CN101996180A - Picture examination and filter system and method - Google Patents
Picture examination and filter system and method Download PDFInfo
- Publication number
- CN101996180A CN101996180A CN2009100562915A CN200910056291A CN101996180A CN 101996180 A CN101996180 A CN 101996180A CN 2009100562915 A CN2009100562915 A CN 2009100562915A CN 200910056291 A CN200910056291 A CN 200910056291A CN 101996180 A CN101996180 A CN 101996180A
- Authority
- CN
- China
- Prior art keywords
- sign indicating
- indicating number
- picture
- file characteristic
- storehouse
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to a picture examination and filter system, and a method. The picture examination and filter system comprises a sensitive library, a picture matching module, a picture examination module and a filter module, wherein the picture matching module is connected with an external original picture and the sensitive library; the picture examination module is connected with the picture matching module and the filter module; and the sensitive library further comprises a picture feature code library and a file feature code library which are connected with the picture matching module respectively. The examination and filter method of the picture examination and filter system comprises the following steps of: generating the sensitive library; extracting original picture information, matching the extracted original picture information with the picture information in the sensitive library, and examining the original picture information; and filtering pictures. By detecting sensitive pictures from words and images, the invention can dynamically update the sensitive library so as to avoid repeated operation and improve the sensitive picture examination and filter efficiency.
Description
Technical field
The present invention relates to the picture detection range, particularly relate to a kind of picture audit filtering system and method.
Background technology
The internet has greatly merged global information, enriched the approach that people obtain information, expanded the scope of people's search information, yet the flame amount also sharply increases in its loaded information, the content that comprises in view of such information may relate to national security, social conduct and influence teen-age physical and mental health, be necessary this type of information is examined
General way has the employing manual examination and verification at present, but the manual examination and verification meeting causes same figure sector-meeting to be examined repeatedly, and workload is big, efficient is low, cost is high.
In the prior art, there is special software to come some sensitization pictures are handled in addition, as detecting porny according to skin color and attitude, yet, because human body complexion and attitude there are differences, thereby influence the result of its detection.
Also have a kind of picture detection system of the prior art and method that thereby the condition code of picture and the condition code of the sensitization picture in the blacklist are compared the detection sensitization picture, this then detection technique only compares the condition code of image, the scope limitation can cause the part sensitization picture to omit.
Summary of the invention
The purpose of this invention is to provide a kind of picture audit filtering system and method, from writings and image, can effectively detect sensitization picture, and can be by the responsive storehouse of artificial dynamic renewal to tackle new sensitization picture, prevented the duplication of labour, the efficient that the sensitization picture audit of raising is filtered.
In order to achieve the above object, technical scheme of the present invention is: a kind of picture audit filtering system comprises responsive storehouse, picture matching module, picture auditing module, filtering module; Described picture matching module is connected with responsive storehouse with the original image of outside, receives original image, and the information of original image and the information in the responsive storehouse are complementary; Described picture auditing module is connected with picture matching module and filtering module respectively, receive the information of the picture of picture matching module, and will be sent to filtering module by the picture of audit, wherein, described responsive storehouse also comprises picture feature sign indicating number storehouse and file characteristic sign indicating number storehouse, picture feature sign indicating number storehouse is connected with the picture matching module respectively with file characteristic sign indicating number storehouse, is respectively applied for the coupling of picture feature sign indicating number and file characteristic sign indicating number.
Above-mentioned picture audit filtering system, wherein, described picture matching module also comprises file characteristic sign indicating number extraction assembly, picture feature sign indicating number extraction assembly, file characteristic sign indicating number enquiring component and picture feature sign indicating number enquiring component;
Described file characteristic sign indicating number extraction assembly is connected with original image, receives the file characteristic sign indicating number in the original image;
Described picture feature sign indicating number extraction assembly is connected with file characteristic sign indicating number enquiring component, receives the picture that does not mate by file characteristic sign indicating number enquiring component;
Described file characteristic sign indicating number enquiring component is connected with file characteristic sign indicating number storehouse with file characteristic sign indicating number extraction assembly respectively, receives the file characteristic sign indicating number from file characteristic sign indicating number extraction assembly, and the file characteristic sign indicating number in this document condition code and the file characteristic sign indicating number storehouse is mated;
Described picture feature sign indicating number enquiring component is connected with picture feature sign indicating number extraction assembly and picture feature sign indicating number storehouse respectively, reception is from the picture feature sign indicating number of picture feature sign indicating number extraction assembly, and the picture feature sign indicating number in this picture feature sign indicating number and the picture feature sign indicating number storehouse is mated.
Above-mentioned picture audit filtering system, wherein, the picture auditing module also comprises file characteristic sign indicating number auditing module and picture feature sign indicating number auditing module;
Described file characteristic sign indicating number auditing module is connected with filtering module with file characteristic sign indicating number enquiring component respectively, receives the file characteristic sign indicating number that file characteristic sign indicating number enquiring component transmits, and will be sent to filtering module by the picture of audit;
Described picture feature sign indicating number audit is connected with filtering module with picture feature sign indicating number enquiring component respectively, receives the picture feature sign indicating number that picture feature sign indicating number enquiring component transmits, and will be sent to filtering module by the picture of audit.
Above-mentioned picture audit filtering system, wherein, described picture feature sign indicating number storehouse and file characteristic sign indicating number storehouse are respectively by different classes of classification.
The method that a kind of picture audit is filtered wherein, comprises the steps:
Step 1. generates responsive storehouse 1;
Have the file characteristic sign indicating number that mates with original image in the step 4. file characteristic sign indicating number storehouse 12, the file characteristic sign indicating number of 31 pairs of original images of file characteristic sign indicating number auditing module is examined;
Step 5. is a sensitization picture by the picture of audit, is filtered by filtering module 4;
Do not have the file characteristic sign indicating number that mates with original image in the step 6. file characteristic sign indicating number storehouse 12,32 pairs of these original images of picture feature sign indicating number extraction assembly carry out the picture feature sign indicating number and extract;
Step 7. picture feature sign indicating number enquiring component 24 is inquired about the picture feature sign indicating number that whether exists with the original image coupling in picture feature sign indicating number storehouse 11;
Have the picture feature sign indicating number that mates with original image in the step 8. picture feature sign indicating number storehouse 11, the picture feature sign indicating number of 32 pairs of original images of picture feature sign indicating number auditing module is examined;
Step 9. is a sensitization picture by the picture of audit, is filtered by filtering module 4.
Above-mentioned a kind of picture audit filter method, wherein, described step 1 is further comprising the steps of:
Step 1.1. obtains a pictures address from the picture library of delete flag;
Step 1.2. judges whether picture file exists and normally, picture file exists and be normal, extracts the file characteristic sign indicating number of picture, and picture file does not exist or be undesired, restarts;
Step 1.3. extraction document condition code;
Step 1.4. sets up file characteristic sign indicating number storehouse;
Step 1.5 is included the file characteristic sign indicating number of extraction file characteristic sign indicating number storehouse in and is judged the identical file characteristic sign indicating number of file characteristic sign indicating number that whether has existed in the file characteristic sign indicating number storehouse and extracted, there is identical file characteristic sign indicating number, file characteristic sign indicating number storehouse value counting adds 1, and get back to step 1.1 and restart, there is not identical file characteristic sign indicating number, updating file condition code storehouse, and carry out next step;
Step 1.6. extracts the picture feature sign indicating number;
Step 1.7. sets up picture feature sign indicating number storehouse;
Step 1.8. includes the picture feature sign indicating number that extracts picture feature sign indicating number storehouse in and judges the picture feature sign indicating number that whether has extraction in the picture feature sign indicating number storehouse, does not exist, and upgrades photo current condition code storehouse; Exist, the count value in picture feature sign indicating number storehouse adds 1, gets back to after the operation to begin to locate to restart, to finish the extraction to all picture feature sign indicating numbers.
Above-mentioned a kind of picture audit filter method, wherein, the extraction of described step 6 picture feature sign indicating number is further comprising the steps of:
Step 6.1. is divided into the N*N piece with picture;
Step 6.2. removes picture periphery piece;
Step 6.3. is by from left to right, and traversal is got in the middle of each piece the color value of any from top to bottom;
The front two of step 6.4. color value after fraction part rounds up, is got its integer divided by 10;
Step 6.5. is with the round values that the obtains character string that to be spliced into a length in order be N*N;
All pieces of step 6.6. carry out MD5 to character string after all handling, and the value that obtains is the picture feature sign indicating number.
The present invention makes it compared with prior art owing to adopt technique scheme, has the following advantages and good effect:
1, picture audit filtering system of the present invention and method are because from Word message and two aspect audits of picture colouring information and deletion sensitization picture, the audit and the deletion scope of sensitization picture have been enlarged, improved the accuracy rate of sensitization picture audit and deletion, simultaneously owing to detect from two aspects, can be judged to be sensitization picture by audit on the one hand, save audit time, improved the efficient of audit.
