CN108805069A - Image detection method and device - Google Patents
Image detection method and device Download PDFInfo
- Publication number
- CN108805069A CN108805069A CN201810561124.5A CN201810561124A CN108805069A CN 108805069 A CN108805069 A CN 108805069A CN 201810561124 A CN201810561124 A CN 201810561124A CN 108805069 A CN108805069 A CN 108805069A
- Authority
- CN
- China
- Prior art keywords
- image
- measured
- detection method
- target database
- database
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Processing Or Creating Images (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of image detection method and device, the image detection device includes several databases, and the image detection method includes：Obtain the image to be measured for user's head portrait that user uploads；Identify the characteristic point in the image to be measured；The characteristic point is compared in target database, the target database is one in several databases；Obtain the similar value of the image to be measured and image in target database；When the similar value of the image to be measured is more than the first preset value, predetermined registration operation is executed.The image detection method and device of the present invention quickly can timely judge whether the head portrait that user uploads is legal, has saved human cost, has improved efficiency and strengthen internet security, further, the application can quickly identify and recognition accuracy higher.
Description
Technical field
The present invention relates to a kind of image detection method and devices.
Background technology
Image recognition technology is a key areas of artificial intelligence.In order to work out the simulation movable meter of mankind's image recognition
Calculation machine program, there has been proposed different image recognition models.Such as Template matching model.This model is thought, identifies some
Image, it is necessary to which the memory pattern for having this image in past experience is called template.If current stimulation can in brain
Template match, this image is also just identified.
Existing website is mostly used image recognition technology in user's uploading pictures, so as in automatic identification image
It relates to Huang, fear information cruelly, prevent the propagation of flame, play the role of purifying network.
But in the prior art, the application field of picture recognition technology is less, and recognition accuracy and speed are relatively low.
Invention content
The technical problem to be solved by the present invention is in order to overcome the application field of picture recognition technology in the prior art less,
Recognition accuracy and the lower defect of speed, providing one kind can quickly identify and the higher image detection method of recognition accuracy
And device.
Technical solution is stated to solve above-mentioned technical problem：
A kind of image detection method, is used for image detection device, and feature is that the image detection device includes several
Database, the image detection method include：
Obtain the image to be measured for user's head portrait that user uploads；
Identify the characteristic point in the image to be measured；
The characteristic point is compared in target database, the target database is in several databases
One；
Obtain the similar value of the image to be measured and image in target database；
When the similar value of the image to be measured is more than the first preset value, predetermined registration operation is executed.
Preferably, several databases are including relating to yellow database, political sensitivity database and fearing database cruelly.
Preferably, the characteristic point in the identification image to be measured includes：
Identify the characteristic point of skin in the image to be measured；
Judge that skin area in image to be measured accounts for the ratio of imagery coverage to be measured and whether is more than the according to the characteristic point of skin
Two preset values, if it is target database then to relate to yellow database described in setting.
Preferably, the characteristic point in the identification image to be measured includes：
Judge that skin area in image to be measured accounts for the ratio of imagery coverage to be measured and whether is more than the according to the characteristic point of skin
Two preset values, if otherwise judging, skin area in image to be measured accounts for the ratio of imagery coverage to be measured and whether is less than third preset value,
The political sensitivity database is set if being less than third preset value and probably database is target database cruelly.
Preferably, the image detection method includes：
Obtain the image for the user's head portrait reported；
It will be compared in newest target database by the characteristic point of report image；
Whether by report image legal, artificial detection is by report image and by the illegal quilt of handmarking if legal if judging
Report image is added in target database, if not rule will be added to by report image in target database.
Preferably, the image detection method includes：
Pass through the text information in Feature point recognition image to be measured；
When the similar value of the image to be measured is more than the first preset value, predetermined registration operation is executed, the predetermined registration operation includes
The contact method that the text information is recorded is reported according to the text information.
Preferably, the image to be measured for obtaining user's head portrait that user uploads includes：
Judge whether the image to be measured is dynamic image, if then press predetermined interval duration intercept the dynamic image with
Obtain sub-image to be measured；
Identify the characteristic point in sub-image to be measured；
The characteristic point is compared in target database, the target database is in several databases
One；
Obtain the similar value of sub-image to be measured and image in target database；
When the similar value of sub-image to be measured is more than the first preset value, predetermined registration operation is executed.
It is illegal and the image to be measured is added to mesh preferably, the predetermined registration operation includes the image mark to be measured
It marks in database.
Preferably, the image detection method includes：
Obtain the association platform of user；
The head portrait for obtaining user in the association platform is image to be measured；
Using the corresponding database of the association platform as target database.
The present invention also provides a kind of image detection device, feature is, the image detection device is for realizing as above
The image detection method.
On the basis of common knowledge of the art, above-mentioned each optimum condition can be combined arbitrarily to get each preferable reality of the present invention
Example.
The positive effect of the present invention is that：The image detection method and device of the present invention quickly can timely judge
Whether the head portrait that user uploads is legal, has saved human cost, has improved efficiency and strengthen internet security, further,
The application can quickly identify and recognition accuracy higher.
Description of the drawings
Fig. 1 is the flow chart of the image detection method of the embodiment of the present invention 1.
Fig. 2 is the flow chart of the image detection method of the embodiment of the present invention 2.
Fig. 3 is the flow chart of the image detection method of the embodiment of the present invention 3.
Fig. 4 is the flow chart of the image detection method of the embodiment of the present invention 5.
Fig. 5 is the flow chart of the image detection method of the embodiment of the present invention 6.
Specific implementation mode
It is further illustrated the present invention below by the mode of embodiment, but does not therefore limit the present invention to the reality
It applies among a range.
Embodiment 1
The present embodiment provides a kind of image detection device, the image detection device includes 3 databases, respectively relates to Huang
Database, political sensitivity database and cruelly fear database.
Wherein, the data for relating to yellow database purchase relate to yellow picture for identification, and political sensitivity database includes political leadership
Human head picture or word, for identification political sensitivity picture, the data of sudden and violent probably database purchase record terrorism, sudden and violent for identification
The picture of power event.
Referring to Fig. 1, using the image detection device, the present embodiment provides a kind of image detection methods, including：
Step 100, the image to be measured for obtaining user's head portrait that user uploads.
The image that the image to be measured is used to use as user's head portrait after registration.
Characteristic point in step 101, the identification image to be measured；
The characteristic point is compared step 102 in target database, and the target database is several numbers
According to one in library；
Step 103, the similar value for obtaining the image to be measured and image in target database；
Step 104 judges whether the similar value of the image to be measured is more than the first preset value, if so then execute step 105,
If otherwise terminating flow.
Step 105 executes predetermined registration operation.
The predetermined registration operation includes notifying user that picture is illegal.
The predetermined registration operation can also include being illegal by the image mark to be measured and being added to the image to be measured
In target database.So as to advanced optimize the recognition capability of database.
Image recognition is carried out using multiple database, recognition accuracy can not only be improved, and recognition speed can be improved.
Embodiment 2
Referring to Fig. 2, the present embodiment is substantially the same manner as Example 1, the difference is that only：
Step 200, the image to be measured for obtaining user's head portrait that user uploads.
The characteristic point of skin in step 201, the identification image to be measured.
Step 202 judges that the ratio that skin area accounts for imagery coverage to be measured in image to be measured is according to the characteristic point of skin
It is no to be more than the second preset value, if so then execute step 203,204 are thened follow the steps if not.
It is target database that yellow database is related to described in step 203, setting, then executes step 205.
Step 204 judges that skin area in image to be measured accounts for the ratio of imagery coverage to be measured and whether is less than third preset value,
If the political sensitivity database is then arranged and probably database is target database cruelly.
The characteristic point is compared step 205 in target database.
Step 206, the similar value for obtaining the image to be measured and image in target database.
Step 207 judges whether the similar value of the image to be measured is more than the first preset value, if so then execute step 208,
If otherwise terminating flow.
Step 208 executes predetermined registration operation.
Embodiment 3
Referring to Fig. 3, the present embodiment is substantially the same manner as Example 1, the difference is that only：
Step 104 is：Judge whether the similar value of the image to be measured is more than the first preset value, if so then execute step
105,306 are thened follow the steps if not.
Step 105 executes predetermined registration operation, then terminates flow.
The image for user's head portrait that step 306, acquisition are reported.
For the violation picture by detection, the application can also receive report.
Step 307 will be compared by the characteristic point of report image in newest target database；
Step 308, judge it is whether legal by report image, if so then execute step 309, then follow the steps 310 if not,
Step 309, artificial detection by report image and by handmarking it is illegal target data is added to by report image
In library, then terminate flow.
Step 310 will be added to by report image in target database.
It is described it is legal be exactly judge image to be measured similar value whether be more than the first preset value.The present embodiment is by receiving act
Report can be improved from recognition capability from now on.
Embodiment 4
The present embodiment is substantially the same manner as Example 1, the difference is that only：
Step 101 includes by the text information in Feature point recognition image to be measured in embodiment 1.
Predetermined registration operation in embodiment 1 in step 105 includes reporting that the text information is recorded according to the text information
Contact method.
The application can not only prevent the propagation of illegal picture, additionally it is possible to effectively hit the spreading side of illegal picture.
Embodiment 5
Referring to Fig. 4, the present embodiment is substantially the same manner as Example 1, the difference is that only：
Step 500, the image to be measured for obtaining user's head portrait that user uploads.
Step 501 judges whether the image to be measured is dynamic image, if so then execute step 502, if otherwise executing reality
Apply the step 101 in example 1.
Step 502 intercepts the dynamic image by predetermined interval duration and obtains sub-image to be measured；
Characteristic point in step 503, identification sub-image to be measured；
The characteristic point is compared step 504 in target database, and the target database is several numbers
According to one in library；
Step 505, the similar value for obtaining sub-image to be measured and image in target database；
Step 506, when the similar value of sub-image to be measured be more than the first preset value when, execute predetermined registration operation.
Embodiment 6
Referring to Fig. 5, the present embodiment is substantially the same manner as Example 1, the difference is that only：
Step 600, the association platform for obtaining user；
Step 601, the head portrait for obtaining user in the association platform are image to be measured；
Step 602, using the corresponding database of the association platform as target database, then execute in embodiment 1
Step 102.
In the present embodiment, existing network registration is mostly registered by way of cross-platform registration, has some credit ratings high
Platform head image information can remove from audit or audited by corresponding database.The time is saved, efficiency is improved.
Although specific embodiments of the present invention have been described above, it will be appreciated by those of skill in the art that these
It is merely illustrative of, protection scope of the present invention is defined by the appended claims.Those skilled in the art is not carrying on the back
Under the premise of from the principle and substance of the present invention, many changes and modifications may be made, but these are changed
Protection scope of the present invention is each fallen with modification.
Claims (10)
1. a kind of image detection method is used for image detection device, which is characterized in that the image detection device includes several numbers
According to library, the image detection method includes：
Obtain the image to be measured for user's head portrait that user uploads；
Identify the characteristic point in the image to be measured；
The characteristic point is compared in target database, the target database is one in several databases
It is a；
Obtain the similar value of the image to be measured and image in target database；
When the similar value of the image to be measured is more than the first preset value, predetermined registration operation is executed.
2. image detection method as described in claim 1, which is characterized in that several databases include relate to yellow database,
Political sensitivity database and cruelly fear database.
3. image detection method as claimed in claim 2, which is characterized in that the characteristic point packet in the identification image to be measured
It includes：
Identify the characteristic point of skin in the image to be measured；
Judge whether in advance more than second skin area accounts for the ratio of imagery coverage to be measured in image to be measured according to the characteristic point of skin
If value, if it is target database then to relate to yellow database described in setting.
4. image detection method as claimed in claim 3, which is characterized in that the characteristic point packet in the identification image to be measured
It includes：
Judge whether in advance more than second skin area accounts for the ratio of imagery coverage to be measured in image to be measured according to the characteristic point of skin
If value, if otherwise judging, skin area in image to be measured accounts for the ratio of imagery coverage to be measured and whether is less than third preset value, if small
The political sensitivity database is then set in third preset value and probably database is target database cruelly.
5. image detection method as described in claim 1, which is characterized in that the image detection method includes：
Obtain the image for the user's head portrait reported；
It will be compared in newest target database by the characteristic point of report image；
Whether by report image legal, artificial detection is reported by report image and by handmarking is illegal if legal if judging
Image is added in target database, if not rule will be added to by report image in target database.
6. image detection method as described in claim 1, which is characterized in that the image detection method includes：
Pass through the text information in Feature point recognition image to be measured；
When the similar value of the image to be measured is more than the first preset value, predetermined registration operation is executed, the predetermined registration operation includes basis
The text information reports the contact method that the text information is recorded.
7. image detection method as described in claim 1, which is characterized in that obtain the shadow to be measured for user's head portrait that user uploads
As including：
Judge whether the image to be measured is dynamic image, intercepts the dynamic image if then pressing predetermined interval duration and obtain
Sub-image to be measured；
Identify the characteristic point in sub-image to be measured；
The characteristic point is compared in target database, the target database is one in several databases
It is a；
Obtain the similar value of sub-image to be measured and image in target database；
When the similar value of sub-image to be measured is more than the first preset value, predetermined registration operation is executed.
8. image detection method as described in claim 1, which is characterized in that the predetermined registration operation includes the image mark to be measured
It is denoted as illegal and the image to be measured is added in target database.
9. image detection method as described in claim 1, which is characterized in that the image detection method includes：
Obtain the association platform of user；
The head portrait for obtaining user in the association platform is image to be measured；
Using the corresponding database of the association platform as target database.
10. a kind of image detection device, which is characterized in that the image detection device is appointed for realizing in such as claim 1 to 9
Image detection method described in meaning one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810561124.5A CN108805069A (en) | 2018-06-04 | 2018-06-04 | Image detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810561124.5A CN108805069A (en) | 2018-06-04 | 2018-06-04 | Image detection method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108805069A true CN108805069A (en) | 2018-11-13 |
Family
ID=64090286
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810561124.5A Pending CN108805069A (en) | 2018-06-04 | 2018-06-04 | Image detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108805069A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101303734A (en) * | 2008-06-25 | 2008-11-12 | 腾讯科技（深圳）有限公司 | System and method for detecting picture |
| CN101996314A (en) * | 2009-08-26 | 2011-03-30 | 厦门市美亚柏科信息股份有限公司 | Content-based human body upper part sensitive image identification method and device |
| CN102073841A (en) * | 2009-11-20 | 2011-05-25 | 中国移动通信集团广东有限公司 | Poor video detection method and device |
| CN103093180A (en) * | 2011-10-28 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Method and system for detecting pornography images |
| US20130156275A1 (en) * | 2011-12-20 | 2013-06-20 | Matthew W. Amacker | Techniques for grouping images |
| CN103294813A (en) * | 2013-06-07 | 2013-09-11 | 北京捷成世纪科技股份有限公司 | Sensitive image search method and device |
| CN105740752A (en) * | 2014-12-11 | 2016-07-06 | 世纪龙信息网络有限责任公司 | Method and system for sensitive image filtering |
| CN106973305A (en) * | 2017-03-20 | 2017-07-21 | 广东小天才科技有限公司 | Method and device for detecting bad content in video |
| CN107203765A (en) * | 2017-03-30 | 2017-09-26 | 腾讯科技（上海）有限公司 | Sensitive Image Detection Method and device |
| CN107864401A (en) * | 2017-11-08 | 2018-03-30 | 北京密境和风科技有限公司 | It is a kind of based on live monitoring method, device, system and terminal device |
| CN108040262A (en) * | 2018-01-25 | 2018-05-15 | 湖南机友科技有限公司 | Live audio and video are reflected yellow method and device in real time |
-
2018
- 2018-06-04 CN CN201810561124.5A patent/CN108805069A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101303734A (en) * | 2008-06-25 | 2008-11-12 | 腾讯科技（深圳）有限公司 | System and method for detecting picture |
| CN101996314A (en) * | 2009-08-26 | 2011-03-30 | 厦门市美亚柏科信息股份有限公司 | Content-based human body upper part sensitive image identification method and device |
| CN102073841A (en) * | 2009-11-20 | 2011-05-25 | 中国移动通信集团广东有限公司 | Poor video detection method and device |
| CN103093180A (en) * | 2011-10-28 | 2013-05-08 | 阿里巴巴集团控股有限公司 | Method and system for detecting pornography images |
| US20130156275A1 (en) * | 2011-12-20 | 2013-06-20 | Matthew W. Amacker | Techniques for grouping images |
| CN103294813A (en) * | 2013-06-07 | 2013-09-11 | 北京捷成世纪科技股份有限公司 | Sensitive image search method and device |
| CN105740752A (en) * | 2014-12-11 | 2016-07-06 | 世纪龙信息网络有限责任公司 | Method and system for sensitive image filtering |
| CN106973305A (en) * | 2017-03-20 | 2017-07-21 | 广东小天才科技有限公司 | Method and device for detecting bad content in video |
| CN107203765A (en) * | 2017-03-30 | 2017-09-26 | 腾讯科技（上海）有限公司 | Sensitive Image Detection Method and device |
| CN107864401A (en) * | 2017-11-08 | 2018-03-30 | 北京密境和风科技有限公司 | It is a kind of based on live monitoring method, device, system and terminal device |
| CN108040262A (en) * | 2018-01-25 | 2018-05-15 | 湖南机友科技有限公司 | Live audio and video are reflected yellow method and device in real time |
Non-Patent Citations (1)
| Title |
|---|
| 叶志强: "《社交电视系统关键支撑技术与应用案例system key supporting technology and application》", 31 January 2016 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102195922B1 (en) | Internet-based facial beautification system | |
| CN106557747B (en) | The method and device of identification insurance single numbers | |
| CN105718869B (en) | The method and apparatus of face face value in a kind of assessment picture | |
| CN109840588A (en) | Neural network model training method, device, computer equipment and storage medium | |
| CN106530194B (en) | Method and device for detecting suspected infringement product picture | |
| CN106484837A (en) | The detection method of similar video file and device | |
| CN110674790B (en) | Abnormal scene processing method and system in video monitoring | |
| CN104636764B (en) | A kind of image latent writing analysis method and its device | |
| CN103839083B (en) | The detection method that a kind of JPEG picture quantization table is distorted | |
| CN103605992B (en) | A kind of nude picture detection method in electric power intranet and extranet interaction | |
| CN107302520A (en) | A kind of dynamic anti-leak of data and method for early warning and system | |
| CN105468922B (en) | A kind of live electromagnetic environment Signal online analysis and intimidation estimating method | |
| CN106921969A (en) | Terminal authenticity verification method, apparatus and system | |
| CN108229493A (en) | Banknote validation method, apparatus and electronic equipment | |
| CN106779449A (en) | A kind of collection register method of labor service real-name management | |
| CN109224444A (en) | Game charater recognition methods, device, storage medium and electronic equipment | |
| CN110493142A (en) | Mobile applications Activity recognition method based on spectral clustering and random forests algorithm | |
| CN110223202A (en) | A kind of method and system of teaching stage property identification and scoring | |
| CN110457601A (en) | The recognition methods and device of social account, storage medium and electronic device | |
| CN109918444A (en) | Training/verifying/management method/system, medium and equipment of model result | |
| CN107742068A (en) | A kind of implicit identity authorization system of the multi-source of smart machine and method | |
| CN106897880B (en) | Account risk assessment method and equipment | |
| CN108805069A (en) | Image detection method and device | |
| CN108921080A (en) | Image-recognizing method, device and electronic equipment | |
| CN105488413A (en) | Malicious code detection method and system based on information gain |
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 |
Application publication date: 20181113 |
|
| RJ01 | Rejection of invention patent application after publication |