CN103312770B - Method for auditing resources of cloud platform - Google Patents
Method for auditing resources of cloud platform Download PDFInfo
- Publication number
- CN103312770B CN103312770B CN201310139000.5A CN201310139000A CN103312770B CN 103312770 B CN103312770 B CN 103312770B CN 201310139000 A CN201310139000 A CN 201310139000A CN 103312770 B CN103312770 B CN 103312770B
- Authority
- CN
- China
- Prior art keywords
- video
- pornographic
- cloud platform
- image
- detected
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 230000033001 locomotion Effects 0.000 claims description 18
- 230000008859 change Effects 0.000 claims description 11
- 239000000284 extract Substances 0.000 claims description 10
- 210000000056 organ Anatomy 0.000 claims description 8
- 230000001235 sensitizing effect Effects 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 7
- 238000003780 insertion Methods 0.000 claims description 6
- 230000037431 insertion Effects 0.000 claims description 6
- 230000001850 reproductive effect Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 4
- 239000012634 fragment Substances 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000037081 physical activity Effects 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 13
- 238000005516 engineering process Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000700605 Viruses Species 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 210000000481 breast Anatomy 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 206010022000 influenza Diseases 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000002574 poison Substances 0.000 description 1
- 231100000614 poison Toxicity 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- GOLXNESZZPUPJE-UHFFFAOYSA-N spiromesifen Chemical compound CC1=CC(C)=CC(C)=C1C(C(O1)=O)=C(OC(=O)CC(C)(C)C)C11CCCC1 GOLXNESZZPUPJE-UHFFFAOYSA-N 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 201000009032 substance abuse Diseases 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a method for auditing resources of a cloud platform. The method comprises the following steps: capturing key data contained in the resources to be audited, wherein the key data comprises the content; performing illegal content detection on the content in parallel by utilizing a signature matching algorithm, an identification matching algorithm and a dynamic characteristic analysis method; if the detection is passed completely, allowing the resources to be shared, if not, forbidding the resources to be shared. Multi-aspect security and legality auditing is performed on the resources/content provide by each terminal, and the updating of the service is monitored in the whole life, so that the security and legality of the resources in the cloud platform are ensured, and absolutely original, legal and secure resources and a good cloud platform environment are provided for all users.
Description
Technical field
The present invention relates to field of cloud calculation, more particularly to a kind of method of cloud platform resource examination ＆ verification.
Background technology
At present, there is various illegal or unsafe resource on network, threaten users.For such case,
Network police is mainly taken to safeguard the safety of network, the webpage of close bad at present.But, due to uploading the money on network daily
Source electrode is more, and distribution is extremely wide, can not possibly be audited without unified solution or all.
The content of the invention
In view of this, the technical problem to be solved in the present invention is to provide a kind of method of cloud platform resource examination ＆ verification, by right
The resource submitted in cloud platform carries out security and legitimacy examination ＆ verification, and closes resource safety-of-life by monitoring management system
Method for terminal service, it is ensured that the security and legitimacy of resource in cloud platform.
To reach above-mentioned purpose, the present invention is achieved through the following technical solutions：
A kind of method of cloud platform resource examination ＆ verification, including：
The critical data that pending nuclear resource is included is captured, the critical data includes content；
The content is carried out parallel in illegal using Signature matching algorithms, identification matching algorithm and dynamic analysis method
Hold detection, such as detect and all pass through, then allow the resource-sharing；Otherwise, the resource-sharing is forbidden.
Further, the critical data that the pending nuclear resource of the crawl is included is to adopt DOM Document Object Model（DOM）Side
What method was completed.
Further, pirate video is detected using the Signature matching algorithms, the algorithm includes,
Segmentation is carried out to video in time-domain and obtains video lens segment；
The brief introduction feature of the video lens segment is extracted as signature, and video is classified, the video lens
The brief introduction feature of fragment includes the semantic information and video content prompt of video lens segment；
According to the original video of the type selecting same type of inquiry video, by the signature of the video with aspect indexing storehouse
In original video signature registered in advance matched, such as the match is successful, then be legal video, is pirate video otherwise.
It is further, described classification is carried out to video to include,
The time occurred in video according to familiar things and space characteristics, video different types are divided into；
The familiar things include personage, building, trees, vehicle and street.
Further, the edit of the pirate video includes that geometric transformation, extraneous information insertion, video sequences change
Become, the insertion of many pictures and video quality change.
Further, detect that the algorithm includes to pornographic image using the identification matching algorithm,
The sensitizing range of human body parts in identification image；
Brightness, tone, shape according to the sensitizing range in digital picture, profile, edge, the feature of roughness,
Skin area is detected；
Skin pixels ratio, tone according to skin area in digital picture, the feature at edge are right in skin area
Human body reproductive organ detected,
Set up according to reproductive organs masterplate or built-in rule, if set up successfully, the image is pornographic image,
Otherwise, it is non-pornographic image；
Further, detect that the algorithm includes to pornographic image using the identification matching algorithm,
Extract each area of skin color in image；
Count dermatoglyph direction histogram, roughness, the texture density feature composition characteristic vector of each area of skin color；
Pornographic image and non-pornographic image are distinguished according to the characteristic vector.
Further, pornographic video is detected using the dynamic analysis method, the dynamic analysis method
Including,
Go to recognize pornographic frame using the recognition methods of pornographic image；
Obtain video dynamic characteristic；
Physical activity is detected according to video dynamic characteristic, so as to judge pornographic video and non-pornographic video.
Further, the method that the kinetic characteristic is obtained includes, extracts motion vector characteristic；Or, extract light stream special
Levy；Or, extract sensitive information point；Or, using the motion model of camera.
Technical scheme, by carrying out security in all directions and conjunction to resource/content that each terminal is provided
Method is audited, and monitors the upgrading renewal of the service all the life, it is ensured that the security and legitimacy of resource in cloud platform, is all
User provides absolute legal, legal, safe resource and a good cloud platform environment.
Description of the drawings
The flow chart of the method that Fig. 1 is audited for the cloud platform resource of the embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawings and examples the invention will be further described.
The method flow diagram that Fig. 1 is audited for the cloud platform resource of the embodiment of the present invention.As shown in figure 1, the method include as
Lower step：
Step 101：The critical data that pending nuclear resource is included is captured, the critical data includes content.
In cloud platform, it is pending that content provider uploads resources into after server of cloud platform etc..Server is first
The critical data that the resource is included is analyzed, the key messages such as Data subject, issuing time, content are such as analyzed.Because cloud is passed
Data form in defeated differs widely, and can adopt DOM Document Object Model（DOM）Method, pre-build a variety of documents
Model, these document models are contrasted with the Doctype of the resource for uploading, and come quick using immediate document model
Analyze the critical data of pending nuclear resource.The mode that this resource is processed increased the efficiency of data processing.
Step 102：Using Signature matching algorithms, identification matching algorithm and dynamic analysis method the content is entered parallel
Row illegal contents are detected.
Wherein, the illegal contents detection includes Pornograph detection and pirate content detection.
Pornograph detection technique includes the detection of pornographic image and the detection of pornographic video.Based on the content for grabbing,
Pornographic image generally has two kinds of recognition methods,
First method：The sensitizing range of human body parts in identification pornographic image；Skin region is detected in the sensitizing range
Domain；In detection of skin regions sensitivity organ；Set up according to sensitive organ masterplate or built-in rule, if set up successfully,
Then the image is pornographic image, is non-pornographic image otherwise.
Wherein, according to the sensitizing range of human body parts in digital picture it is brightness, tone, shape, profile, edge, thick
The feature of rugosity, detects in skin area to female breast etc..Skin picture according to sensitizing range in digital picture
Plain ratio, tone, the feature at edge, detect in skin area to human body reproductive organ, according to reproductive organs masterplate or
Built-in rule is set up, if set up successfully, the image is pornographic image, is non-pornographic image otherwise；
The particular content of human body parts is to attempt to understand people in pornographic image by algorithm in the identification pornographic image
What the particular content of body portion was realized.
Second method：Extract each area of skin color in image；The dermatoglyph direction for counting each area of skin color is straight
Fang Tu, roughness, texture density feature composition characteristic vector；Pornographic image and non-pornographic figure are distinguished according to the characteristic vector
Picture..
The main feature of pornographic image is containing a large amount of human body skin areas.Each area of skin color of image is marked,
Both redundant information can have been filtered, resource overhead when feature extraction and machine learning can have been reduced again.Therefore, split in the picture
It is the important prerequisite for carrying out feature extraction to go out area of skin color.Area of skin color in pornographic image accounts for the large percentage of entire image,
And in normal picture area of skin color to account for the ratio of entire image then corresponding less, using the size of area of skin color as a feature
Value；Another of pornographic image is mainly characterized by containing a large amount of smooth, area of skin color without substantially periodicity and directionality, performance
The pixel ratio of i.e. texture-free is more on " grain direction histogram ".By the addition of " grain direction histogram " characteristic value, can
To reduce the erroneous judgement to the image with a large amount of skin pixels but without dermatoglyph.It is " thick in the area of skin color of pornographic image
Rugosity " value shows that color change is slow in colour of skin area than larger, and interior in a big way the consistent of color is kept.Normal picture
Color has typically fluctuated in spatial domain, just changes significantly within the scope of less, so the value of " roughness " is general
It is all smaller., used as a kind of special texture, its " texture density " is than general non-broca scale as much smaller for dermatoglyph.
Statistical analysis technique has an obvious advantage for Texture classification and Texture Segmentation, and with processing speed it is fast the characteristics of, thus make
With " grain direction histogram ", " roughness ", " the texture density " based on statistics as another characteristic value.
The detection of pornographic video can utilize the recognition methods of pornographic image to go to recognize pornographic frame, then using the dynamic of video
Step response detecting the erotic activity of human body, so as to judge pornographic video and non-pornographic video.The video motion characteristic is carried
Take including following several schemes：
Scheme 1：Motion vector characteristic is extracted, the size and Orientation of motion vector is analyzed and is counted.
If neighbor frame difference method is exactly to extract a kind of algorithm of motion vector, the adjacent two field pictures respective pixel phase of general principle
Respective pixel value is subtracted each other, in the case of ambient brightness change less, if respective pixel value difference very little, then it is assumed that be herein
Static, if the pixel value changes in image-region somewhere are very big, it is believed that caused by motion problems in image, by these areas
Field mark out, carries out motion vector characteristic extraction.
Scheme 2：Optical-flow Feature is extracted, the size and Orientation of motion is passed judgment on.
Optical flow method detects that the general principle of moving object is：A speed arrow is given to each pixel in image
Amount, which forms an image motion field, in a particular moment of motion, the point on image and the point one on three-dimensional body
One correspondence, this corresponding relation can be obtained by projection relation, according to the velocity feature of each pixel, image can be entered
Mobile state is analyzed.If without moving object in image, light stream vector is continually varying in whole image region.Work as image
In when having moving object, there is relative motion in target and image background, velocity certainty and neighborhood that moving object is formed
Background velocity vector is different, so as to detect moving object and position.The advantage of optical flow method is that light stream not only carries motion
The movable information of object, but also the abundant information about scenery three-dimensional structure is carried, it can not know appointing for scene
In the case of what information, Moving Objects are detected.
Scheme 3：Sensitive information point is extracted, sensitive information point is tracked and trajectory analysis.
Scheme 4：Using the motion model of camera, the pushing away, draw, shaking of camera, shifting movement type are judged.
The characteristics of pirate content detection technique is by analyzing video piracy, in inquiry video, edits for every class is pirate
Means are detected；According to pirate edit blocking and change degree to video original contents, screen is selectively extracted
The brief introduction feature of frame is used as signature；Then the signature of inquiry video is signed with original video registered in advance in aspect indexing storehouse
Matched, such as the match is successful, then be legal video, is pirate video otherwise.Whole signature search procedure, using cloud computing
Rational calculating task coordinated allocation is carried out, efficient parallel computation is realized, so as to improve effectiveness of retrieval.
In the present embodiment, pirate video is detected using the Signature matching algorithms, examined using video shot boundary
Method of determining and calculating, carries out segmentation and obtains video lens segment in time-domain to video；The brief introduction for extracting the video lens segment is special
Levy as signature, and video is classified, the brief introduction feature of the video lens fragment includes the semanteme of video lens segment
Information and video content prompt；According to the original video of the type selecting same type of inquiry video, by the label of the video
Name is matched with original video signature registered in advance in aspect indexing storehouse, and such as the match is successful, then be legal video, otherwise,
For pirate video.
It is described classification is carried out to video to include, the time occurred in video according to familiar things and space characteristics, regarding
Frequency is divided into different types；The familiar things include personage, building, trees, vehicle and street.
The key technology of video frequency searching during the video shot boundary algorithm.Regarded by shot segmentation or Shot change
Frequency is detected.In the present embodiment, video is split in time-domain, split the video lens segment for obtaining, as source video
Subset, extract the semantic information of video that segmentation is obtained, video is classified such that it is able to the content according to video, according to
The designator of video content, selectively carries out the matching of video, reduces the number of times of video matching, according to personage, building
Time and space characteristics that the familiar things such as thing, trees, vehicle, street occur in video, video is divided into different classes
Type.Then, according to the type of inquiry video, the original video for selecting same type is matched, and greatly improves matching
Efficiency.
The edit of pirate video is mainly 5 big class：Geometric transformation（Such as the change of screen frame displaying ratio）, extra letter
The insertion of breath（Such as captions and TV station's station symbol）, picture-in-picture（As simultaneously small one and large one two videos are played）, under video quality
Drop（Such as noise）With the change of video sequences（Such as fall frame and insertion frame）.When video is inquired about, for every class piracy edit
Detected.
Step 103：Judge whether content detection all passes through, such as all pass through, execution step 105, otherwise execution step
104。
Step 104：Forbid the resource-sharing, reduce the credit grade of resource uploader, and reported and submitted according to violation content
Relevant departments are processed.
Step 105：Allow the resource-sharing；
Step 106：Judge whether the resource has renewal or upgrade, if any execution step 101；Otherwise, to resource other
Aspect is supervised.
By above-mentioned many-side cloud platform resource is audited parallel, it is ensured that the legitimacy and security of content.
Additionally, present disclosure review mechanism can be monitored management, the i.e. resource to servicing to cloud platform
The monitoring for being used, or see that it whether there is the illegal operation for violating safety, and renewal of upgrading to it is also carried out examination ＆ verification,
Ensure the safety-of-life and legitimacy of resource.
Pass（platform as a service）The program development framework provided under pattern is, it is ensured that its security, is somebody's turn to do
, it is ensured that it will not serve the devil, the program to developing should have certain function monitor to the program developed under pattern, under preventing pass
The possibility that utilized by Malware of cloud platform.
In order to solve the problems, such as that installing external network after fire wall can not access internal network server, and for setting up
Buffering area between non-security system and security system, DMZ（demilitarized zone）The web application for using, leads to
Such a DMZ regions are crossed, internal network is more efficiently protected.This network design, compared with general fire wall scheme,
Again many one outposts of the tax office, can filter the attack initiated based on protocol bug for attacker, precisely control user's
Access, reduce the order abuse risk exposed because of useless order opening.