2, the responsive storehouse of picture audit filtering system of the present invention and method institute foundation can manually be upgraded dynamically, and picture audit filtering system is grown with each passing hour, and science is proper.
Description of drawings
Fig. 1 is a picture audit filtering system one-piece construction synoptic diagram of the present invention.
Fig. 2 is a picture audit filtering system inner structure synoptic diagram of the present invention.
Fig. 3 is the process flow diagram of picture audit filter method of the present invention.
Fig. 4 is the process flow diagram that the responsive storehouse in the picture audit filtering system of the present invention is set up.
Fig. 5 is the process flow diagram that the picture feature sign indicating number extracts in the picture audit filter method of the present invention.
Fig. 6 is the enforcement illustration of picture audit filter method of the present invention.
Embodiment
See also accompanying drawing 1,, comprise responsive storehouse 1, picture matching module 2, picture auditing module 3, filtering module 4, picture feature sign indicating number storehouse 11 and file characteristic sign indicating number storehouse 12 for picture audit filtering system one-piece construction synoptic diagram of the present invention.
Wherein, be provided with picture feature sign indicating number storehouse 11 and file characteristic sign indicating number storehouse 12 in the responsive storehouse 1, picture feature sign indicating number storehouse 11 and file characteristic sign indicating number storehouse 12 are respectively by different classes of classification.As political class sensitization picture, the pornographic class sensitization picture of violence etc., to doing above-mentioned classification mark in the storehouse, to realize detecting different classes of sensitization picture at different demands.Picture matching module 2 is connected with file characteristic sign indicating number storehouse 12 with the original image of outside and the picture feature sign indicating number storehouse 11 in the responsive storehouse 1 respectively, receive original image, and with the condition code of original image respectively with picture feature sign indicating number storehouse 11 and file characteristic sign indicating number storehouse 12 in condition code be complementary; Picture auditing module 3 is connected with picture matching module 2 and filtering module 4 respectively, receives the information of the picture of picture matching module 2, and will be sent to filtering module 4 by the picture of audit.
See also accompanying drawing 2, be picture audit filtering system inner structure synoptic diagram, picture matching module 2 comprises file characteristic sign indicating number extraction assembly 21, picture feature sign indicating number extraction assembly 23, file characteristic sign indicating number enquiring component 22 and picture feature sign indicating number enquiring component 24, and picture auditing module 3 comprises file characteristic sign indicating number auditing module 31 and picture feature sign indicating number auditing module 32.
Wherein, file characteristic sign indicating number extraction assembly 21 is connected with original image, receives the file characteristic sign indicating number in the original image.Picture feature sign indicating number extraction assembly 23 is connected with file characteristic sign indicating number enquiring component 22, receives the picture that does not mate by file characteristic sign indicating number enquiring component 22.File characteristic sign indicating number enquiring component 22 is connected with file characteristic sign indicating number storehouse 12 with file characteristic sign indicating number extraction assembly 21 respectively, reception is from the file characteristic sign indicating number of file characteristic sign indicating number extraction assembly 21, and the file characteristic sign indicating number in this document condition code and the file characteristic sign indicating number storehouse 12 is mated.Picture feature sign indicating number enquiring component 24 is connected with picture feature sign indicating number extraction assembly 23 and picture feature sign indicating number storehouse 11 respectively, reception is from the picture feature sign indicating number of picture feature sign indicating number extraction assembly 23, and the picture feature sign indicating number in this picture feature sign indicating number and the picture feature sign indicating number storehouse 11 is mated.File characteristic sign indicating number auditing module 31 is connected with filtering module 4 with file characteristic sign indicating number enquiring component 22 respectively, receives the file characteristic sign indicating number that file characteristic sign indicating number enquiring component 22 transmits, and will be sent to filtering module 4 by the picture of audit; Picture feature sign indicating number audit 32 is connected with filtering module 4 with picture feature sign indicating number enquiring component 24 respectively, receives the picture feature sign indicating number that picture feature sign indicating number enquiring component 24 transmits, and will be sent to filtering module 4 by the picture of audit.
See also accompanying drawing 3, the process flow diagram for picture audit filter method comprises the steps:
Generate responsive storehouse 1; File characteristic sign indicating number extraction assembly 21 extracts the file characteristic sign indicating number of original image, the md5 value of this document condition code for obtaining according to file content, and the identical expression file of md5 value is also identical; File characteristic sign indicating number enquiring component 22 is inquired about the file characteristic sign indicating number that whether exists with the original image coupling in file characteristic sign indicating number storehouse 12; If there is the file characteristic sign indicating number that mates with original image in the file characteristic sign indicating number storehouse 12, the file characteristic sign indicating number of 31 pairs of original images of file characteristic sign indicating number auditing module is examined, and whether the audit document condition code is counted greater than setting value, is audit greater than setting value and passes through; Picture by audit is a sensitization picture, is filtered by filtering module 4; If there is not file characteristic sign indicating number in the file characteristic sign indicating number storehouse 12 with the original image coupling, 32 pairs of these original images of picture feature sign indicating number extraction assembly carry out the picture feature sign indicating number and extract, the extraction of picture feature sign indicating number is that this numerical value can be used to a corresponding unique pictures according to the abstract numerical value that comes out of pixel color, positional information; Picture feature sign indicating number enquiring component 24 is inquired about the picture feature sign indicating number that whether exists with the original image coupling in picture feature sign indicating number storehouse 11; If there is the picture feature sign indicating number that mates with original image in the picture feature sign indicating number storehouse 11, the picture feature sign indicating number of 32 pairs of original images of picture feature sign indicating number auditing module is examined, and whether audit picture feature yardage number is greater than setting value; Picture by audit is a sensitization picture, is filtered by filtering module 4.
Please cooperate referring to Fig. 4, generate responsive storehouse 1 and may further comprise the steps:
From the picture library of delete flag, obtain a pictures address; Judge whether picture file exists and normally,, extract the file characteristic sign indicating number of picture,, restart if picture file does not exist or be undesired if picture file exists and be normal; When picture file existence and normal, extraction document condition code; Set up file characteristic sign indicating number storehouse; The file characteristic sign indicating number of extraction is included in file characteristic sign indicating number storehouse and judged the identical file characteristic sign indicating number of file characteristic sign indicating number that whether has existed in the file characteristic sign indicating number storehouse and extracted, if there is identical file characteristic sign indicating number, file characteristic sign indicating number storehouse value counting adds 1, and get back to step 1.1 and restart, there is not identical file characteristic sign indicating number, updating file condition code storehouse, and carry out next step; Extract the picture feature sign indicating number; Set up picture feature sign indicating number storehouse; The picture feature sign indicating number that extracts is included in picture feature sign indicating number storehouse and judged the picture feature sign indicating number that whether has extraction in the picture feature sign indicating number storehouse, if there is no, upgrade photo current condition code storehouse; If exist, the count value in picture feature sign indicating number storehouse adds 1, gets back to after the operation to begin to locate to restart, to finish the extraction to all picture feature sign indicating numbers.The present invention sets up responsive storehouse in the audit mode that manually adds program, upgrades responsive storehouse dynamically, constantly improves the information in the responsive storehouse, guarantees to examine and delete the comprehensive of sensitive information.
Please cooperate referring to Fig. 5, the extraction of picture feature sign indicating number is further comprising the steps of:
Picture is divided into the N*N piece; Remove picture periphery piece; By from left to right, traversal is got in the middle of each piece the color value of any from top to bottom, and this color value refers to the color index value of certain point on the image, represents current colouring information, is the rgb value that returns this point with integer; The front two of color value after fraction part rounds up, is got its integer divided by 10; With the round values that the obtains character string that to be spliced into a length in order be N*N; After all pieces are all handled, character string is carried out MD5, the value that obtains is the picture feature sign indicating number.Can judge by the distribution of colouring information on the picture whether two pictures are similar, so the picture feature sign indicating number can effectively be used for carrying out the audit of picture.
As shown in Figure 6, be a concrete example of using of the present invention, picture matching module 2 extracts the file characteristic sign indicating number of this figure, with the file characteristic sign indicating number coupling in this document condition code and the responsive storehouse 1 that generated, the match is successful, and 31 pairs of this document condition codes of file characteristic sign indicating number auditing module are examined, if file characteristic yardage number is greater than setting value, illustrating has the file characteristic sign indicating number identical with this figure to exist in the responsive storehouse 1, promptly this figure is a sensitization picture, thereby delivers to filtering module 4 and filter; If file characteristic sign indicating number coupling is unsuccessful, picture matching module 2 extracts the picture feature sign indicating number of this figure, picture is divided into the N*N piece, remove picture periphery piece, press from left to right, traversal is got in the middle of each piece the color value of any from top to bottom, this color value refers to the color index value of certain point on the image, represent current colouring information, it is the rgb value that returns this point with integer, whether the distribution decidable picture by colouring information on the picture is similar, therefore picture matching module 2 mates picture feature sign indicating number in the picture feature sign indicating number of this figure and the responsive storehouse 1, if the match is successful, examine through picture feature sign indicating number auditing module 32, as picture feature yardage number greater than setting value, the existence picture feature sign indicating number identical with this figure in the responsive storehouse is described, also promptly this figure is a sensitization picture, delivering to filtering module 4 filters, the responsive storehouse of picture audit filtering system of the present invention and method can be upgraded at any time by the mode that manually adds program, guaranteed the science of picture audit, the present invention's mode of examining two aspect dual fail-safes from file characteristic sign indicating number audit and picture feature sign indicating number is examined in addition, has guaranteed the accuracy that picture is examined.
Claims (7)
1. a picture audit filtering system comprises responsive storehouse (1), picture matching module (2), picture auditing module (3), filtering module (4); Described picture matching module (2) is connected with responsive storehouse (1) with the original image of outside, receives original image, and the information of original image and the information in responsive storehouse (1) are complementary; Described picture auditing module (3) is connected with picture matching module (2) and filtering module (4) respectively, receive the information of the picture of picture matching module (2), and will be sent to filtering module (4) by the picture of audit, it is characterized in that described responsive storehouse (1) also comprises picture feature sign indicating number storehouse (11) and file characteristic sign indicating number storehouse (12), picture feature sign indicating number storehouse (11) is connected with picture matching module (2) respectively with file characteristic sign indicating number storehouse (12), is respectively applied for the coupling of picture feature sign indicating number and file characteristic sign indicating number.
2. picture audit filtering system as claimed in claim 1, it is characterized in that described picture matching module (2) also comprises file characteristic sign indicating number extraction assembly (21), picture feature sign indicating number extraction assembly (23), file characteristic sign indicating number enquiring component (22) and picture feature sign indicating number enquiring component (24);
Described file characteristic sign indicating number extraction assembly (21) is connected with original image, receives the file characteristic sign indicating number in the original image;
Described picture feature sign indicating number extraction assembly (23) is connected with file characteristic sign indicating number enquiring component (22), receives the picture that does not mate by file characteristic sign indicating number enquiring component (22);
Described file characteristic sign indicating number enquiring component (22) is connected with file characteristic sign indicating number storehouse (12) with file characteristic sign indicating number extraction assembly (21) respectively, reception is from the file characteristic sign indicating number of file characteristic sign indicating number extraction assembly (21), and the file characteristic sign indicating number in this document condition code and the file characteristic sign indicating number storehouse (12) is mated;
Described picture feature sign indicating number enquiring component (24) is connected with picture feature sign indicating number extraction assembly (23) and picture feature sign indicating number storehouse (11) respectively, reception is from the picture feature sign indicating number of picture feature sign indicating number extraction assembly (23), and the picture feature sign indicating number in this picture feature sign indicating number and the picture feature sign indicating number storehouse (11) is mated.
3. picture audit filtering system as claimed in claim 1 is characterized in that picture auditing module (3) also comprises file characteristic sign indicating number auditing module (31) and picture feature sign indicating number auditing module (32);
Described file characteristic sign indicating number auditing module (31) is connected with filtering module (4) with file characteristic sign indicating number enquiring component (22) respectively, receives the file characteristic sign indicating number that file characteristic sign indicating number enquiring component (22) transmits, and will be sent to filtering module (4) by the picture of audit;
Described picture feature sign indicating number audit (32) is connected with filtering module (4) with picture feature sign indicating number enquiring component (24) respectively, receives the picture feature sign indicating number that picture feature sign indicating number enquiring component (24) transmits, and will be sent to filtering module (4) by the picture of audit.
4. picture audit filtering system as claimed in claim 1, it is characterized in that, described picture feature sign indicating number storehouse (11) and file characteristic sign indicating number storehouse (12) are respectively by different classes of classification, and this classification comprises political class sensitization picture, the pornographic class sensitization picture of violence, and above-mentioned classification mark is done in the storehouse.
5. a method that is used for the picture audit filtration of picture audit filtering system is characterized in that, comprises the steps:
Step 1. generates responsive storehouse (1);
Step 2. file characteristic sign indicating number extraction assembly (21) extracts the file characteristic sign indicating number of original image;
Whether step 3. file characteristic sign indicating number enquiring component (22) inquiry in file characteristic sign indicating number storehouse (12) exists the file characteristic sign indicating number with the original image coupling;
If there is the file characteristic sign indicating number that mates with original image in the step 4. file characteristic sign indicating number storehouse (12), file characteristic sign indicating number auditing module (31) is examined the file characteristic sign indicating number of original image;
Step 5. is a sensitization picture by the picture of audit, is filtered by filtering module (4);
If there is not the file characteristic sign indicating number that mates with original image in the step 6. file characteristic sign indicating number storehouse (12), picture feature sign indicating number extraction assembly (32) carries out the picture feature sign indicating number to this original image and extracts;
Whether step 7. picture feature sign indicating number enquiring component (24) inquiry in picture feature sign indicating number storehouse (11) exists the picture feature sign indicating number with the original image coupling;
If there is the picture feature sign indicating number that mates with original image in the step 8. picture feature sign indicating number storehouse (11), picture feature sign indicating number auditing module (32) is examined the picture feature sign indicating number of original image;
Step 9. is a sensitization picture by the picture of audit, is filtered by filtering module (4).
6. a kind of picture audit filter method as claimed in claim 5 is characterized in that described step 1 is further comprising the steps of:
Step 1.1. obtains a pictures address from the picture library of delete flag;
Step 1.2. judges whether picture file exists and normally, picture file exists and be normal, extracts the file characteristic sign indicating number of picture, and picture file does not exist or be undesired, restarts;
Step 1.3. extraction document condition code;
Step 1.4. sets up file characteristic sign indicating number storehouse;
Step 1.5 is included the file characteristic sign indicating number that extracts in file characteristic sign indicating number storehouse, and judge whether existed in the file characteristic sign indicating number storehouse and the identical file characteristic sign indicating number of file characteristic sign indicating number that extracts, if there is identical file characteristic sign indicating number, file characteristic sign indicating number storehouse value counting adds 1, and get back to step 1.1 and restart, if there is no identical file characteristic sign indicating number, updating file condition code storehouse, and carry out next step;
Step 1.6. extracts the picture feature sign indicating number;
Step 1.7. sets up picture feature sign indicating number storehouse;
Step 1.8. includes the picture feature sign indicating number that extracts picture feature sign indicating number storehouse in and judges the picture feature sign indicating number that whether has extraction in the picture feature sign indicating number storehouse, does not exist, and upgrades photo current condition code storehouse; Exist, the count value in picture feature sign indicating number storehouse adds 1, gets back to after the operation to begin to locate to restart, to finish the extraction to all picture feature sign indicating numbers.
7. a kind of picture audit filter method as claimed in claim 5 is characterized in that the extraction of described step 6 picture feature sign indicating number is further comprising the steps of:
Step 6.1. is divided into the N*N piece with picture;
Step 6.2. removes picture periphery piece;
Step 6.3. is by from left to right, and traversal is got in the middle of each piece the color value of any from top to bottom;
The front two of step 6.4. color value after fraction part rounds up, is got its integer divided by 10;
Step 6.5. is with the round values that the obtains character string that to be spliced into a length in order be N*N;
All pieces of step 6.6. carry out MD5 to character string after all handling, and the value that obtains is the picture feature sign indicating number.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009100562915A CN101996180A (en) | 2009-08-12 | 2009-08-12 | Picture examination and filter system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009100562915A CN101996180A (en) | 2009-08-12 | 2009-08-12 | Picture examination and filter system and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101996180A true CN101996180A (en) | 2011-03-30 |
Family
ID=43786347
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009100562915A Pending CN101996180A (en) | 2009-08-12 | 2009-08-12 | Picture examination and filter system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101996180A (en) |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102902675A (en) * | 2011-07-26 | 2013-01-30 | 腾讯科技(深圳)有限公司 | Picture content approval method and device |
CN103116752A (en) * | 2013-02-25 | 2013-05-22 | 新浪网技术(中国)有限公司 | Picture auditing method and system |
CN104331414A (en) * | 2014-09-28 | 2015-02-04 | 北京云巢动脉科技有限公司 | Sensitive data filtering method and system based on data blocks |
CN105159902A (en) * | 2014-04-04 | 2015-12-16 | 卡姆芬德公司 | Image processing method based on priority |
CN105260423A (en) * | 2015-09-28 | 2016-01-20 | 北京橙鑫数据科技有限公司 | Duplicate removal method and apparatus for electronic cards |
CN105654057A (en) * | 2015-12-31 | 2016-06-08 | 中国建设银行股份有限公司 | Picture auditing system and picture auditing method based on picture contents |
CN105808630A (en) * | 2014-12-31 | 2016-07-27 | 广州市动景计算机科技有限公司 | Android application auditing method and apparatus |
CN107729924A (en) * | 2017-09-25 | 2018-02-23 | 平安科技(深圳)有限公司 | Picture review probability interval generation method and picture review decision method |
CN107992599A (en) * | 2017-12-13 | 2018-05-04 | 厦门市美亚柏科信息股份有限公司 | File comparison method and system |
CN108470028A (en) * | 2017-02-23 | 2018-08-31 | 北京唱吧科技股份有限公司 | A kind of picture examination method and apparatus |
CN108765085A (en) * | 2018-05-30 | 2018-11-06 | 杭州骑轻尘信息技术有限公司 | Vehicle order checking method, device and readable storage medium storing program for executing |
CN108960782A (en) * | 2018-07-10 | 2018-12-07 | 北京木瓜移动科技股份有限公司 | content auditing method and device |
CN109033416A (en) * | 2018-08-06 | 2018-12-18 | 钱文浩 | Electronic auditing automatic trigger method |
CN109344258A (en) * | 2018-11-28 | 2019-02-15 | 中国电子科技网络信息安全有限公司 | A kind of intelligent self-adaptive sensitive data identifying system and method |
CN109639742A (en) * | 2019-02-01 | 2019-04-16 | 温州大学 | A kind of information spy system |
CN109902243A (en) * | 2019-03-04 | 2019-06-18 | 广州博士信息技术研究院有限公司 | A kind of data filtering method for digging with first screening function |
CN110083733A (en) * | 2019-03-16 | 2019-08-02 | 平安城市建设科技(深圳)有限公司 | Picture examination method, apparatus, equipment and computer readable storage medium |
CN110992446A (en) * | 2019-12-04 | 2020-04-10 | 杭州三体视讯科技有限公司 | Picture auditing method |
CN111079816A (en) * | 2019-12-11 | 2020-04-28 | 北京金山云网络技术有限公司 | Image auditing method and device and server |
CN111967245A (en) * | 2020-07-09 | 2020-11-20 | 福建亿榕信息技术有限公司 | Method and device for automatically auditing verification document and computer equipment |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1761205A (en) * | 2005-11-18 | 2006-04-19 | 郑州金惠计算机系统工程有限公司 | System for detecting eroticism and unhealthy images on network based on content |
CN101303734A (en) * | 2008-06-25 | 2008-11-12 | 腾讯科技(深圳)有限公司 | System and method for detecting picture |
-
2009
- 2009-08-12 CN CN2009100562915A patent/CN101996180A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1761205A (en) * | 2005-11-18 | 2006-04-19 | 郑州金惠计算机系统工程有限公司 | System for detecting eroticism and unhealthy images on network based on content |
CN101303734A (en) * | 2008-06-25 | 2008-11-12 | 腾讯科技(深圳)有限公司 | System and method for detecting picture |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102902675B (en) * | 2011-07-26 | 2017-10-10 | 腾讯科技(深圳)有限公司 | Picture content approval method and device |
CN102902675A (en) * | 2011-07-26 | 2013-01-30 | 腾讯科技(深圳)有限公司 | Picture content approval method and device |
CN103116752A (en) * | 2013-02-25 | 2013-05-22 | 新浪网技术(中国)有限公司 | Picture auditing method and system |
CN105159902A (en) * | 2014-04-04 | 2015-12-16 | 卡姆芬德公司 | Image processing method based on priority |
CN104331414A (en) * | 2014-09-28 | 2015-02-04 | 北京云巢动脉科技有限公司 | Sensitive data filtering method and system based on data blocks |
CN105808630B (en) * | 2014-12-31 | 2020-03-31 | 广州市动景计算机科技有限公司 | Android application auditing method and device |
CN105808630A (en) * | 2014-12-31 | 2016-07-27 | 广州市动景计算机科技有限公司 | Android application auditing method and apparatus |
CN105260423A (en) * | 2015-09-28 | 2016-01-20 | 北京橙鑫数据科技有限公司 | Duplicate removal method and apparatus for electronic cards |
CN105654057A (en) * | 2015-12-31 | 2016-06-08 | 中国建设银行股份有限公司 | Picture auditing system and picture auditing method based on picture contents |
CN108470028A (en) * | 2017-02-23 | 2018-08-31 | 北京唱吧科技股份有限公司 | A kind of picture examination method and apparatus |
CN108470028B (en) * | 2017-02-23 | 2020-08-18 | 北京唱吧科技股份有限公司 | Picture auditing method and device |
CN107729924A (en) * | 2017-09-25 | 2018-02-23 | 平安科技(深圳)有限公司 | Picture review probability interval generation method and picture review decision method |
CN107992599A (en) * | 2017-12-13 | 2018-05-04 | 厦门市美亚柏科信息股份有限公司 | File comparison method and system |
CN108765085A (en) * | 2018-05-30 | 2018-11-06 | 杭州骑轻尘信息技术有限公司 | Vehicle order checking method, device and readable storage medium storing program for executing |
CN108960782A (en) * | 2018-07-10 | 2018-12-07 | 北京木瓜移动科技股份有限公司 | content auditing method and device |
CN109033416A (en) * | 2018-08-06 | 2018-12-18 | 钱文浩 | Electronic auditing automatic trigger method |
CN109344258A (en) * | 2018-11-28 | 2019-02-15 | 中国电子科技网络信息安全有限公司 | A kind of intelligent self-adaptive sensitive data identifying system and method |
CN109344258B (en) * | 2018-11-28 | 2021-11-12 | 中国电子科技网络信息安全有限公司 | Intelligent self-adaptive sensitive data identification system and method |
CN109639742A (en) * | 2019-02-01 | 2019-04-16 | 温州大学 | A kind of information spy system |
CN109902243A (en) * | 2019-03-04 | 2019-06-18 | 广州博士信息技术研究院有限公司 | A kind of data filtering method for digging with first screening function |
CN110083733A (en) * | 2019-03-16 | 2019-08-02 | 平安城市建设科技(深圳)有限公司 | Picture examination method, apparatus, equipment and computer readable storage medium |
CN110992446A (en) * | 2019-12-04 | 2020-04-10 | 杭州三体视讯科技有限公司 | Picture auditing method |
CN111079816A (en) * | 2019-12-11 | 2020-04-28 | 北京金山云网络技术有限公司 | Image auditing method and device and server |
CN111967245A (en) * | 2020-07-09 | 2020-11-20 | 福建亿榕信息技术有限公司 | Method and device for automatically auditing verification document and computer equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101996180A (en) | Picture examination and filter system and method | |
US8095547B2 (en) | Method and apparatus for detecting spam user created content | |
CN102801859B (en) | Method and device for identifying junk short message, and mobile communication terminal with device | |
US20150339518A1 (en) | Electronic device, electronic device control method, and computer-readable recording medium having stored thereon electronic device control program | |
US20180060314A1 (en) | Multibyte heterogeneous log preprocessing | |
CN105808251A (en) | Virtual function table based method and system for bypassing security detection by hijack | |
CN106021301B (en) | Data comparison system and method for different file formats | |
Gurgel-Gonçalves et al. | Automated identification of insect vectors of Chagas disease in Brazil and Mexico: the Virtual Vector Lab | |
CN106502984A (en) | A kind of method and device of field new word discovery | |
US8311271B2 (en) | Image processing apparatus and image processing method | |
CN105930313A (en) | Method and device for processing notification message | |
CN110069769A (en) | Using label generating method, device and storage equipment | |
CN116955720A (en) | Data processing method, apparatus, device, storage medium and computer program product | |
CN110782169A (en) | Method and device for updating business process | |
CN104346616A (en) | Character recognition device and character recognition method | |
CN107368500A (en) | Data pick-up method and system | |
CN110795069A (en) | Code analysis method, intelligent terminal and computer readable storage medium | |
CN107273432A (en) | A kind of patent article integration system and method based on big data | |
CN108280085B (en) | Data deduplication method and device | |
WO2018077059A1 (en) | Barcode identification method and apparatus | |
CN205302388U (en) | Tourism passenger management system concerning foreign affairs | |
CN113901898A (en) | Image stable sampling method, device, equipment and medium in face recognition scene | |
CN112579944A (en) | Page jump processing method and device and electronic equipment | |
US20170031924A1 (en) | Search device, method and program recording medium | |
CN112733523A (en) | Document sending method, device, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110330 |