CN103093213B - Video file classification method and terminal - Google Patents

Video file classification method and terminal Download PDF

Info

Publication number
CN103093213B
CN103093213B CN201310032781.8A CN201310032781A CN103093213B CN 103093213 B CN103093213 B CN 103093213B CN 201310032781 A CN201310032781 A CN 201310032781A CN 103093213 B CN103093213 B CN 103093213B
Authority
CN
China
Prior art keywords
video file
image information
human face
face image
bar
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
Application number
CN201310032781.8A
Other languages
Chinese (zh)
Other versions
CN103093213A (en
Inventor
曾元清
夏威
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Oppo Mobile Telecommunications Corp Ltd
Original Assignee
Guangdong Oppo Mobile Telecommunications Corp Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdong Oppo Mobile Telecommunications Corp Ltd filed Critical Guangdong Oppo Mobile Telecommunications Corp Ltd
Priority to CN201310032781.8A priority Critical patent/CN103093213B/en
Publication of CN103093213A publication Critical patent/CN103093213A/en
Application granted granted Critical
Publication of CN103093213B publication Critical patent/CN103093213B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Processing Or Creating Images (AREA)
  • Television Signal Processing For Recording (AREA)

Abstract

The invention discloses a kind of video file classification method, comprise the following steps: extract all of human face image information in video file;Human face image information described in comparison and the original records bar of storage in data base successively;Update described original records bar according to comparison result or generate new record bar;Generating corresponding file according to the described original records bar after updating or described new record bar, each file deposits the video file containing identical human face image information.Compared with prior art, the video file classification method of the present invention achieves and the video file including identical facial image is divided into a class, facilitate user search specific someone all videos, easy and simple to handle, save the lookup time and Consumer's Experience is effective.The present invention discloses a kind of video file classification terminal.

Description

Video file classification method and terminal
Technical field
The present invention relates to document classification technical field, relate more specifically to a kind of video file classification method and end End.
Background technology
The most perfect along with mobile terminal function, people are possible not only to utilize mobile terminal to take pictures, more Achieve and utilize mobile terminal to carry out recorded video, as household or friend's recorded video.
But, the video that user is recorded by current mobile terminal be usually with the time name, be i.e. with time Between successively to visual classification, which results in when mobile terminal stores more video file, very The difficult video picking out the someone wished to from various video file, needs user to click on one by one and regards Frequency makes a look up, and complex operation, loses time and Consumer's Experience is the highest.
Therefore, it is necessary to provide a kind of video file classification method and terminal to overcome drawbacks described above.
Summary of the invention
It is an object of the invention to provide a kind of video file classification method, thus realize including identical face figure The video file of picture is divided into a class, and then facilitates user to search the video of specific someone, easy and simple to handle, joint Save time and improve Consumer's Experience.
It is a further object of the present invention to provide a kind of video file classification terminal, thus realize to include same person The video file of face image is divided into a class, and then facilitates user to search the video of specific someone, easy and simple to handle, Time-consuming and improve Consumer's Experience.
For achieving the above object, the invention provides a kind of video file classification method, comprise the following steps:
Extract all of human face image information in video file;
Human face image information described in comparison and the original records bar of storage in data base successively;
Update described original records bar according to comparison result or generate new record bar;
Corresponding file, Mei Yisuo is generated according to the described original records bar after updating or described new record bar State file and deposit the video file containing identical human face image information.
Compared with prior art, the video file classification method of the present invention is by carrying out face knowledge to video file Not thus extract the face images information in video file, the most successively by each human face image information Compare with the original records bar in data base, and update original records bar according to comparison result or generate new Record strip, the new record bar finally according to the original records bar updated or generation generates corresponding file, its In in each file storage have the video file comprising identical human face image information, it is achieved thereby that will include The video file of identical human face image information is divided into a class and stores up in same file presss from both sides, and facilitates user and looks into Look for all video files of specific someone, easy and simple to handle, time-consuming and improve Consumer's Experience.
It is preferred that described original records bar or described new record bar include following information: facial image mark, The mark of the described video file belonging to human face image information, described human face image information and described facial image The time point that information occurs in described video file.
It is preferred that also include before described " extracting all of human face image information in video file ": judge Whether described video file exists facial image, extracts all of face in video file according to judged result Image information or add an invalid value in the attribute of described video file.
It is preferred that described " update described original records bar according to comparison result or generate new record bar " is concrete For:
When described human face image information is identical with the human face image information in described original records bar, extract also Add the mark of described video file belonging to described human face image information and described human face image information described The time point occurred in video file updates described original records bar to described original records bar with realization;
When described human face image information is different from the human face image information in described original records bar, according to newly Facial image mark, described human face image information, extraction described human face image information belonging to described regard The time point generation that the mark of frequency file and described human face image information occur in described video file is described newly Record strip.
It is preferred that described facial image is designated numeral or letter.
Correspondingly, present invention also offers a kind of video file classification terminal, including:
Data base, for storing video file, original records bar and the file of generation;
Extraction module, is used for extracting all of human face image information in video file;
Comparing module, the described human face image information extracted for extraction module described in comparison successively is with described The original records bar of storage in data base;
Record strip processing module, for updating described original records bar according to comparison result or generating new record bar To be stored in described data base;
File generation module, for generating according to the described original records bar after updating or described new record bar Corresponding file, each described file deposits the video file containing identical human face image information.
It is preferred that described original records bar or described new record bar include following information: facial image mark, The mark of the described video file belonging to human face image information, described human face image information and described facial image The time point that information occurs in described video file.
It is preferred that described terminal also includes:
Judging treatmenting module, for judging whether there is face figure in the described video file in described data base Picture also indicates the work of described extraction module according to judged result or adds one in the attribute of described video file Invalid value.
It is preferred that described record strip processing module specifically includes:
Record strip updating block, for when the face figure in described human face image information with described original records bar When information is identical, extract and add mark and the institute of described video file belonging to described human face image information State time point that human face image information occurs in described video file to described original records bar to realize renewal Described original records bar;
Record strip signal generating unit, for when the face figure in described human face image information with described original records bar During as information difference, according to new facial image mark, described human face image information, the described face of extraction Mark and the described human face image information of the described video file belonging to image information go out in described video file Existing time point generates described new record bar.
It is preferred that described facial image is designated numeral or letter.
By description below and combine accompanying drawing, the present invention will become more fully apparent, and these accompanying drawings are used for explaining Embodiments of the invention.
Accompanying drawing explanation
Fig. 1 is the flow chart of video file classification method one embodiment of the present invention.
Fig. 2 is the flow chart of another embodiment of video file classification method of the present invention.
Fig. 3 is the structured flowchart of video file classification terminal one embodiment of the present invention.
Detailed description of the invention
With reference now to accompanying drawing, describing embodiments of the invention, element numbers similar in accompanying drawing represents similar unit Part.
Refer to Fig. 1, describe the workflow of video file classification method one embodiment of the present invention.Such as Fig. 1 Shown in, comprise the following steps:
Step S101, extracts all of human face image information in video file;Wherein carry out recognition of face and then Extract human face image information and have employed face recognition technology based on video flowing, particularly as follows: by video file solution Code becomes video flowing, is identified each two field picture in video flowing successively, finds out all of facial image letter Breath, each key feature dot information during wherein human face image information is specially facial image, such as the face information of people Deng;
Step S102, successively human face image information described in comparison and the original records bar of storage in data base;Its In middle data base, the original records bar of storage includes following information: facial image mark, human face image information, The time that the mark of the video file belonging to human face image information and human face image information occur in video file Point, wherein the mark of video file can be that title or the attribute etc. of video file can represent this video file Arbitrary parameter;Facial image mark can be digital, alphabetical or other, different facial image mark generations The human face image information that table is different, the most different facial image marks represents different people, such as some record Bar is: 1+aa+ video A+03:45, and wherein 1 is the facial image letter that face image identification, aa represent correspondence When breath, video A represent video file mark, 03:45 represents that human face image information occurs in video A Between point be moment of 3 minutes and 45 seconds;Each key feature dot information during wherein aa represents facial image;
Step S103, updates original records bar according to comparison result or generates new record bar;Particularly as follows: work as institute State human face image information consistent with the human face image information in described original records bar (such as: record strip 1) time, The mark extracting the video file at described human face image information place and the time point occurred in video file are also Add to original records bar (record strip 1), it is achieved the renewal to original records bar (record strip 1);When When described human face image information is inconsistent with the human face image information in described original records bar, according to new people The mark of the video file belonging to face image identification, described human face image information, described human face image information and The time point that described human face image information occurs in video file generates new record bar;
Step S104, generates corresponding file according to the original records bar after updating or new record bar, wherein Original records bar after each renewal or the most corresponding file of new record bar, in file, storage has All video files that in corresponding record strip, human face image information is corresponding, it may be assumed that achieve and will include same person The video file of face image information is stored in a file, completes video file according to facial image Classify, facilitate user to search.
Refer to Fig. 2, describe the flow chart of another embodiment of video file classification method of the present invention.Such as Fig. 2 Shown in, comprise the following steps:
Step S201, starts mobile terminal, and in the automatic scan database of mobile terminal system, all of storage regard Frequency file also takes first video file;As, in data base, storage has video file A, B and C, takes first Individual video file A;
Step S202, carries out recognition of face and judges whether there is facial image in video file video file, If it is not, then perform step S203, otherwise, perform step S204;As, video file A is carried out face knowledge Not and judge whether video file A exists facial image, if it is not, then perform step S203, otherwise, hold Row step S204;
Step S203, adds an invalid value in the attribute of video file;As, step S202 judges video literary composition Part A does not exist facial image, therefore in the attribute of video file A, adds an invalid value;
Step S204, extracts all of human face image information in video file;As, extract video file A In two human face image information aa comprising and bb;
Step S205, successively by the human face image information in video file and data base storage all initially Record strip is compared;As, in data base, storage has 3 original records bars, is respectively as follows: 1+aa+ video D+04:21,2+bb+ video E+03:20 and 3+cc+ video F+02:21, first by the face in video file A Image information aa is compared successively with 3 original records bars in data base, then by video file A Human face image information bb and 3 original records bar comparisons successively in data base;
Step S206, it is judged that whether there is described human face image information in original records bar, if existing, then holds Row step S207, otherwise, then perform step S208;
Step S207, extracts and adds the mark of video file belonging to described human face image information and regard at this The time point occurred in frequency file is in the original records bar comprising described human face image information, it is achieved to initially The renewal of record strip;Human face image information aa and bb and people is obtained after video file A is identified The time point that face image information aa and bb occur in video file A, such as respectively 04:32 and 02:13, Human face image information aa is compared successively with the original records bar in data base, obtains data base comprises Identical human face image information aa, now is updated obtaining to original records bar 1+aa+ video D+04:21 Original records bar after renewal: 1+aa+ video D+04:21+ video A+04:32;Afterwards by human face image information Bb and 3 original records bars are compared successively, complete original records bar: 2+bb+ according to comparison result The renewal of video E+03:20, the original records bar after being updated: 2+bb+ video E+03:20+ video A+02:13;That is: each record strip is corresponding to a facial image, therefore when video file exists multiple face During image, the mark of this video file can be present in multiple record strip;
Step S208, according to new facial image mark, described human face image information, described facial image letter The mark of the described video file belonging to breath and described human face image information occur in described video file time Between point generate described new record bar;In the present embodiment, facial image mark is digital and arranges in order, such as: In data base, in existing original records bar, facial image mark includes 1-6, then face in the new record bar generated Image identification is 7, such as: if being identified obtaining human face image information dd to video file A, after comparison Find data base does not exist identical human face image information dd, extract human face image information dd at video file The time point occurred in A is 01:30, and generates new record bar: 7+dd+ video A+01:30;
Step S209, it is judged that whether described video file is last, the most then perform step S211, Otherwise, then step S210 is performed;As, it is judged that whether video file A is last in data base, if It is then to perform step S211, otherwise, perform step S210;
Step S210, takes next video file and performs step S202;As, take video file B;
Step S211, generates respectively according to the new record bar of each original records bar after updating or generation accordingly File, in file storage have correspondence record strip in all of video file;As: after updating Original records bar: 1+aa+ video D+04:21+ video A+04:32 generates corresponding file, file Entitled: 1+D+A, it is, of course, also possible to be other naming rules, wherein file 1+D+A store and have Video file D and the concordance list of video file A, every index in this concordance list represents a video respectively File, clicks on every index and just can open the video file of correspondence, and video file D and A includes same One face information aa;According to new record bar: 7+dd+ video A+01:30 generates corresponding file 7+A, literary composition In part folder 7+A, storage has video file A;
Step S212, completes the categorizing process to video file.
Wherein, when requiring to look up the video file of someone, can be carried out by the photographic head of mobile terminal Take pictures, from photo, extract corresponding human face image information and the original records bar with database purchase is carried out Comparison, find the original records bar comprising identical human face image information, and each original records bar is all to having One file, in file, storage has all video files comprising described human face image information, Jin Erke The time point occurred in video file with all video files and human face image information that find this people, Simple to operation and save the time.
Refer to Fig. 3, describe the structured flowchart of video file classification terminal one embodiment of the present invention.Such as Fig. 3 Shown in, video file classification terminal 100 includes:
Data base 10, the original records bar after storing all of video file, original records bar, renewal And the file generated;
Extraction module 11, is used for extracting all of human face image information in video file;Specifically have employed based on The face recognition process of video flowing;
Comparing module 12, the described human face image information extracted for extraction module 11 described in comparison successively with In described data base 10 storage original records bar, judge whether data base 10 exists to comprise described face The original records bar of image information;
Record strip processing module 13, updates described original records bar according to comparison result or generates new record bar; Concrete, record strip processing module 13 includes record strip updating block 131 and record strip signal generating unit 132; Record strip updating block 131 is for the facial image in described human face image information with described original records bar When information is identical, it may be assumed that when data base 10 exists the original records bar including described human face image information, carry Take and add the mark of video file belonging to described human face image information and described human face image information at video The time point occurred in file, to described original records bar, completes the renewal to original records bar;Record strip is raw Unit 132 is become to be used for when described human face image information is different from the human face image information in original records bar, According to new facial image mark, described human face image information, extraction described human face image information belonging to The time point that the mark of described video file and described human face image information occur in described video file generates Described new record bar;
File generation module 14, generates corresponding file according to the original records bar after updating or new record bar Folder, in the most each file, storage has the video that in corresponding record strip, all of video file mark is corresponding File, thus realize the classification to video file;
Judging treatmenting module 15, is used for judging whether there is facial image in described video file and according to judgement Result instruction extraction module 11 works or adds an invalid value in the attribute of video file, as added invalid value 0。
Wherein, original records bar or new record bar include following information: facial image mark, facial image are believed The mark of the video file belonging to breath, human face image information and described human face image information are at described video file The time point of middle appearance, wherein facial image is designated numeral or letter.
As described above it can be seen that video file classification terminal 100 first passes through judging treatmenting module 15 Video file being carried out recognition of face and judges whether video file exists facial image, if existing, then leading to Cross extraction module 11 and extract all of human face image information in video file, afterwards by comparing module 12 The human face image information that will identify that is compared with the original records bar in data base 10 successively, it is judged that institute State human face image information the most identical with the human face image information in original records bar, if identical, then by note Record bar updating block 131 adds the mark of the video file belonging to described human face image information and described face figure As, in the time point extremely existing original records bar that information occurs in video file, if differing, then passing through Record strip signal generating unit 132 generates corresponding new record bar and preserves to data base 10, then passes through file Generation module 14 is the file that the original records bar after each renewal or new record bar generate correspondence respectively, literary composition Part folder includes the video file that the mark of all video files in the record strip of correspondence is corresponding, finally includes The video file of identical facial image is divided into a class and is stored in same file folder, facilitates user and searches All video files of specific someone and the time point occurred in video, time-consuming and Consumer's Experience effect The best.
Above in association with most preferred embodiment, invention has been described, but the invention is not limited in disclosed above Embodiment, and amendment, the equivalent combinations that the various essence according to the present invention is carried out should be contained.

Claims (10)

1. a video file classification method, it is characterised in that comprise the following steps:
Extract all of human face image information in video file;
Human face image information described in comparison and the original records bar of storage in data base successively;
Update described original records bar according to comparison result or generate new record bar;
Corresponding file, Mei Yisuo is generated according to the described original records bar after updating or described new record bar State file and deposit the video file containing identical human face image information.
2. video file classification method as claimed in claim 1, it is characterised in that described original records bar Or described new record bar includes following information: facial image mark, human face image information, described facial image The mark of the described video file belonging to information and described human face image information occur in described video file Time point.
3. video file classification method as claimed in claim 2, it is characterised in that described " extraction video All of human face image information in file " also include before:
Judge whether described video file exists facial image, extract described video file according to judged result In all of human face image information or add an invalid value in the attribute of described video file.
4. video file classification method as claimed in claim 2, it is characterised in that described " according to comparison Result updates described original records bar or generates new record bar " particularly as follows:
When described human face image information is identical with the human face image information in described original records bar, extract also Add the mark of described video file belonging to described human face image information and described human face image information described The time point occurred in video file updates described original records bar to described original records bar with realization;
When described human face image information is different from the human face image information in described original records bar, according to newly Facial image mark, described human face image information, extraction described human face image information belonging to described regard The time point generation that the mark of frequency file and described human face image information occur in described video file is described newly Record strip.
5. video file classification method as claimed in claim 2, it is characterised in that described facial image mark Know for digital or alphabetical.
6. a video file classification terminal, it is characterised in that described terminal includes:
Data base, for storing video file, original records bar and the file of generation;
Extraction module, is used for extracting all of human face image information in video file;
Comparing module, the described human face image information extracted for extraction module described in comparison successively is with described The original records bar of storage in data base;
Record strip processing module, for updating described original records bar according to comparison result or generating new record bar To be stored in described data base;
File generation module, for generating according to the described original records bar after updating or described new record bar Corresponding file, each described file deposits the video file containing identical human face image information.
7. video file classification terminal as claimed in claim 6, it is characterised in that described original records bar Or described new record bar includes following information: facial image mark, human face image information, described facial image The mark of the described video file belonging to information and described human face image information occur in described video file Time point.
8. video file classification terminal as claimed in claim 7, it is characterised in that also include:
Judging treatmenting module, for judging whether there is face figure in the described video file in described data base Picture also indicates the work of described extraction module according to judged result or adds one in the attribute of described video file Invalid value.
9. video file classification terminal as claimed in claim 7, it is characterised in that described record strip processes Module specifically includes:
Record strip updating block, for when the face figure in described human face image information with described original records bar When information is identical, extract and add mark and the institute of described video file belonging to described human face image information State time point that human face image information occurs in described video file to described original records bar to realize renewal Described original records bar;
Record strip signal generating unit, for when the face figure in described human face image information with described original records bar During picture information difference, according to new facial image mark, described human face image information, the described face of extraction Mark and the described human face image information of the described video file belonging to image information go out in described video file Existing time point generates described new record bar.
10. video file classification terminal as claimed in claim 7, it is characterised in that described facial image mark Know for digital or alphabetical.
CN201310032781.8A 2013-01-28 2013-01-28 Video file classification method and terminal Active CN103093213B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310032781.8A CN103093213B (en) 2013-01-28 2013-01-28 Video file classification method and terminal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310032781.8A CN103093213B (en) 2013-01-28 2013-01-28 Video file classification method and terminal

Publications (2)

Publication Number Publication Date
CN103093213A CN103093213A (en) 2013-05-08
CN103093213B true CN103093213B (en) 2016-11-16

Family

ID=48205761

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310032781.8A Active CN103093213B (en) 2013-01-28 2013-01-28 Video file classification method and terminal

Country Status (1)

Country Link
CN (1) CN103093213B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106604051A (en) * 2016-12-20 2017-04-26 广州华多网络科技有限公司 Live channel recommending method and device
WO2020173314A1 (en) * 2019-02-27 2020-09-03 杭州海康威视数字技术股份有限公司 Personnel statistical method and device and electronic device
CN111553191A (en) * 2020-03-30 2020-08-18 深圳壹账通智能科技有限公司 Video classification method and device based on face recognition and storage medium
CN115460446A (en) * 2022-08-19 2022-12-09 上海爱奇艺新媒体科技有限公司 Alignment method and device for multiple paths of video signals and electronic equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101278286A (en) * 2005-09-30 2008-10-01 皇家飞利浦电子股份有限公司 Method and apparatus for enabling browsing of images
CN102355588A (en) * 2011-08-25 2012-02-15 中兴通讯股份有限公司 Method and system for synchronously browsing electronic photo albums
CN102682111A (en) * 2012-05-11 2012-09-19 广东欧珀移动通信有限公司 Picture retrieval method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101278286A (en) * 2005-09-30 2008-10-01 皇家飞利浦电子股份有限公司 Method and apparatus for enabling browsing of images
CN102355588A (en) * 2011-08-25 2012-02-15 中兴通讯股份有限公司 Method and system for synchronously browsing electronic photo albums
CN102682111A (en) * 2012-05-11 2012-09-19 广东欧珀移动通信有限公司 Picture retrieval method

Also Published As

Publication number Publication date
CN103093213A (en) 2013-05-08

Similar Documents

Publication Publication Date Title
CN103530652B (en) A kind of video categorization based on face cluster, search method and system thereof
US10242250B2 (en) Picture ranking method, and terminal
JP5212610B2 (en) Representative image or representative image group display system, method and program thereof, and representative image or representative image group selection system, method and program thereof
JP4643735B1 (en) Electronic device and video processing method
CN106126617A (en) A kind of video detecting method and server
CN112738556B (en) Video processing method and device
KR20090010855A (en) System and method for saving digital contents classified with person-based clustering
WO2017080007A1 (en) Intelligent analysis and processing system of identification code and processing method therefor
KR101832680B1 (en) Searching for events by attendants
CN103093213B (en) Video file classification method and terminal
CN105684046B (en) Generate image composition
CN106127222B (en) A kind of the similarity of character string calculation method and similitude judgment method of view-based access control model
EP3989158A1 (en) Method, apparatus and device for video similarity detection
US20150278248A1 (en) Personal Information Management Service System
CN112000024B (en) Method, device and equipment for controlling household appliance
CN104331515B (en) A kind of method and system for automatically generating tourism diary
WO2020135756A1 (en) Video segment extraction method, apparatus and device, and computer-readable storage medium
CN111368867A (en) Archive classification method and system and computer readable storage medium
CN108124478A (en) Picture searching method and apparatus
CN110825821B (en) Personnel relationship query method and device, electronic equipment and storage medium
CN109635688B (en) Method and system for managing books on bookshelf based on image recognition
CN110580297A (en) Merchant and dish matching method and device based on dish image and electronic equipment
KR101138873B1 (en) Method and apparatus for reducing the number of photo in photo album
WO2022241987A1 (en) Image retrieval method and apparatus
CN115983873A (en) Big data based user data analysis management system and method

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
CP01 Change in the name or title of a patent holder

Address after: Changan town in Guangdong province Dongguan 523860 usha Beach Road No. 18

Patentee after: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS Corp.,Ltd.

Address before: Changan town in Guangdong province Dongguan 523860 usha Beach Road No. 18

Patentee before: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS Corp.,Ltd.

CP01 Change in the name or title of a patent holder