In DMZ schemes, including two fire walls, external firewall keeps out the attack of external network, and manages all outer
Access of portion's network to DMZ.Interior firewall manages DMZ for the access of internal network.Interior firewall is internal network
3rd road security perimeter (above including external firewall and Bastion Host), when external firewall fails, it can be with
Play the function of protection internal network.And inside LAN, the access for Internet is by interior firewall and positioned at DMZ's
Bastion Host is controlled.In such structure, a hacker must pass through three independent regions, and (external firewall, inside are anti-
Wall with flues and Bastion Host) LAN can be reached.Attack difficulty to greatly reinforce, the security of respective inner network is also just significantly
Strengthen.Resource uploader uploads resource, then it is audited by cloud platform content auditing system, if the resource has
Poison, due to the design of cloud platform DMZ security architecture, the virus can not be threatened the final resource composition in high in the clouds, and can be by
The resource is destroyed immediately.
Technical scheme, by carrying out security in all directions and conjunction to resource/content that each terminal is provided
Method is audited, and monitors the upgrading renewal of the service all the life, it is ensured that the security and legitimacy of resource in cloud platform, is all
User provides absolute legal, legal, safe resource and a good cloud platform environment.
All or part of content in the technical scheme that above example is provided can instruct the hard of correlation by program
Completing, described program can be stored in a computer read/write memory medium part, and the program is accordingly introduced
The step of perform, described storage medium, such as：ROM/RAM, magnetic disc, CD etc..
Above are only presently preferred embodiments of the present invention and institute's application technology principle, any technology people for being familiar with the art
In the technical scope of present disclosure, the change or replacement that can be readily occurred in all should be covered in protection scope of the present invention member
It is interior.
Claims (8)
1. a kind of method that cloud platform resource is audited, it is characterised in that include：
The critical data that pending nuclear resource is included is captured, the critical data includes content；
Pirate video is detected using Signature matching algorithms, pornographic image is detected using identification matching algorithm, profit
Pornographic video is detected with dynamic analysis method, wherein, pirate video is examined using the Signature matching algorithms
Survey, the algorithm includes,
Segmentation is carried out to video in time-domain and obtains video lens segment；
The brief introduction feature of the video lens segment is extracted as signature, and video is classified, the video lens fragment
Brief introduction feature including video lens segment semantic information and video content prompt；
According to the original video of the type selecting same type of inquiry video, by the signature of the video with pre- in aspect indexing storehouse
The original video signature first registered is matched, and such as the match is successful, then be legal video, is pirate video otherwise；
Such as detect and all pass through, then allow the resource-sharing；Otherwise, the resource-sharing is forbidden.
2. the method for cloud platform resource according to claim 1 examination ＆ verification, it is characterised in that the pending nuclear resource bag of the crawl
The critical data for containing is completed using the method for DOM Document Object Model (DOM).
3. the method that cloud platform resource according to claim 1 is audited, it is characterised in that described that classification bag is carried out to video
Include,
The time occurred in video according to familiar things and space characteristics, video different types are divided into；
The familiar things include personage, building, trees, vehicle and street.
4. the method that cloud platform resource according to claim 1 is audited, it is characterised in that editor's hand of the pirate video
Section includes geometric transformation, extraneous information insertion, video sequences change, the insertion of many pictures and video quality change.
5. the method that cloud platform resource according to claim 1 is audited, it is characterised in that using the identification matching algorithm
Pornographic image is detected, the algorithm includes,
The sensitizing range of human body parts in identification image；
Brightness, tone, shape according to the sensitizing range in digital picture, profile, edge, the feature of roughness, to skin
Detected in skin region；
Skin pixels ratio, tone according to skin area in digital picture, the feature at edge, to human body in skin area
Reproductive organs detected,
Set up according to reproductive organs masterplate or built-in rule, if set up successfully, the image is pornographic image, no
Then, it is non-pornographic image.
6. the method that cloud platform resource according to claim 1 is audited, it is characterised in that using the identification matching algorithm
Pornographic image is detected, the algorithm includes,
Extract each area of skin color in image；
Count dermatoglyph direction histogram, roughness, the texture density feature composition characteristic vector of each area of skin color；
Pornographic image and non-pornographic image are distinguished according to the characteristic vector.
7. the method that cloud platform resource according to claim 1 is audited, it is characterised in that using the dynamic analysis
Method detects that the dynamic analysis method includes to pornographic video,
Go to recognize pornographic frame using the recognition methods of pornographic image；
Obtain video dynamic characteristic；
Physical activity is detected according to video dynamic characteristic, so as to judge pornographic video and non-pornographic video.
8. the method that cloud platform resource according to claim 7 is audited, it is characterised in that the acquisition side of the dynamic characteristic
Method includes：Extract motion vector characteristic；Or, extract Optical-flow Feature；Or, extract sensitive information point；Or, using camera
Motion model, judge the pushing away, draw, shaking of camera, shifting movement type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310139000.5A CN103312770B (en) | 2013-04-19 | 2013-04-19 | Method for auditing resources of cloud platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310139000.5A CN103312770B (en) | 2013-04-19 | 2013-04-19 | Method for auditing resources of cloud platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103312770A CN103312770A (en) | 2013-09-18 |
| CN103312770B true CN103312770B (en) | 2017-05-03 |
Family
ID=49137547
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310139000.5A Expired - Fee Related CN103312770B (en) | 2013-04-19 | 2013-04-19 | Method for auditing resources of cloud platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103312770B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103686244A (en) * | 2013-12-26 | 2014-03-26 | 乐视网信息技术（北京）股份有限公司 | Video data managing method and system |
| US9668020B2 (en) * | 2014-04-07 | 2017-05-30 | The Nielsen Company (Us), Llc | Signature retrieval and matching for media monitoring |
| CN104158844A (en) * | 2014-07-14 | 2014-11-19 | 上海东方延华节能技术服务股份有限公司 | Remote real-time monitoring system |
| WO2017107209A1 (en) * | 2015-12-25 | 2017-06-29 | 王晓光 | Method and system for image recognition in video software |
| CN105868747B (en) * | 2016-06-27 | 2018-04-13 | 厦门久凌创新科技有限公司 | Cloud computing system |
| US9872056B1 (en) * | 2016-12-16 | 2018-01-16 | Google Inc. | Methods, systems, and media for detecting abusive stereoscopic videos by generating fingerprints for multiple portions of a video frame |
| CN108171222B (en) * | 2018-02-11 | 2020-08-25 | 清华大学 | Real-time video classification method and device based on multi-stream neural network |
| WO2020010568A1 (en) * | 2018-07-12 | 2020-01-16 | 深圳齐心集团股份有限公司 | Big data artificial intelligence analysis device |
| CN109766508B (en) * | 2018-12-28 | 2021-09-21 | 广州华多网络科技有限公司 | Information auditing method and device and electronic equipment |
| CN110572693A (en) * | 2019-08-23 | 2019-12-13 | 贵州省广播电视信息网络股份有限公司 | Media asset transcoding method based on artificial intelligence |
| CN111967493A (en) * | 2020-06-29 | 2020-11-20 | 北京百度网讯科技有限公司 | Image auditing method and device, electronic equipment and storage medium |
| CN112333189B (en) * | 2020-11-06 | 2021-07-30 | 北京木铭春天文化传媒有限公司 | Data processing method based on Internet of things communication and live broadcast platform and cloud computing center |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102143120A (en) * | 2010-02-01 | 2011-08-03 | 中兴通讯股份有限公司 | Method and device for realizing grading multi-media files |
| CN102292726A (en) * | 2009-01-23 | 2011-12-21 | 日本电气株式会社 | video identifier extracting device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1323370C (en) * | 2004-05-28 | 2007-06-27 | 中国科学院计算技术研究所 | Method for detecting pornographic images |
| CN100363943C (en) * | 2004-06-21 | 2008-01-23 | 南开大学 | Color image matching analytical method based on color content and distribution |
| CN101296148B (en) * | 2008-06-26 | 2011-01-05 | 蓝汛网络科技（北京）有限公司 | Verification method, system and device for validity of multimedia contents |
| CN101299690A (en) * | 2008-06-26 | 2008-11-05 | 蓝汛网络科技（北京）有限公司 | Method, system and apparatus for monitoring multimedia contents |
| CN102073841B (en) * | 2009-11-20 | 2012-08-01 | 中国移动通信集团广东有限公司 | Poor video detection method and device |
| CN102184419B (en) * | 2011-04-13 | 2013-05-01 | 深圳市迈科龙影像技术有限公司 | Pornographic image recognizing method based on sensitive parts detection |
| CN102236796B (en) * | 2011-07-13 | 2014-01-22 | Tcl集团股份有限公司 | Method and system for sorting defective contents of digital video |
| CN102970164B (en) * | 2012-11-20 | 2016-05-11 | 无锡成电科大科技发展有限公司 | A kind of cloud platform management monitoring system and method |
-
2013
- 2013-04-19 CN CN201310139000.5A patent/CN103312770B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102292726A (en) * | 2009-01-23 | 2011-12-21 | 日本电气株式会社 | video identifier extracting device |
| CN102143120A (en) * | 2010-02-01 | 2011-08-03 | 中兴通讯股份有限公司 | Method and device for realizing grading multi-media files |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103312770A (en) | 2013-09-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103312770B (en) | Method for auditing resources of cloud platform | |
| Verdoliva | Media forensics and deepfakes: an overview | |
| US8619135B2 (en) | Detection of abnormal behaviour in video objects | |
| US20170289624A1 (en) | Multimodal and real-time method for filtering sensitive media | |
| Zabłocki et al. | Intelligent video surveillance systems for public spaces–a survey | |
| US20060239506A1 (en) | Line textured target detection and tracking with applications to "Basket-run" detection | |
| CN113255518B (en) | Video abnormal event detection method and chip | |
| CN111866600B (en) | Method, device and system for protecting privacy | |
| Kim et al. | Defending Physical Adversarial Attack on Object Detection via Adversarial Patch-Feature Energy | |
| KR20160116585A (en) | Method and apparatus for blocking harmful area of moving poctures | |
| Frejlichowski et al. | SmartMonitor: An approach to simple, intelligent and affordable visual surveillance system | |
| Patil et al. | AlexNet based pirate detection system | |
| Upadhyay et al. | Learning based video authentication using statistical local information | |
| Sitara et al. | Automated camera sabotage detection for enhancing video surveillance systems | |
| CN115546906A (en) | System and method for detecting human face activity in image and electronic equipment | |
| CN113596354B (en) | Image processing method, image processing device, computer equipment and storage medium | |
| CN113095272B (en) | Living body detection method, living body detection device, living body detection medium and computer program product | |
| Kaur | Background subtraction in video surveillance | |
| Tang et al. | C-14: assured timestamps for drone videos | |
| Xie et al. | PROVES: Establishing Image Provenance using Semantic Signatures | |
| Gupta et al. | Unusual Activity Detection Using Machine Learning | |
| Fischer et al. | Privacy-aware smart video surveillance revisited | |
| Mancas | Attention-based dense crowds analysis | |
| Bheechook et al. | The need for automatic detection of uncommon behaviour in surveillance systems: A short review | |
| Kandavalli et al. | Design and implementation of colour texture‐based multiple object detection using morphological gradient approach |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210113 Address after: No.2006 Xiyuan Avenue, Chengdu, Sichuan 611731 Patentee after: University of Electronic Science and technology of Sichuan foundation for education development Address before: Room 402, area a, Liye building, science and Technology Park, China sensor network university, Taike Park, Wuxi New District, Wuxi City, Jiangsu Province, 214135 Patentee before: WUXI UESTC TECHNOLOGY DEVELOPMENT Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170503 |