CN106326824A - Method and device for automatically screening pictures - Google Patents
Method and device for automatically screening pictures Download PDFInfo
- Publication number
- CN106326824A CN106326824A CN201510394966.2A CN201510394966A CN106326824A CN 106326824 A CN106326824 A CN 106326824A CN 201510394966 A CN201510394966 A CN 201510394966A CN 106326824 A CN106326824 A CN 106326824A
- Authority
- CN
- China
- Prior art keywords
- region
- detection zone
- detected
- picture
- location
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000012216 screening Methods 0.000 title claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000001514 detection method Methods 0.000 claims description 83
- 238000012360 testing method Methods 0.000 claims description 25
- 238000013519 translation Methods 0.000 claims description 19
- 238000012544 monitoring process Methods 0.000 claims description 11
- 238000005194 fractionation Methods 0.000 claims description 6
- 238000003475 lamination Methods 0.000 claims description 6
- 238000011282 treatment Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 description 7
- 230000002950 deficient Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000003760 hair shine Effects 0.000 description 2
- 208000003464 asthenopia Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000005056 cell body Anatomy 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/162—Detection; Localisation; Normalisation using pixel segmentation or colour matching
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
-
- 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
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and device for automatically screening pictures. The method for automatically screening pictures comprises the following steps: firstly, performing background color processing and reducing pixel interference caused by the background; then, confirming locating areas in target pictures and removing pictures in which the locating areas are deviated; and then confirming areas to be tested according to the locating areas and judging whether the areas to be tested meet the setting requirements or not, and removing pictures that the areas to be tested do not meet the setting requirements. The device for automatically screening pictures comprises a color processing module, a locating area screening module and an area to be detected screening module. The method and the device for automatically screening pictures can automatically select pictures, reduce the workload of manual examination, and improve the accuracy rate.
Description
Technical field
The invention belongs to technical field of image processing, particularly relate to the one when putting veritification photo on record certainly
The method and device of dynamic screening picture.
Background technology
In order to strengthen the management to internet site, and the management to online operation electricity business, generally need
The operator of the head of a station of website, on-line shop is put on record.When putting on record, examination & verification is needed to put on record application
The veritification photograph of putting on record that people submits to, screens out background serious, portrait severe deviations, mould defective, reflective
The defective photos such as paste, reduce the system pressure of follow-up system explication de texte picture.
Prior art for put on record veritify photograph examination & verification typically by artificial screening, when examination & verification amount exceedes
During some, the visual fatigue of people can cause false judgment to steeply rise.
Summary of the invention
It is an object of the invention to provide the method and device of a kind of automatic screening picture, to solve existing skill
Art uses artificial screening workload big, the technical problem easily made mistakes.
To achieve these goals, technical solution of the present invention is as follows:
A kind of method of automatic screening picture, including:
Determine the region, location in Target Photo, remove the picture of region, location position deviation;
Region to be detected is determined, it is determined that whether region to be detected meets sets requirement according to region, location,
Remove region to be detected to be unsatisfactory for setting the picture required.
Further, before described method region, location in determining Target Photo, further comprise the steps of:
Carry out picture color process, reduce the pixel interference that background causes.By removing what background caused
Pixel interference, can carry out automatic screening more accurately.
Further, described Target Photo is veritification photograph of putting on record, described in put on record veritification according to be blue background,
Described carry out picture color process, reduce the pixel interference that background causes, including step:
By described veritification of putting on record according to splitting into R, tri-Color Channel layers of G, B;
By R, the G Color Channel lamination of fractionation also, B color channel layer is removed.
Target Photo of the present invention is veritification photograph of putting on record, described in put on record veritification according to be blue background, institute
State location region behaviour face region, described according to location region determine region to be detected, including step:
The distance set to left by human face region, using the region after translation as left side detection zone;
The distance set to right translation by human face region, using the region after translation as right side detection zone.
Further, whether described judgement region to be detected meets setting requirement, including step:
Take the test point setting quantity at left side detection zone, left side detection zone etc. is cut by described test point
Cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of detection zone on the right side of traversal, if Blue color domain pixel is more than the first threshold set, and
Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge to meet.
The region described to be detected of the present invention also includes that being located at picture specifies the left side shoulder detection in region
District and right shoulder detection zone.
Further, whether described judgement region to be detected meets the requirement set, including step:
Take the test point setting quantity at left side shoulder detection zone, described test point is by left side detection zone etc.
Cut and cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of traversal right shoulder detection zone, if Blue color domain pixel is more than the first threshold set
It is worth, and Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge satisfied.
The invention allows for the device of a kind of automatic screening picture, including:
Region, location screening module, for determining the region, location in Target Photo, removes region, location
The picture of position deviation;
Region to be detected screening module, for determining region to be detected according to region, location, it is determined that to be checked
Survey whether region meets setting requirement, remove region to be detected and be unsatisfactory for setting the picture required.
Further, described device also includes color treatments module, is used for carrying out picture color process,
Reduce the pixel interference that background causes.
Target Photo of the present invention is veritification photograph of putting on record, described in put on record veritification according to be blue background, institute
State color treatments module row picture color to process, reduce the pixel interference that background causes, perform following behaviour
Make:
By described veritification of putting on record according to splitting into R, tri-Color Channel layers of G, B;
By R, the G Color Channel lamination of fractionation also, B color channel layer is removed.
Region behaviour face region, location of the present invention, described region to be detected screening module is according to location
Region determines region to be detected, performs to operate as follows:
The distance set to left by human face region, using the region after translation as left side detection zone;
The distance set to right translation by human face region, using the region after translation as right side detection zone.
Further, screening module in described region to be detected judges whether region to be detected meets setting and want
Ask, perform to operate as follows:
Take the test point setting quantity at left side detection zone, left side detection zone etc. is cut by described test point
Cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of detection zone on the right side of traversal, if Blue color domain pixel is more than the first threshold set, and
Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge to meet.
Equally, described region to be detected also include being located at picture specify region left side shoulder detection zone and
Right shoulder detection zone.
Further, screening module in described region to be detected judges whether region to be detected meets setting and want
Ask, perform to operate as follows:
Take the test point setting quantity at left side shoulder detection zone, described test point is by left side detection zone etc.
Cut and cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of traversal right shoulder detection zone, if Blue color domain pixel is more than the first threshold set
It is worth, and Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge satisfied.
The method and device of a kind of automatic screening picture that the present invention proposes, true by recognition of face mode
Determine the location of face in photo, and obtain the pixel samples of key area in picture with face for reference,
And sample is done colour recognition, screen out background by the method defective, the photograph of portrait position's deviation
Sheet, improves examination & verification speed and examination & verification accuracy while reducing manual examination and verification pressure.
Accompanying drawing explanation
Fig. 1 is the method flow diagram of automatic screening picture of the present invention;
Fig. 2 is embodiment of the present invention area schematic to be detected;
Fig. 3 is that a kind of of present invention region to be detected judges embodiment schematic diagram;
Fig. 4 is the apparatus structure schematic diagram of automatic screening picture of the present invention.
Detailed description of the invention
With embodiment, technical solution of the present invention is described in further details below in conjunction with the accompanying drawings, real below
Execute example and do not constitute limitation of the invention.
According to the requirement of Ministry of Industry and Information, veritification of putting on record is according to needing to be front-face and hat-free photos half body photograph, and background is simultaneously
The wrongly written or mispronounced character curtain of the blue end of Access Service Provider's qualification title printing.The general thought of the present invention is to pass through face
Recognition method determines the location of face in photo, and with face for reference to obtaining key area in picture
Pixel samples, and sample is done colour recognition, screen out background by the method defective, people's image position
Put the photo of deviation.
As it is shown in figure 1, the method for the present embodiment a kind of automatic screening picture, comprise the steps:
S1, Target Photo is carried out picture color process, reduce the pixel interference that background causes.
The veritification of putting on record that needs are carried out auditing by the present embodiment is according to as Target Photo, owing to background is blue
Color wrongly written or mispronounced character, in order to reduce color interference, prepares for location region recognition, picture color to be carried out
Process, reduce the pixel interference that large area blueness curtain causes.
It is specifically divided into following steps:
Target Photo is split into R, tri-Color Channel layers of G, B;
By R, the G Color Channel lamination of fractionation also, B color channel layer is removed.
Thus eliminate blue background, it is beneficial to identify location region, i.e. human face region.
S2, the region, location determined in Target Photo, remove the picture of region, location position deviation.
The present embodiment is to veritify for putting on record to shine into capable picture screening, therefore by the face district in photo
People, as region, location, when human face region photo within the specified range is only qualified photo, is worked as in territory
Face region no longer specify in the range of time regard as position deviation, thus remove the photograph of portrait position's deviation
Sheet.
Target Photo is carried out face coupling, determines that human face region and coordinate thereof are current comparative maturity
Technology, is not discussed here.After determining human face region, it is judged that whether human face region is specifying
In the range of, if, illustrate that portrait position is normal, need to carry out follow-up judgement, otherwise
Directly judge that this Target Photo is undesirable, and remove.
The present embodiment human face region refers to the rectangular area that face determines, the appointment scope of human face region
It is:
The distance at the top margin image top edge, rectangular area that face determines, more than picture height
15%;
The distance at the bottom margin image base edge, rectangular area that face determines, more than picture height
25%;
The left side, rectangular area that face the determines distance away from image left side edge, more than picture width
25%;
The distance of the rectangular area rightmargin image right side edge that face determines, more than picture width
25%.
Illustrate, for other targets it should be noted that the present embodiment shines for veritification of putting on record
Picture, needs basis that the specific requirement of this picture delimited appointment scope.
S3, determine region to be detected according to region, location, it is determined that whether region to be detected meets and set
Requirement, removes region to be detected and is unsatisfactory for setting the picture required.
The region to be detected of the present embodiment includes left side detection zone, right side detection zone, the detection of left side shoulder
District and right shoulder detection zone.
The present embodiment is as follows for the determination method of left side detection zone, right side detection zone:
The distance set to left by human face region, using the region after translation as left side detection zone;
The distance set to right translation by human face region, using the region after translation as right side detection zone.
Specifically, as illustrated in fig. 2, it is assumed that the width of face rectangular area is Aw, left side detection zone,
Right side detection zone is defined as the left and right translation A of face rectangular areaw* 0.6 times of distance, to ensure to avoid
Face side projection and ear hair et al. soma is disturbed.
After determining left side detection zone, right side detection zone, respectively it is judged:
For left side detection zone, taking the test point setting quantity at left side detection zone, described test point will
Point cuttings such as left side detection zone, for any monitoring point, if all in Blue color domain, then judge full
Foot.The present embodiment takes 16 test points, if as it is shown on figure 3,16 test points are all at blue color
In territory, it is determined that meet.
For right side detection zone, the pixel of detection zone on the right side of traversal, if Blue color domain pixel is more than setting
Fixed first threshold, and Blue color domain with white gamut pixels is added more than setting Second Threshold, then
Judge to meet.The present embodiment first threshold is 70%, and Second Threshold is 95%, i.e. Blue color domain pixel
More than 70%, and Blue color domain+white gamut pixels is added and judges to meet more than total pixel 95%.
Further, the region to be detected of the present embodiment also includes left side shoulder detection zone and right shoulder
Detection zone, to prevent in picture shoulder outside picture.Wherein left side shoulder detection zone, at picture
The lower left corner, height is the 25% of picture height, and width is the 10% of picture width;Right shoulder detects
District, in the lower right corner of picture, height is the 25% of picture height, and width is the 10% of picture width.
And for left side shoulder detection zone and the judgement of right shoulder detection zone, the most respectively with left side detection zone,
Right side detection zone is identical, i.e. left side shoulder detection zone uses and takes the test point setting quantity, if all existed
In Blue color domain, then judge to meet, and right shoulder detection zone uses and travels through all pixels, if blue
Color gamut pixels is more than the first threshold set, and Blue color domain is added more than setting with white gamut pixels
Fixed Second Threshold, then judge to meet.
In sum, the present embodiment has following 2 points for the criterion veritifying photograph of putting on record:
1) face rectangular area is within the specified range;
2) left side detection zone, right side detection zone, left side shoulder detection zone, right shoulder detection zone is all
Judgement is passed through.
Meet above 2 points, it is determined that picture passes through, enter next step examination & verification;Otherwise it is considered as defective,
Directly return.
It should be noted that for different types of picture, it is fixed to select according to actual application places
Region, position and region to be detected, specify scope and the position in district to be detected, all for region, location
Selecting according to actual demand, correspondingly its standard judged is the most different, by the side of the present invention
Method can complete automatic screening.The present embodiment is only illustrated as a example by veritification of putting on record is shone, the party
Method can also be applied to other certificate photos, or the screening of other pictures.
The device of the present embodiment a kind of automatic screening picture as shown in Figure 4, including:
Region, location screening module, for determining the region, location in Target Photo, removes region, location
The picture of position deviation;
Region to be detected screening module, for determining region to be detected according to region, location, it is determined that to be checked
Survey whether region meets setting requirement, remove region to be detected and be unsatisfactory for setting the picture required.
Further, the device of the present embodiment also includes color treatments module, is used for carrying out picture color
Process, reduce the pixel interference that background causes.
Owing to the present embodiment is as a example by Target Photo shines for veritification of putting on record, veritification of putting on record is carried on the back according to for blueness
Scape, color treatments module row picture color processes, and reduces the pixel interference that background causes, performs as follows
Operation:
By described veritification of putting on record according to splitting into R, tri-Color Channel layers of G, B;
By R, the G Color Channel lamination of fractionation also, B color channel layer is removed.
Positioning region behaviour face region in the present embodiment, screening module in region to be detected is according to region, location
Determine region to be detected, perform to operate as follows:
The distance set to left by human face region, using the region after translation as left side detection zone;
The distance set to right translation by human face region, using the region after translation as right side detection zone.
After determining left side detection zone and right side detection zone, region to be detected screening module judges to be checked
Survey whether region meets setting requirement, the following operation of execution:
Take the test point setting quantity at left side detection zone, left side detection zone etc. is cut by described test point
Cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of detection zone on the right side of traversal, if Blue color domain pixel is more than the first threshold set, and
Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge to meet.
The present embodiment region to be detected also include being located at picture specify region left side shoulder detection zone and
Right shoulder detection zone.
Then, region to be detected screening module judges whether region to be detected meets and sets requirement, performs such as
Lower operation:
Take the test point setting quantity at left side shoulder detection zone, described test point is by left side detection zone etc.
Cut and cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of traversal right shoulder detection zone, if Blue color domain pixel is more than the first threshold set
It is worth, and Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge satisfied.
Above example is only limited in order to technical scheme to be described, is not carrying on the back
In the case of present invention spirit and essence thereof, those of ordinary skill in the art work as can be according to the present invention
Make various corresponding change and deformation, but these change accordingly and deformation all should belong to institute of the present invention
Attached scope of the claims.
Claims (14)
1. the method for an automatic screening picture, it is characterised in that including:
Determine the region, location in Target Photo, remove the picture of region, location position deviation;
Region to be detected is determined, it is determined that whether region to be detected meets sets requirement according to region, location,
Remove region to be detected to be unsatisfactory for setting the picture required.
Method the most according to claim 1, it is characterised in that described method is determining target
Before region, location in picture, further comprise the steps of:
Carry out picture color process, reduce the pixel interference that background causes.
Method the most according to claim 2, it is characterised in that described Target Photo is for putting on record
Veritify photograph, described in veritification of putting on record according to being blue background, described in carry out picture color process, reduce background
The pixel interference caused, including step:
By described veritification of putting on record according to splitting into R, tri-Color Channel layers of G, B;
By R, the G Color Channel lamination of fractionation also, B color channel layer is removed.
Method the most according to claim 1, it is characterised in that described Target Photo is for putting on record
Veritify photograph, described in put on record veritification according to be blue background, region behaviour face region, described location, described
Region to be detected is determined, including step according to region, location:
The distance set to left by human face region, using the region after translation as left side detection zone;
The distance set to right translation by human face region, using the region after translation as right side detection zone.
