those skilled in the art will be understood that; above-mentioned better embodiment is only used for the present invention is described; not be used for limiting protection scope of the present invention; any within the spirit and principles in the present invention scope; any modification of doing, equivalence replacement, improvement etc., all should be included within the scope of the present invention.
Claims (5)
1. a news video automatic strip-cutting device, comprising:
Extraction module, for automatically extracting mute frames all in news video;
Memory module, for prestoring sound and the head portrait of ewscaster, and stores mute frame and time point thereof, also stores frame and the time point thereof of the frame of sound containing ewscaster and time point and the head portrait containing ewscaster simultaneously;
Feature identification module, for extracting the frame containing ewscaster's sound in news video and extracting the frame containing ewscaster's head portrait in news video according to described ewscaster's head portrait according to the sound characteristic of described ewscaster;
Judge module, for judging whether the described mute frame in news video is the frame comprising ewscaster's head portrait, wherein, described judge module is also used in when described mute frame is the frame comprising ewscaster's head portrait and judges whether the frame after described mute frame comprises head portrait and the sound of ewscaster;
Split module, when comprising ewscaster's head portrait and sound for the frame after described mute frame, described mute frame is set to start frame, and from the time point that described start frame is corresponding, news video is split.
2. news video automatic strip-cutting device as claimed in claim 1, it is characterized in that, described judge module is also for judging when described mute frame is not the frame comprising ewscaster's head portrait whether described mute frame is last mute frame, if not, then read next mute frame and continue to judge whether described mute frame is the frame comprising ewscaster's head portrait.
3. news video automatic strip-cutting device as claimed in claim 2, it is characterized in that, when described judge module does not also comprise ewscaster's head portrait and sound for the frame after described mute frame, then represent that the frame after described mute frame is still mute frame, continue to judge whether described mute frame is the frame comprising ewscaster's head portrait.
4. news video automatic strip-cutting device as claimed in claim 1, it is characterized in that, described ewscaster and described sound are one_to_one corresponding.
5. a method for news video automatic strip, comprising:
Extract the mute frame in described news video;
Read the sound characteristic of the ewscaster of news video;
The frame containing ewscaster's sound in news video is extracted according to the sound characteristic of described ewscaster;
Read the frame containing ewscaster's head portrait in news video;
The frame containing ewscaster's head portrait in news video is extracted according to described ewscaster's head portrait;
Judge in described mute frame whether be the frame comprising described ewscaster's head portrait;
If comprise the frame of described ewscaster's head portrait in described mute frame, then judge whether comprise ewscaster's sound and head portrait in the frame after described mute frame;
If comprise ewscaster's sound and head portrait in the frame after described mute frame, then described mute frame is set to start frame, and from the time point that described start frame is corresponding, news video is split;
If do not comprise ewscaster's sound and head portrait in the frame after mute frame, then continue to judge the frame whether comprising described ewscaster's head portrait in mute frame;
If described mute frame does not comprise the frame of described ewscaster's head portrait, then judge whether mute frame is last mute frame;
If mute frame is not last mute frame, then read next mute frame; If mute frame is last mute frame, then split end.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310032025.5A CN103079041B (en) | 2013-01-25 | 2013-01-25 | The method of news video automatic strip-cutting device and news video automatic strip |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310032025.5A CN103079041B (en) | 2013-01-25 | 2013-01-25 | The method of news video automatic strip-cutting device and news video automatic strip |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103079041A CN103079041A (en) | 2013-05-01 |
| CN103079041B true CN103079041B (en) | 2016-01-27 |
Family
ID=48155440
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310032025.5A Active CN103079041B (en) | 2013-01-25 | 2013-01-25 | The method of news video automatic strip-cutting device and news video automatic strip |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103079041B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101031035A (en) * | 2006-03-03 | 2007-09-05 | 广州市纽帝亚资讯科技有限公司 | Method for cutting news video unit automatically based on video sequence analysis |
| CN101616264A (en) * | 2008-06-27 | 2009-12-30 | 中国科学院自动化研究所 | News video categorization and system |
| CN101855897A (en) * | 2007-11-14 | 2010-10-06 | 皇家飞利浦电子股份有限公司 | A method of determining a starting point of a semantic unit in an audiovisual signal |
| CN102073631A (en) * | 2009-11-19 | 2011-05-25 | 凌坚 | Video news unit dividing method by using association rule technology |
| CN102547139A (en) * | 2010-12-30 | 2012-07-04 | 北京新岸线网络技术有限公司 | Method for splitting news video program, and method and system for cataloging news videos |
-
2013
- 2013-01-25 CN CN201310032025.5A patent/CN103079041B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101031035A (en) * | 2006-03-03 | 2007-09-05 | 广州市纽帝亚资讯科技有限公司 | Method for cutting news video unit automatically based on video sequence analysis |
| CN101855897A (en) * | 2007-11-14 | 2010-10-06 | 皇家飞利浦电子股份有限公司 | A method of determining a starting point of a semantic unit in an audiovisual signal |
| CN101616264A (en) * | 2008-06-27 | 2009-12-30 | 中国科学院自动化研究所 | News video categorization and system |
| CN102073631A (en) * | 2009-11-19 | 2011-05-25 | 凌坚 | Video news unit dividing method by using association rule technology |
| CN102547139A (en) * | 2010-12-30 | 2012-07-04 | 北京新岸线网络技术有限公司 | Method for splitting news video program, and method and system for cataloging news videos |
Non-Patent Citations (1)
| Title |
|---|
| 新闻视频场景分割技术研究;李默;《中国人民解放军信息工程大学硕士毕业论文》;20060220;论文正文第39页倒数第2－3段、论文正文第40页倒数第1－4段、论文正文第42页第3－7段以及图4－2和图4－3 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103079041A (en) | 2013-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9396397B2 (en) | Methods and systems for object-recognition and link integration in a composite video stream | |
| US10241585B2 (en) | Smart watch and gesture input method for the smart watch | |
| CN103488983A (en) | Business card OCR data correction method and system based on knowledge base | |
| RU2017102521A (en) | HUMAN RECOGNITION METHOD, DEVICE AND TERMINAL | |
| US20150319373A1 (en) | Method and device to compose an image by eliminating one or more moving objects | |
| CN110691202A (en) | Video editing method, device and computer storage medium | |
| CN111798836B (en) | Method, device, system, equipment and storage medium for automatically switching languages | |
| CN106331559A (en) | Method and system for intelligent video recognition on aeration of sewage reservoir | |
| CN113889113A (en) | Sentence dividing method and device, storage medium and electronic equipment | |
| CN102098449A (en) | Method for realizing automatic inside segmentation of TV programs by utilizing mark detection | |
| CN102724516B (en) | Adaptive signal processing method, device and decoder | |
| WO2011038099A3 (en) | System and method for document location and recognition | |
| CN105550207B (en) | Information popularization method and device | |
| CN113515536A (en) | Map updating method, device, equipment, server and storage medium | |
| EP2738675A3 (en) | System and method for efficient resource management of a signal flow programmed digital signal processor code | |
| CN103079041B (en) | The method of news video automatic strip-cutting device and news video automatic strip | |
| CN111241217A (en) | Data processing method, device and system | |
| CN103986981B (en) | The recognition methods of the plot fragment of multimedia file and device | |
| CN103064935A (en) | System and method for multimedia data parallel processing | |
| CN104734797A (en) | Time synchronization method and electronic devices | |
| CN113242431A (en) | Road side perception oriented labeled data preprocessing method | |
| CN110633663B (en) | Method for automatically cutting multi-mode data in sign language video | |
| CN109635707A (en) | A kind of video lens extracting method based on feature identification | |
| EP2521071A3 (en) | Method and system for text segmentation | |
| CN115080770A (en) | Multimedia data processing method and device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |