CN103810195A - Index generating method and system - Google Patents
Index generating method and system Download PDFInfo
- Publication number
- CN103810195A CN103810195A CN201210446980.9A CN201210446980A CN103810195A CN 103810195 A CN103810195 A CN 103810195A CN 201210446980 A CN201210446980 A CN 201210446980A CN 103810195 A CN103810195 A CN 103810195A
- Authority
- CN
- China
- Prior art keywords
- event
- information
- video
- scene
- concern
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/71—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/7837—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using objects detected or recognised in the video content
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an index generating method and system. The index generating method comprises the steps of identifying scene information in video information, obtaining an attention object related to the scene information, identifying events related to the attention object, selecting an attention event in the events, extracting event description information and key frames related to the attention event, utilizing the event description information to generate an XML (extensive markup language) mark file, and utilizing the XML mark file and the key frames to generate an index. The index file is generated by identifying the scene information in the video information and querying the object and the events related to the scene information, so the information resources in the video information are fully utilized and convenience conditions are provided for realizing efficient and fast video indexing.
Description
Technical field
The present invention relates to field of video processing, particularly relate to a kind of index generation method and system.
Background technology
Along with the development of network monitoring business, the monitoring site of laying is more and more, and the massive video data collecting is that carrying out of network monitoring business (for example China Telecom's global eyes business) provides powerful guarantee.But at present the video data of magnanimity is not effectively utilized, user can only be by watching in real time video, carrying out manual retrieval by playback video, further finds object and the event be concerned about watching in process.
Therefore, this traditional browsing mode far can not meet access and the query demand of people to video content, and people more and more wish in video library, to find fast own interested video segment.Meanwhile, the video data collecting is to be only positioned over simply system backstage, video data is not further analyzed, thereby the abundant information containing in video information also cannot be utilized.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of index generation method and system.Scene information by identification in video information, and the inquiry object and the event that are associated with scene information, and then generating indexes file, thus make full use of the information resources in video information, for realizing efficiently, the video frequency searching condition of providing convenience fast.
According to an aspect of the present invention, provide a kind of index generation method, comprising:
Scene information in identification video information;
Obtain the perpetual object being associated with scene information;
The event that identification is associated with described perpetual object;
Select the concern event in described event;
Extract the event description information and the key frame that are associated with described concern event;
Utilize described event description Information generation expandable mark language XML tab file;
Utilize described XML tag file and described key frame generating indexes.
According to a further aspect in the invention, provide a kind of index generation system, comprising:
Scene Recognition unit, for identifying the scene information of video information;
Object acquisition unit, for obtaining the perpetual object being associated with scene information;
Event recognition unit, for identifying the event being associated with described perpetual object;
Selected cell, for selecting the concern event of described event;
Extraction unit, for extracting the event description information and the key frame that are associated with described concern event;
Tab file generation unit, for utilizing described event description Information generation expandable mark language XML tab file;
Index generation unit, for utilizing described XML tag file and described key frame generating indexes.
The present invention is by the scene information in identification video information, obtain the perpetual object being associated with scene information, the event that identification is associated with described perpetual object, select the concern event in described event, extract the event description information and the key frame that are associated with described concern event, utilize described event description Information generation expandable mark language XML tab file, utilize described XML tag file and described key frame generating indexes.Scene information by identification in video information, and the inquiry object and the event that are associated with scene information, and then generating indexes file, thus make full use of the information resources in video information, for realizing efficiently, the video frequency searching condition of providing convenience fast.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of an embodiment of index generation method of the present invention.
Fig. 2 is the schematic diagram of an embodiment of video semanteme characteristic information of the present invention.
Fig. 3 is the schematic diagram of another embodiment of index generation method of the present invention.
Fig. 4 is the schematic diagram of an embodiment of index generation system of the present invention.
Fig. 5 is the schematic diagram of another embodiment of index generation system of the present invention.
Embodiment
With reference to the accompanying drawings the present invention is described more fully, exemplary embodiment of the present invention is wherein described.
Fig. 1 is the schematic diagram of an embodiment of index generation method of the present invention.As shown in Figure 1, the index generation method step of the present embodiment is as follows:
The index generation method providing based on the above embodiment of the present invention, by the scene information in identification video information, obtain the perpetual object being associated with scene information, the event that identification is associated with described perpetual object, select the concern event in described event, extract the event description information and the key frame that are associated with described concern event, utilize described event description Information generation expandable mark language XML tab file, utilize described XML tag file and described key frame generating indexes.Scene information by identification in video information, and the inquiry object and the event that are associated with scene information, and then generating indexes file, thus make full use of the information resources in video information, for realizing efficiently, the video frequency searching condition of providing convenience fast.
Wherein, in step 101, the scene information by identification in video information, can select corresponding object or event targetedly, thereby reaches efficiently, recognition effect fast.
Because those skilled in the art understand the scene information of how identifying in video information, therefore do not launch explanation here.
Preferably, by default scene partitioning table, can determine fast the object that needs concern under corresponding scene.
The schematic diagram that table 1 is scene partitioning table one embodiment of the present invention.As shown in table 1, scene can be divided into and substantially be divided into outdoor and indoor two classes.For outdoor scene and indoor scene, can also further determine the type of scene, and be defined in the object type that should pay close attention under such scene.For example, if scene is company gate, the object that now should pay close attention to is behaved or car.And in the time that scene is bank counter, the object that now should pay close attention to is behaved or thing.
Table 1
For the object that needs in current scene to pay close attention to, identify the event being associated with the object of paying close attention to, and further judge whether the event identifying is concern event.If the event identifying is concern event, extract the event description information and the key frame that are associated with concern event.Utilize described event description Information generation expandable mark language XML tab file, utilize described XML tag file and described key frame generating indexes.Thereby when being convenient to user and retrieving, can utilize key frame to browse, experience thereby improve user.
Here it should be noted that, because those skilled in the art understand how video information is processed to identify perpetual object, dependent event, extract event description information and key frame, therefore do not launch explanation here.
Describe with a concrete example below.In the time identifying current scene and be walkway, according to scene partitioning table determine now should pay close attention to as if people.In this section of video, the object of appearance (being people) is first, second and third.By the event of identifying object first, second and third, discovery first and second are being fought, and third passes by thus.The concern event of presetting owing to fighting, therefore using first and second as perpetual object.And then the event description information that is associated with first and second of extraction, and with the key frame of fighting and being associated.Utilize described event description Information generation expandable mark language XML tab file, utilize described XML tag file and described key frame generating indexes.Thereby user utilizes this index, can easily retrieve the video content of fighting about first and second, and can browse by key frame.
Preferably, event description information comprises number of objects, object's position and the object relationship of the perpetual object being associated with concern event.For example, in upper example, number of objects represents to participate in the number of fighting, and object's position represents to participate in fighting the position of personnel in video, and object relationship represents to participate in the relation of fighting between personnel, for example join, adjacent, cover, comprise etc.
Preferably, event description information also comprises object model, object color and the Obj State of the perpetual object being associated with concern event, object model, object color and the Obj State of for example key object.For example, in upper example, object model represents to participate in fighting the fat or thin of personnel, and object color represents to participate in the personnel's that fight color (for example clothing color), and Obj State represents to participate in fighting the current state of personnel, for example move, static, stand, fall down to the ground etc.
Owing to comprising number of objects, object's position and the object relationship of perpetual object in event description information, and the information such as the object model of perpetual object, object color and Obj State, thereby contribute to user to carry out more specifically and retrieval more efficiently.
Fig. 2 is the schematic diagram of an embodiment of video semanteme characteristic information of the present invention.Wherein video semanteme characteristic information comprises scene information and event description information two parts, and event description information can be expanded as required.By such describing mode, can make it to be more suitable for concrete application.
As a kind of example, utilize the form of expandable mark language XML tab file of described event description Information generation as follows:
Label is wherein:
Event: identified event, class attribute specification event type, as fight, hover etc.;
Frame: key frame, with the position of offset attribute record key frame, take frame as unit;
Relation: the relation in video between object.As: join, adjacent, cover, comprise etc.;
Amount: mark number of objects;
Location: the position of attribute-bit object in video;
Object: the object of mark participation event;
Color: mark object color;
Size: indicated object size, or people's object is fat or thin;
State: the current state of indicated object, as: mobile, static, stand, fall down to the ground etc.
Still, take the above-described event of fighting as example, its XML file is as follows:
Fig. 3 is the schematic diagram of another embodiment of index generation method of the present invention.Wherein in embodiment illustrated in fig. 2, step 301,303-308 are identical with the step 101-107 in embodiment illustrated in fig. 1.Wherein:
Step 301, the scene information in identification video information.
Step 302, carries out pre-service to video, to improve video image quality.
Step 303, obtains the perpetual object being associated with scene information.
Step 304, the event that identification is associated with described perpetual object.
Step 305, selects the concern event in described event.
Step 306, extracts the event description information and the key frame that are associated with described concern event.
Step 307, utilizes described event description Information generation expandable mark language XML tab file.
Step 308, utilizes described XML tag file and described key frame generating indexes.
Wherein by video is carried out to pre-service, the irrelevant information in can removal of images, recovers useful real information, strengthens detectability for information about, reduced data to greatest extent, thus improve the reliability that feature extraction, image are cut apart, mated and identify.Preprocessing process generally has the steps such as digitizing, geometric transformation, normalization, level and smooth, recovery and enhancing.For example can adopt gaussian filtering to eliminate noise, adopt histogram equalization to improve picture quality and make image sharpening etc.
Fig. 4 is the schematic diagram of an embodiment of index generation system of the present invention.In the embodiment shown in fig. 4, index generation system comprises scene Recognition unit 401, object acquisition unit 402, event recognition unit 403, selected cell 404, extraction unit 405, tab file generation unit 506 and index generation unit 407.Wherein:
Tab file generation unit 406, for utilizing described event description Information generation expandable mark language XML tab file.
The index generation system providing based on the above embodiment of the present invention, by the scene information in identification video information, obtain the perpetual object being associated with scene information, the event that identification is associated with described perpetual object, select the concern event in described event, extract the event description information and the key frame that are associated with described concern event, utilize described event description Information generation expandable mark language XML tab file, utilize described XML tag file and described key frame generating indexes.Scene information by identification in video information, and the inquiry object and the event that are associated with scene information, and then generating indexes file, thus make full use of the information resources in video information, for realizing efficiently, the video frequency searching condition of providing convenience fast.
Preferably, event description information comprises number of objects, object's position and the object relationship of the perpetual object being associated with concern event.
Preferably, event description information also comprises object model, object color and the Obj State of the perpetual object being associated with concern event.
Fig. 5 is the schematic diagram of another embodiment of index generation system of the present invention.Compared with embodiment illustrated in fig. 4, in the embodiment shown in fig. 5, system also comprises video pre-filtering unit 501, for after scene recognition unit 401 is identified the scene information of video information, video is carried out to pre-service, to improve video image quality, then denoted object acquiring unit 402 obtains the perpetual object being associated with scene information.
By implementing the present invention, can obtain following beneficial effect:
1. for later retrieval system provides retrieval foundation.
2. can in magnanimity video library, find fast own interested video segment.
3. can get rid of fast video redundancy
4. can fully excavate the information that video data is expressed.
5. reduce the cost of labor that video is searched.
Description of the invention provides for example with for the purpose of describing, and is not exhaustively or limit the invention to disclosed form.Many modifications and variations are obvious for the ordinary skill in the art.Selecting and describing embodiment is for better explanation principle of the present invention and practical application, thereby and makes those of ordinary skill in the art can understand the present invention's design to be suitable for the various embodiment with various modifications of special-purpose.
Claims (8)
1. an index generation method, is characterized in that, comprising:
Scene information in identification video information;
Obtain the perpetual object being associated with scene information;
The event that identification is associated with described perpetual object;
Select the concern event in described event;
Extract the event description information and the key frame that are associated with described concern event;
Utilize described event description Information generation expandable mark language XML tab file;
Utilize described XML tag file and described key frame generating indexes.
2. method according to claim 1, is characterized in that:
After the step of the scene information in identification video information, also comprise:
Video is carried out to pre-service, to improve video image quality;
Then carry out the step of obtaining the perpetual object being associated with scene information.
3. method according to claim 1 and 2, is characterized in that:
Described event description information comprises number of objects, object's position and the object relationship of the perpetual object being associated with concern event.
4. method according to claim 3, is characterized in that:
Described event description information also comprises object model, object color and the Obj State of the perpetual object being associated with concern event.
5. an index generation system, is characterized in that, comprising:
Scene Recognition unit, for identifying the scene information of video information;
Object acquisition unit, for obtaining the perpetual object being associated with scene information;
Event recognition unit, for identifying the event being associated with described perpetual object;
Selected cell, for selecting the concern event of described event;
Extraction unit, for extracting the event description information and the key frame that are associated with described concern event;
Tab file generation unit, for utilizing described event description Information generation expandable mark language XML tab file;
Index generation unit, for utilizing described XML tag file and described key frame generating indexes.
6. system according to claim 5, it is characterized in that: system also comprises video pre-filtering unit, for after the scene information of scene recognition unit identification video information, video is carried out to pre-service, to improve video image quality, then denoted object acquiring unit obtains the perpetual object being associated with scene information.
7. according to the system described in claim 5 or 6, it is characterized in that:
Described event description information comprises number of objects, object's position and the object relationship of the perpetual object being associated with concern event.
8. system according to claim 7, is characterized in that:
Described event description information also comprises object model, object color and the Obj State of the perpetual object being associated with concern event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210446980.9A CN103810195B (en) | 2012-11-09 | 2012-11-09 | index generation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210446980.9A CN103810195B (en) | 2012-11-09 | 2012-11-09 | index generation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103810195A true CN103810195A (en) | 2014-05-21 |
| CN103810195B CN103810195B (en) | 2017-12-12 |
Family
ID=50706978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210446980.9A Active CN103810195B (en) | 2012-11-09 | 2012-11-09 | index generation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103810195B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105450978A (en) * | 2014-06-24 | 2016-03-30 | 杭州海康威视数字技术股份有限公司 | Method and device for achieving structural description in video monitoring system |
| CN105631043A (en) * | 2016-01-26 | 2016-06-01 | 公安部第一研究所 | Video processing method and device |
| CN106488336A (en) * | 2016-11-21 | 2017-03-08 | 株洲中车时代电气股份有限公司 | A kind of engine video frequency transmission method based on event corner and system |
| CN107707931A (en) * | 2016-08-08 | 2018-02-16 | 阿里巴巴集团控股有限公司 | Generated according to video data and explain data, data synthesis method and device, electronic equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101075969A (en) * | 2007-07-20 | 2007-11-21 | 中国建设银行股份有限公司 | Method, apparatus and system for controlling data transmission |
| CN101119482A (en) * | 2007-09-28 | 2008-02-06 | 北京智安邦科技有限公司 | Overall view monitoring method and apparatus |
| CN101226664A (en) * | 2008-02-02 | 2008-07-23 | 北京海鑫科金高科技股份有限公司 | Intelligent monitoring system and method for self-help bank and ATM |
| CN101334845A (en) * | 2007-06-27 | 2008-12-31 | 中国科学院自动化研究所 | Video frequency behaviors recognition method based on track sequence analysis and rule induction |
| CN101449587A (en) * | 2006-06-08 | 2009-06-03 | 汤姆逊许可公司 | Detection of scene switching for video |
| CN101610396A (en) * | 2008-06-16 | 2009-12-23 | 北京智安邦科技有限公司 | Intellective video monitoring device module and system and method for supervising thereof with secret protection |
| JP4387913B2 (en) * | 2004-09-28 | 2009-12-24 | 株式会社エヌ・ティ・ティ・データ | Abnormality detection device, abnormality detection method and abnormality detection program |
| CN101625694A (en) * | 2009-08-17 | 2010-01-13 | 中国科学院地理科学与资源研究所 | Method and system for storing various standard geological metadata |
-
2012
- 2012-11-09 CN CN201210446980.9A patent/CN103810195B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4387913B2 (en) * | 2004-09-28 | 2009-12-24 | 株式会社エヌ・ティ・ティ・データ | Abnormality detection device, abnormality detection method and abnormality detection program |
| CN101449587A (en) * | 2006-06-08 | 2009-06-03 | 汤姆逊许可公司 | Detection of scene switching for video |
| CN101334845A (en) * | 2007-06-27 | 2008-12-31 | 中国科学院自动化研究所 | Video frequency behaviors recognition method based on track sequence analysis and rule induction |
| CN101075969A (en) * | 2007-07-20 | 2007-11-21 | 中国建设银行股份有限公司 | Method, apparatus and system for controlling data transmission |
| CN101119482A (en) * | 2007-09-28 | 2008-02-06 | 北京智安邦科技有限公司 | Overall view monitoring method and apparatus |
| CN101226664A (en) * | 2008-02-02 | 2008-07-23 | 北京海鑫科金高科技股份有限公司 | Intelligent monitoring system and method for self-help bank and ATM |
| CN101610396A (en) * | 2008-06-16 | 2009-12-23 | 北京智安邦科技有限公司 | Intellective video monitoring device module and system and method for supervising thereof with secret protection |
| CN101625694A (en) * | 2009-08-17 | 2010-01-13 | 中国科学院地理科学与资源研究所 | Method and system for storing various standard geological metadata |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105450978A (en) * | 2014-06-24 | 2016-03-30 | 杭州海康威视数字技术股份有限公司 | Method and device for achieving structural description in video monitoring system |
| CN105631043A (en) * | 2016-01-26 | 2016-06-01 | 公安部第一研究所 | Video processing method and device |
| CN107707931A (en) * | 2016-08-08 | 2018-02-16 | 阿里巴巴集团控股有限公司 | Generated according to video data and explain data, data synthesis method and device, electronic equipment |
| CN106488336A (en) * | 2016-11-21 | 2017-03-08 | 株洲中车时代电气股份有限公司 | A kind of engine video frequency transmission method based on event corner and system |
| CN106488336B (en) * | 2016-11-21 | 2019-11-12 | 株洲中车时代电气股份有限公司 | A kind of engine video frequency transmission method and system based on event corner |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103810195B (en) | 2017-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Goëau et al. | Pl@ ntnet mobile app | |
| US9042646B2 (en) | Detecting recurring themes in consumer image collections | |
| Chen et al. | Visual summaries of popular landmarks from community photo collections | |
| CN101692706B (en) | Intelligent storage equipment for security monitoring | |
| Daly et al. | Mining and Classifying Image Posts on Social Media to Analyse Fires. | |
| Kaneko et al. | Event photo mining from twitter using keyword bursts and image clustering | |
| CN102902771A (en) | Method, device and server for searching pictures | |
| CN101021903A (en) | Video caption content analysis system | |
| CN103604271A (en) | Intelligent-refrigerator based food recognition method | |
| CN103050025A (en) | Mobile terminal learning method and learning system thereof | |
| CN105721944A (en) | News information recommendation method for smart television | |
| CN104903892A (en) | Searching system and searching method for object-based images | |
| CN104581437A (en) | Video abstract generation and video backtracking method and system | |
| CN103347167A (en) | Surveillance video content description method based on fragments | |
| CN103069420A (en) | Detecting recurring events in consumer image collections | |
| CN102760169A (en) | Method for detecting advertising slots in television direct transmission streams | |
| CN105100748A (en) | Video monitoring system and video monitoring method | |
| CN104361586A (en) | Electric pole and tower maintenance and pre-warning system based on panoramas | |
| CN103631932A (en) | Method for detecting repeated video | |
| CN101646070A (en) | Novel intelligent security-protecting and monitoring method and special equipment thereof | |
| CN104486585B (en) | A kind of city magnanimity monitor video management method and system based on GIS | |
| CN103810195A (en) | Index generating method and system | |
| CN106021393B (en) | The grid equipment Standard Information Searching method and system of facing mobile apparatus | |
| Rizzo et al. | What Fresh Media Are You Looking For? Retrieving Media Items from Multiple Social Networks | |
| CN103428537A (en) | Video processing method and video processing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |