CN109218721A - A kind of mutation video detecting method compared based on frame - Google Patents
A kind of mutation video detecting method compared based on frame Download PDFInfo
- Publication number
- CN109218721A CN109218721A CN201811416278.1A CN201811416278A CN109218721A CN 109218721 A CN109218721 A CN 109218721A CN 201811416278 A CN201811416278 A CN 201811416278A CN 109218721 A CN109218721 A CN 109218721A
- Authority
- CN
- China
- Prior art keywords
- video
- frame
- sample
- characteristic value
- mutation
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
Abstract
The invention discloses a kind of mutation video detecting methods compared based on frame, for the illicit video grasped, it samples the every frame data of original sample video and generates every frame characteristic value and deposit in the database, the sample database is all based on for the detection of target mobile phones below, sample database needs constantly study, updates；Target mobile phones multi-medium data is acquired to locally, cellphone multimedia data is traversed, to target mobile phones video acquisition frame data and generates characteristic value and sample storehouse matching, finally export hit results.The present invention is directed to illegal mutation video file, provides useful supplement for MD5 detection mode, being capable of effective monitoring, strike and the propagation wantonly for blocking illegal file.
Description
Technical field
The invention discloses a kind of mutation video detecting methods compared based on frame, are related to mobile terminal safety technology neck
Domain.
Background technique
In recent years, with the fast development of intelligent mobile terminal, letter of the multimedias such as voice, image and video due to carrying
Breath amount is bigger, it is more convenient to link up, and using more and more extensive in social networks and instant messaging tools, law-breaker propagates and has
The image of sensitive text；Or sensitive information is transmitted by image file using Information Hiding Techniques, to escape monitoring and strike.
Current general detection mode be by scanning cell phone multimedia file, will be in the MD5 value and video library of video file
The comparison of face MD5 value, so that it is determined that whether the mobile phone has illicit video, which is quickly, accurately.But illicit video is being propagated through
In journey or when artificial change video attribute, such as change video resolution, video truncation, video-splicing, the MD5 value hair of video
Changing, generic way are just examined and are not measured the video, thus can not effective monitoring, strike and the biography wantonly for blocking illegal file
It broadcasts.
Summary of the invention
The technical problems to be solved by the present invention are: in view of the drawbacks of the prior art, providing a kind of change compared based on frame
Kind video detecting method, this method can effectively detect the corresponding mutation of discovery on the basis of the illicit video grasped
Video provides useful supplement for MD5 detection mode, being capable of effective monitoring, strike and the propagation wantonly for blocking illegal file.
The present invention uses following technical scheme to solve above-mentioned technical problem:
A kind of mutation video detecting method compared based on frame, the specific steps of the method include:
Step 1: sampling to obtain the feature of raw sample data by sampling function to each frame of original sample video
The characteristic value of the raw sample data is stored sample introduction database by value；
The sample mode are as follows: every frame picture is reduced into 8x8 size, then gray processing, binaryzation (big law OTSU), most
After calculate hash value, obtain the characteristic value of every frame picture；
Step 2: obtaining all multimedia files in target mobile phones, the multimedia file in target mobile phones is traversed；
Step 3: being adopted to the video file in multimedia file by each frame of the sampling function to target video
Sample, by the spy of original sample video stored in the characteristic value and sample database that sample obtained each frame sampling data
Value indicative is matched；
Step 4: when the characteristic value for being matched to identical raw sample data and sampled data in sample database
When, the frame is determined for frame hit, and stepping detection is successively carried out to several frames for setting quantity after hit frame, stepping is detected
To characteristic value successively with hit frame in database after the characteristic values of corresponding several frames be compared, when passing through it was found that
When characteristic value is all consistent, determines target video and original sample video is same video clip, i.e., the target video is sample
The mutation illicit video file of a certain video in video library.
As a further preferred embodiment of the present invention, the sampling function is using the big law algorithm of OTSU to original sample number
According to progress binary conversion treatment.
As a further preferred embodiment of the present invention, in the step 4: the characteristic value that stepping detects successively with number
It is compared according to the characteristic value of corresponding several frames after hit frame in library, comparison result is all consistent and refers to, contrast function is returned
The result returned is true.Wherein, whether it is genuine judgment method that the contrast function returns the result are as follows: two frame video characteristic values Hammings
Distance is less than the threshold value of setting.
It as a further preferred embodiment of the present invention, in the method, is adjusted by parameter, control sampling and compares speed
Degree, rate of false alarm and rate of failing to report.
As a further preferred embodiment of the present invention, in the method, sample database is by way of artificial or AI
Carry out machine learning or update.
The invention adopts the above technical scheme compared with prior art, has following technical effect that the present invention is based on current
The MD5 mode disadvantage existing in concrete application scene provided for illegal mutation video file for MD5 detection mode
Useful supplement, being capable of effective monitoring, strike and the propagation wantonly for blocking illegal file.
Detailed description of the invention
Fig. 1 is in the present invention, and mutation video frame compares schematic diagram.
Specific embodiment
Embodiments of the present invention are described below in detail, the example of the embodiment is shown in the accompanying drawings, wherein from beginning
Same or similar element or element with the same or similar functions are indicated to same or similar label eventually.Below by ginseng
The embodiment for examining attached drawing description is exemplary, and for explaining only the invention, and is not construed as limiting the claims.
Technical solution of the present invention is described in further detail with reference to the accompanying drawing:
Technical solution disclosed in this invention mainly includes sample database being generated to video sampling and to target mobile phones multimedia
File detects two parts, and specific step is as follows for the method:
1, each frame of original sample video is sampled to obtain characteristic value by certain sampling function, characteristic value is stored in
In sample database.
2, all multimedia files of target mobile phones are obtained to locally, traverse target mobile phones multimedia file.
3, for video file, also each frame of target video is sampled in the same way, the spy that each frame obtains
It is matched in value indicative and sample database.
If 4, a certain frame hit (finding identical characteristic value in database), to several after the frame
Frame successively carries out stepping detection, obtained characteristic value successively in database hit frame after corresponding several frames feature
Value is compared, if be all consistent (do not imply that completely the same, contrast function, which returns, true to be thought to be consistent) be then considered as it is same
Video clip, the i.e. target video are the mutation illicit video file of a certain video in Sample video library.
5, it for sampling and comparing speed, rate of false alarm and rate of failing to report, can be adjusted by parameter to control.
As shown in Figure 1, present invention design is based on video frame comparison technology, mutation video is detected, specific implementation step
It is rapid as follows:
1) for the illicit video grasped, the every frame data of sample video simultaneously generate every frame characteristic value there are databases
In, behind the sample database is all based on for the detection of target mobile phones, sample database needs constantly study, updates.
2) acquisition target mobile phones multi-medium data traverses cellphone multimedia data, to target mobile phones video acquisition to locally
Frame data simultaneously generate characteristic value and sample storehouse matching, finally export hit results.
Embodiments of the present invention are explained in detail above in conjunction with attached drawing, but the present invention is not limited to above-mentioned implementations
Mode within the knowledge of a person skilled in the art can also be without departing from the purpose of the present invention
It makes a variety of changes.The above described is only a preferred embodiment of the present invention, not making limit in any form to the present invention
System, although the present invention has been disclosed as a preferred embodiment, however, it is not intended to limit the invention, any skill for being familiar with this profession
Art personnel, without departing from the scope of the present invention, be modified when the technology contents using the disclosure above or
It is modified to the equivalent embodiment of equivalent variations, but without departing from the technical solutions of the present invention, technology according to the present invention is real
Matter, within the spirit and principles in the present invention, any simple modifications, equivalent substitutions and improvements to the above embodiments
Deng, fall within the scope of protection of the technical scheme of the present invention within.
Claims (6)
1. a kind of mutation video detecting method compared based on frame, which is characterized in that the specific steps of the method include:
Step 1: each frame of original sample video is sampled to obtain the characteristic value of raw sample data by sampling function, it will
The characteristic value of the raw sample data stores sample introduction database；
The sample mode are as follows: every frame picture is reduced into 8x8 size, and then gray processing, binaryzation (big law OTSU), are finally counted
Hash value is calculated, the characteristic value of every frame picture is obtained；
Step 2: obtaining all multimedia files in target mobile phones, the multimedia file in target mobile phones is traversed；
Step 3: the video file in multimedia file is sampled by each frame of the sampling function to target video, it will
Sample the characteristic value of original sample video stored in the characteristic value and sample database of obtained each frame sampling data
It is matched；
Step 4: sentencing when being matched to the characteristic value of identical raw sample data and sampled data in sample database
The fixed frame is frame hit, successively carries out stepping detection to several frames for setting quantity after hit frame, stepping is detected
Characteristic value is successively compared with the characteristic value of corresponding several frames after hit frame in database, when process is it was found that feature
When value is all consistent, determines target video and original sample video is same video clip, i.e., the target video is Sample video
The mutation illicit video file of a certain video in library.
2. a kind of mutation video detecting method compared based on frame as described in claim 1, it is characterised in that: the sampling letter
Number carries out binary conversion treatment to raw sample data using the big law algorithm of OTSU.
3. a kind of mutation video detecting method compared based on frame as described in claim 1, which is characterized in that the step 4
In: the characteristic value that stepping detects successively is compared with the characteristic value of corresponding several frames after hit frame in database,
Comparison result is all consistent and refers to, the result that contrast function returns is true.
4. a kind of mutation video detecting method compared based on frame as claimed in claim 3, which is characterized in that the comparison letter
Whether it is genuine judgment method that number returns the result are as follows: two frame video characteristic values Hamming distances are less than the threshold value of setting.
5. a kind of mutation video detecting method compared based on frame as claimed in claim 1 or 3, it is characterised in that: the side
It in method, is adjusted by parameter, control sampling and comparison speed, rate of false alarm and rate of failing to report.
6. a kind of mutation video detecting method compared based on frame as claimed in claim 1 or 3, it is characterised in that: the side
In method, sample database carries out machine learning or update by way of artificial or AI.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811416278.1A CN109218721A (en) | 2018-11-26 | 2018-11-26 | A kind of mutation video detecting method compared based on frame |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811416278.1A CN109218721A (en) | 2018-11-26 | 2018-11-26 | A kind of mutation video detecting method compared based on frame |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109218721A true CN109218721A (en) | 2019-01-15 |
Family
ID=64993798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811416278.1A Pending CN109218721A (en) | 2018-11-26 | 2018-11-26 | A kind of mutation video detecting method compared based on frame |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109218721A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110769276A (en) * | 2019-11-07 | 2020-02-07 | 成都国腾实业集团有限公司 | Frame extraction slice video detection method based on MD5 |
| CN113971229A (en) * | 2021-10-20 | 2022-01-25 | 成都智元汇信息技术股份有限公司 | Frame comparison method analysis method and device |
| US11431638B2 (en) * | 2019-10-14 | 2022-08-30 | Sandvine Corporation | System and method for monitoring and managing video stream content |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101540907A (en) * | 2009-04-28 | 2009-09-23 | 无锡天脉聚源传媒科技有限公司 | Method for auditing safety of internet video |
| CN102222227A (en) * | 2011-04-25 | 2011-10-19 | 中国华录集团有限公司 | Video identification based system for extracting film images |
| CN103596016A (en) * | 2013-11-20 | 2014-02-19 | 韩巍 | Multimedia video data processing method and device |
| CN105005772A (en) * | 2015-07-20 | 2015-10-28 | 北京大学 | Video scene detection method |
| CN107318041A (en) * | 2017-06-29 | 2017-11-03 | 深圳市茁壮网络股份有限公司 | The method and system that a kind of Video security is played |
-
2018
- 2018-11-26 CN CN201811416278.1A patent/CN109218721A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101540907A (en) * | 2009-04-28 | 2009-09-23 | 无锡天脉聚源传媒科技有限公司 | Method for auditing safety of internet video |
| CN102222227A (en) * | 2011-04-25 | 2011-10-19 | 中国华录集团有限公司 | Video identification based system for extracting film images |
| CN103596016A (en) * | 2013-11-20 | 2014-02-19 | 韩巍 | Multimedia video data processing method and device |
| CN105005772A (en) * | 2015-07-20 | 2015-10-28 | 北京大学 | Video scene detection method |
| CN107318041A (en) * | 2017-06-29 | 2017-11-03 | 深圳市茁壮网络股份有限公司 | The method and system that a kind of Video security is played |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11431638B2 (en) * | 2019-10-14 | 2022-08-30 | Sandvine Corporation | System and method for monitoring and managing video stream content |
| US20220393982A1 (en) * | 2019-10-14 | 2022-12-08 | Sandvine Corporation | System and method for monitoring and managing video stream content |
| US11743195B2 (en) * | 2019-10-14 | 2023-08-29 | Sandvine Corporation | System and method for monitoring and managing video stream content |
| US20230370384A1 (en) * | 2019-10-14 | 2023-11-16 | Sandvine Corporation | System and method for monitoring and managing video stream content |
| CN110769276A (en) * | 2019-11-07 | 2020-02-07 | 成都国腾实业集团有限公司 | Frame extraction slice video detection method based on MD5 |
| CN113971229A (en) * | 2021-10-20 | 2022-01-25 | 成都智元汇信息技术股份有限公司 | Frame comparison method analysis method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109218721A (en) | A kind of mutation video detecting method compared based on frame | |
| CN104091388B (en) | A kind of paper money discrimination method based on magnetic image and device | |
| CN105095856B (en) | Face identification method is blocked based on mask | |
| CN103761748B (en) | Anomaly detection method and device | |
| CN103902476B (en) | A kind of webpage back door detection method and system based on non-credit | |
| CN105635085B (en) | Safe big data analysis system and method based on dynamical health degree model | |
| CN102571487B (en) | Distributed bot network scale measuring and tracking method based on multiple data sources | |
| CN105335722A (en) | Detection system and detection method based on depth image information | |
| Qiu et al. | Multi-view convolutional neural network for data spoofing cyber-attack detection in distribution synchrophasors | |
| CN104636764B (en) | A kind of image latent writing analysis method and its device | |
| CN108197575A (en) | A kind of abnormal behaviour recognition methods detected based on target detection and bone point and device | |
| CN109063625A (en) | A kind of face critical point detection method based on cascade deep network | |
| CN109977895A (en) | A kind of wild animal video object detection method based on multi-characteristic fusion | |
| CN109450957A (en) | A kind of low speed Denial of Service attack detection method based on cloud model | |
| CN111191535B (en) | Pedestrian detection model construction method based on deep learning and pedestrian detection method | |
| CN106611151A (en) | A human face identification method and apparatus | |
| CN109389105A (en) | A kind of iris detection and viewpoint classification method based on multitask | |
| CN110490150A (en) | A kind of automatic auditing system of picture violating the regulations and method based on vehicle retrieval | |
| CN108921826A (en) | The transmission line of electricity that super-pixel segmentation is combined with deep learning invades object detecting method | |
| CN108876810A (en) | The method that algorithm carries out moving object detection is cut using figure in video frequency abstract | |
| CN109660814A (en) | A kind of detection method that video foreground deletion is distorted | |
| CN103366163B (en) | Face detection system and method based on incremental learning | |
| CN102254396A (en) | Intrusion detection method and device based on video | |
| CN105427303B (en) | A kind of vision measurement and method of estimation of substation's legacy | |
| CN104980211B (en) | A kind of signal processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190115 |