Method the most according to claim 4, it is characterised in that described judgement region to be detected
Whether meet and set requirement, including step:
Take the test point setting quantity at left side detection zone, left side detection zone etc. is cut by described test point
Cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of detection zone on the right side of traversal, if Blue color domain pixel is more than the first threshold set, and
Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge to meet.
Method the most according to claim 4, it is characterised in that described region to be detected is also wrapped
Include left side shoulder detection zone and the right shoulder detection zone being located at picture appointment region.
Method the most according to claim 6, it is characterised in that described judgement region to be detected
Whether meet the requirement set, including step:
Take the test point setting quantity at left side shoulder detection zone, described test point is by left side detection zone etc.
Cut and cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of traversal right shoulder detection zone, if Blue color domain pixel is more than the first threshold set
It is worth, and Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge satisfied.
8. the device of an automatic screening picture, it is characterised in that including:
Region, location screening module, for determining the region, location in Target Photo, removes region, location
The picture of position deviation;
Region to be detected screening module, for determining region to be detected according to region, location, it is determined that to be checked
Survey whether region meets setting requirement, remove region to be detected and be unsatisfactory for setting the picture required.
Device the most according to claim 8, it is characterised in that described device also includes color
Processing module, is used for carrying out picture color process, reduces the pixel interference that background causes.
Device the most according to claim 9, it is characterised in that described Target Photo is for putting on record
Veritify photograph, described in put on record veritification according to be blue background, described color treatments module row picture color process,
Reduce the pixel interference that background causes, perform to operate as follows:
By described veritification of putting on record according to splitting into R, tri-Color Channel layers of G, B;
By R, the G Color Channel lamination of fractionation also, B color channel layer is removed.
11. devices according to claim 8, it is characterised in that described Target Photo is for putting on record
Veritify photograph, described in put on record veritification according to be blue background, region behaviour face region, described location, described in treat
Screening module in detection region determines region to be detected according to region, location, the following operation of execution:
The distance set to left by human face region, using the region after translation as left side detection zone;
The distance set to right translation by human face region, using the region after translation as right side detection zone.
12. devices according to claim 11, it is characterised in that described region to be detected is sieved
Modeling block judges whether region to be detected meets and sets requirement, the following operation of execution:
Take the test point setting quantity at left side detection zone, left side detection zone etc. is cut by described test point
Cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of detection zone on the right side of traversal, if Blue color domain pixel is more than the first threshold set, and
Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge to meet.
13. devices according to claim 11, it is characterised in that described region to be detected is also
Including the left side shoulder detection zone and the right shoulder detection zone that are located at picture appointment region.
14. devices according to claim 13, it is characterised in that described region to be detected is sieved
Modeling block judges whether region to be detected meets and sets requirement, the following operation of execution:
Take the test point setting quantity at left side shoulder detection zone, described test point is by left side detection zone etc.
Cut and cut, for any monitoring point, if all in Blue color domain, then judge to meet;
The pixel of traversal right shoulder detection zone, if Blue color domain pixel is more than the first threshold set
It is worth, and Blue color domain is added more than the Second Threshold set with white gamut pixels, then judge satisfied.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510394966.2A CN106326824B (en) | 2015-07-02 | 2015-07-02 | A kind of method and device of automatic screening picture |
PCT/CN2016/086866 WO2017000827A1 (en) | 2015-07-02 | 2016-06-23 | Method and device for automatically screening pictures |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510394966.2A CN106326824B (en) | 2015-07-02 | 2015-07-02 | A kind of method and device of automatic screening picture |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106326824A true CN106326824A (en) | 2017-01-11 |
CN106326824B CN106326824B (en) | 2019-04-19 |
Family
ID=57607893
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510394966.2A Active CN106326824B (en) | 2015-07-02 | 2015-07-02 | A kind of method and device of automatic screening picture |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106326824B (en) |
WO (1) | WO2017000827A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107424128A (en) * | 2017-06-20 | 2017-12-01 | 南京泰立瑞信息科技有限公司 | A kind of method and system that picture rich in detail frame is extracted from the fundus video recorded |
CN110473263A (en) * | 2018-05-09 | 2019-11-19 | 北京京东尚科信息技术有限公司 | The checking method of picture, device and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101615241A (en) * | 2008-06-24 | 2009-12-30 | 上海银晨智能识别科技有限公司 | A kind of screening technique of certificate photograph |
US7925884B2 (en) * | 2008-06-30 | 2011-04-12 | International Business Machines Corporation | Fraud detection in seals |
CN103345622A (en) * | 2013-07-09 | 2013-10-09 | 浙江省公安厅居民身份证制作中心 | Method for controlling quality of character pictures on second-generation identification cards |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101650782A (en) * | 2009-09-16 | 2010-02-17 | 上海电力学院 | Method for extracting front human face outline based on complexion model and shape constraining |
-
2015
- 2015-07-02 CN CN201510394966.2A patent/CN106326824B/en active Active
-
2016
- 2016-06-23 WO PCT/CN2016/086866 patent/WO2017000827A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101615241A (en) * | 2008-06-24 | 2009-12-30 | 上海银晨智能识别科技有限公司 | A kind of screening technique of certificate photograph |
US7925884B2 (en) * | 2008-06-30 | 2011-04-12 | International Business Machines Corporation | Fraud detection in seals |
CN103345622A (en) * | 2013-07-09 | 2013-10-09 | 浙江省公安厅居民身份证制作中心 | Method for controlling quality of character pictures on second-generation identification cards |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107424128A (en) * | 2017-06-20 | 2017-12-01 | 南京泰立瑞信息科技有限公司 | A kind of method and system that picture rich in detail frame is extracted from the fundus video recorded |
CN107424128B (en) * | 2017-06-20 | 2020-12-29 | 南京泰立瑞信息科技有限公司 | Method and system for extracting clear image frame from recorded eye fundus video |
CN110473263A (en) * | 2018-05-09 | 2019-11-19 | 北京京东尚科信息技术有限公司 | The checking method of picture, device and system |
Also Published As
Publication number | Publication date |
---|---|
WO2017000827A1 (en) | 2017-01-05 |
CN106326824B (en) | 2019-04-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20230039805A1 (en) | Method and device for testing product quality | |
CN105913093B (en) | A kind of template matching method for Text region processing | |
CN105808416B (en) | A kind of automated testing method and system of man-machine figure interactive interface | |
US20140050387A1 (en) | System and Method for Machine Vision Inspection | |
JP2018198053A5 (en) | ||
TWI709085B (en) | Method, device, computer readable storage medium and computing equipment for damage segmentation of vehicle damage image | |
CN109146873B (en) | Learning-based intelligent detection method and device for defects of display screen | |
CN108896545B (en) | Gluing detection method and device and computer readable storage medium | |
CN105069453A (en) | Image correction method and apparatus | |
CN104458748A (en) | Aluminum profile surface defect detecting method based on machine vision | |
CN114494257B (en) | Gluing detection method, device, equipment and storage medium | |
EP3779874A3 (en) | System and method for automated surface assessment | |
CN113870267B (en) | Defect detection method, defect detection device, computer equipment and readable storage medium | |
US20190272627A1 (en) | Automatically generating image datasets for use in image recognition and detection | |
CN106469300B (en) | A kind of color spot detection recognition method | |
CN113096060B (en) | Positioning method and device for abnormal color lamp beads and storage medium | |
CN105844281A (en) | Automatic form set parameter acquiring system and method | |
CN112967221A (en) | Shield constructs section of jurisdiction production and assembles information management system | |
CN106326824A (en) | Method and device for automatically screening pictures | |
CN107346417A (en) | Method for detecting human face and device | |
CN115147362A (en) | Detection method, detection device and detection system for display panel | |
CN105678301A (en) | Method, system and device for automatically identifying and segmenting text image | |
CN107948392B (en) | Terminal and display fault processing method | |
CN103605973A (en) | Image character detection and identification method | |
CN101398462B (en) | Plaster bonding wire automatic detection system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |