CN107403155A - A kind of rapid classification sorting technique and device - Google Patents
A kind of rapid classification sorting technique and device Download PDFInfo
- Publication number
- CN107403155A CN107403155A CN201710618685.XA CN201710618685A CN107403155A CN 107403155 A CN107403155 A CN 107403155A CN 201710618685 A CN201710618685 A CN 201710618685A CN 107403155 A CN107403155 A CN 107403155A
- Authority
- CN
- China
- Prior art keywords
- face
- image
- facial image
- recognition
- deep learning
- 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
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of rapid classification sorting technique and device.Methods described includes obtaining at least image that collecting device uploads；Face datection is carried out to an at least image, determines the target facial image in described image；The face recognition result of the target facial image will be contrasted with the recognition of face deep learning model pre-established, and the comparing result of acquisition is classified, according to classification results, control the display of display device.Rapid classification sorting technique provided in an embodiment of the present invention and device, field of safety check is incorporated into applied to safe examination system, and by face recognition technology, improves safety check efficiency, the precision and efficiency that reduce safety check labor intensity, lift safety check management.
Description
Technical field
The present invention relates to safety check technical field, and in particular to a kind of rapid classification sorting technique and device.
Background technology
The occasion that detector gate is applied in daily life is more and more, for example, applied to subway, civil aviaton, station,
The large public activity place such as harbour, official mission, museum, stadium, the application in incorporated business is also more next in recent years
More popularize.
By taking subway safety check as an example, subway safety check passenger safety check is mainly that the mode for leaning on " detector gate+hand inspection " carries out safety check, but
Be as subway construction constantly promotes, increasing people by the subway, volume of the flow of passengers sustainable growth.Bus's flow has become subway
Normality, bring huge challenge to subway safety check work of searching for dangerous goods.
Through measuring and calculating, average " inspection people " speed of subway safety check at present is about 15 seconds/people, one, a machine (single-view X-ray machine+gold
Category detector gate) safety check ability be about 300 people/hour, it is about 22.5 minutes that peak period passenger, which averagely waits the inspection time,.In existing skill
In art using safety inspector to enter the station everyone will carry out hand inspection, in the case of more than passenger, this safety check speed is just very slow,
Efficiency is very low, and passenger enters the station needs waiting for a long time.Closely touch safety check has not only been invaded passenger's dignity and caused and multiplied
Visitor is discontented, has also had a strong impact on real-time percent of pass, causes personnel's congestion in station, forms new potential safety hazard, while aggravate
The work load of security staff.How quick security check, strengthen the safety check in the case of large passenger flow and search for dangerous goods work, it has also become subway safety check
The problem urgently improved and solved.
Therefore, a kind of method how is proposed, can be in the case where the volume of the flow of passengers be big, moreover it is possible to safety check is quickly and safely carried out,
As urgent problem to be solved.
The content of the invention
For in the prior art the defects of, the embodiments of the invention provide a kind of rapid classification sorting technique and device.
On the one hand, the embodiment of the present invention provides a kind of rapid classification sorting technique, applied to safe examination system, including：
Obtain at least image that collecting device uploads；
Face datection is carried out to an at least image, determines the target facial image in described image；
By to the face recognition result of the target facial image and the recognition of face deep learning model pre-established
Contrasted, and the comparing result of acquisition is classified, according to classification results, control the display of display device.
On the other hand, the embodiment of the present invention provides a kind of rapid classification sorter, applied to safe examination system, including：
Acquiring unit, for obtaining an at least image for collecting device upload；
Determining unit, for carrying out Face datection to an at least image, determine the target face in described image
Image；
Judging unit, for by the face recognition result of the target facial image and the recognition of face pre-established
Deep learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls display device
Display.
Rapid classification sorting technique provided in an embodiment of the present invention and device, know applied to safe examination system, and by face
Other technology is incorporated into field of safety check, is known by the face recognition result to the target facial image and the face pre-established
Other deep learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls display device
Display, so as to which which kind of security action reminded security staff to be adapted to use, improve safety check efficiency, reduce safety check labor intensity, carry
Rise the precision and efficiency of safety check management.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are this hairs
Some bright embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can be with root
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the schematic flow sheet of rapid classification sorting technique provided in an embodiment of the present invention；
Fig. 2 is the structural representation of rapid classification sorter provided in an embodiment of the present invention.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
Part of the embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is the schematic flow sheet of quick security check method provided in an embodiment of the present invention, as shown in figure 1, methods described bag
Include：
S101, obtain at least image that collecting device uploads；
When there is personnel to pass through detector gate, it can be gathered installed in the collecting device of the diverse location of detector gate and pass through personnel
An at least image, and collecting device can be uploaded onto the server at least image collected by transmission network.Wherein,
Collecting device can be video camera etc., and collecting device is per second to capture more than ten frames, and specific collection frame number can be according to the need of user
Seek sets itself.
S102, Face datection is carried out to an at least image, determine the target facial image in described image；
After server receives at least image that the collecting device uploads, the face in an at least image can be carried out
Detect, there may be a facial image in each image, it is also possible to there are multiple facial images, determined from multiple facial images
Target facial image.
S103, by the face recognition result of the target facial image and the recognition of face deep learning that pre-establishes
Model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the display of display device.
Target facial image in server determines described image, face knowledge is carried out to the target facial image
Not, the facial recognition data of the target facial image is obtained, server is by obtained facial recognition data with pre-establishing
Recognition of face learning model contrasted, obtain different comparing results, obtained comparing result classified, according to not
With classification results control the display of display device.
For example, when obtained human face recognition model is contrasted with the recognition of face learning model pre-established, may
Certain met in recognition of face learning model is a kind of, it is also possible to does not meet face learning model, can thus obtain different pairs
Than result, and the lamp that different comparing results will correspond to the different colours of display device is shown, while represents different safety checks
Grade.
Rapid classification sorting technique provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Alternatively, the recognition of face deep learning model is established using the steps：
Gather the different types of face picture for training；
Multilayer feature is carried out to the face picture using multistage convolutional neural networks algorithm to extract to obtain training data；
The training data is trained using deep learning method to obtain the recognition of face deep learning model.
On the basis of above-described embodiment, it is necessary first to establish recognition of face deep learning model, specific establishment step
As described below：
The face picture without type for training is gathered, for example, the human face photo of the employee of collection company A, gathers B
The human face photo of the employee of company or resident's photo of some cell or enterprises and institutions of Party and government offices work people
The photo or process statistics of member calculates, and calculates in certain time period, in the face that same place often occurs
Image etc., and these photos are stored according to respective category classification；To collect be used for train different classifications it is every
One photo is handled, using multistage convolutional neural networks algorithm carry out multilayer feature extract to obtain each photo it is corresponding
Training data；The training data calculated is trained using deep learning method, so as to obtain the people
Face identifies deep learning model.
Rapid classification sorting technique provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Alternatively, it is described that Face datection is carried out to an at least image, determine the target face figure in described image
As being specially：Face datection is carried out to an at least image, the facial image of certain predetermined rule will be met, is defined as described
Target facial image in image.
On the basis of above-described embodiment, server is examined after an at least image is received to the face in image
Survey, it is a kind of it might be that at a time, during to being taken pictures by the personnel of detector gate, it may also photograph and passing through
Personnel bystander, so a just more than facial image in photo in this case；Alternatively possible situation is, due to
Diverse location on detector gate is installed with collecting device, is from different angles so for one to be ready the personnel that pass through
Degree shooting, the display of face is just different in the photo of each angle shot.
It is above-mentioned it is possible in the case of, server will carry out Face datection to face different in image, according to one
Default rule, server are judged described face, if meeting this default rule, decide that this face is
Target facial image, multi-faceted face can be shot, detected, so as to improve accuracy.
Alternatively, the certain predetermined rule is the maximum facial image of the weighted value of the facial image.
On the basis of above-described embodiment, server detects according to certain preset rules to the face in image,
If meet certain predetermined rule, it is determined that be target facial image, wherein described certain predetermined rule is the facial image
The maximum facial image of weighted value.
For example, the weight of the facial image, the size of the area occupied of face, or face can be referred to
Definition etc..If had in the image for having multiple faces, the weighted value of at least two faces is identical, and server can be with
Machine determines that one is target facial image.
Rapid classification sorting technique provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Alternatively, the recognition of face deep learning model pre-established is：The mould that the face picture of user's collection is established
The model of suspect or the model of a suspect after type and public security system networking.
, it is necessary to pre-establish recognition of face deep learning model on the basis of above-described embodiment, and this study mould
Type is built upon on the basis of different sample sets, for example, collection sample can be some company, some cell personnel mould
The photo sample of type or some suspects after being networked with public security system or once carried
Photo of some a suspects of certain of dangerous goods etc..
Rapid classification sorting technique provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Alternatively, it is described according to the classification results, the display of display device is controlled, is specially：
If the face recognition result of the target facial image meets the recognition of face deep learning model pre-established
In with public security system network after suspect model, then the display device shown with first state；
If the face recognition result of the target facial image meets the recognition of face deep learning model pre-established
In a suspect model, then the display device shown with first state；
If the face recognition result of the target facial image meets the recognition of face deep learning model pre-established
In user collection face picture establish model, then the display device shown with the second state；
If the face recognition result of the target facial image does not meet the recognition of face deep learning mould pre-established
Type, then the display device shown with the third state.
On the basis of above-described embodiment, server is according to by the face recognition result of the target facial image and in advance
The recognition of face deep learning model first established is contrasted, and the comparing result of acquisition is classified, according to classification
As a result, the display of display device is controlled.
Display device in the embodiment of the present invention can be display lamp or display screen, or other can be used
Come the equipment shown.
If the face recognition result of the target facial image meets the recognition of face deep learning model pre-established
In with public security system network after suspect model, then the display device shown with first state, remind
The rank of safety inspector's safety check is the superlative degree, and safety inspector can be performed corresponding safety check according to the state of display, can be examined with hand, also may be used
So that the personnel passed through are carried out with safety check using Terahertz human body safety check instrument, the time of general safety check is 20s or so；
If the face recognition result of the target facial image meets the recognition of face deep learning model pre-established
In a suspect model, then the display device is shown with first state, reminds the rank of safety inspector's safety check as most
Advanced, safety inspector can perform corresponding safety check according to the state of display, can be examined with hand, can also use Terahertz human body safety check
Instrument to carry out the personnel passed through safety check, and the time of general safety check is 20s or so；
If the face recognition result of the target facial image meets the recognition of face deep learning model pre-established
In user collection face picture establish model, then the display device shown with the second state, remind safety inspector
The rank of safety check is quick level, and safety inspector can be performed corresponding safety check according to the state of display, can be examined with hand, can also used
Terahertz human body safety check instrument to carry out the personnel passed through safety check, and the time of general safety check is 7s or so, wherein gather here
Human face photo can be some companies, the resident of some cell, the photo of enterprises and institutions of Party and government offices staff or process
Algorithm calculates, and in certain time period, be in the same localities the facial image often occurred；
If the face recognition result of the target facial image does not meet the recognition of face deep learning mould pre-established
Type, then the display device shown with the third state, remind safety inspector's safety check rank be regular grade, safety inspector can basis
The state of display performs corresponding safety check, can be examined with hand, can also using Terahertz human body safety check instrument come to the personnel passed through
Safety check is carried out, the time of general safety check is 15s or so.In this case, it is this not meet recognition of face deep learning model
Target face may be exactly some tourisms personnel etc..
Rapid classification sorting technique provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Fig. 2 is the structural representation of rapid classification sorter provided in an embodiment of the present invention, as shown in Fig. 2 the dress
Put including acquiring unit 10, determining unit 11 and judging unit 12, wherein：
Acquiring unit 10 is used at least image for obtaining collecting device upload；Determining unit 11 is used for described at least one
Image carries out Face datection, determines the target facial image in described image；Judging unit 12 is used for will be to the target person
The face recognition result of face image is contrasted with the recognition of face deep learning model pre-established, and by pair of acquisition
Classified than result, according to classification results, control the display of display device.
At least one collecting device is installed in the diverse location of detector gate, and display is also equipped with detector gate and is set
Standby, when there is personnel to pass through detector gate, collecting device can gather at least image by personnel, and collecting device can pass through
At least image collected is uploaded to acquiring unit 10 by transmission network；Now, acquiring unit 10 is got on collecting device
At least image passed, determining unit 11 can detect to the face in an at least image, may there is one in each image
Facial image, it is also possible to have multiple facial images, target facial image is determined from multiple facial images；It is described determining
Target facial image in image, judging unit 12 carry out recognition of face to the target facial image, obtain the target person
The facial recognition data of face image, server is by obtained facial recognition data and the recognition of face learning model that pre-establishes
Contrasted, obtain different comparing results, obtained comparing result is classified, controlled according to different classification results
The display of display device.
For example, when obtained human face recognition model is contrasted with the recognition of face learning model pre-established, may
Certain met in recognition of face learning model is a kind of, it is also possible to does not meet face learning model, can thus obtain different pairs
Than result, and the lamp that different comparing results will correspond to the different colours of display device is shown, while represents different safety checks
Grade.
Rapid classification sorter provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Alternatively, on the basis of above-described embodiment, described device includes acquiring unit 10, determining unit 11, judges list
Member 12 and recognition of face deep learning model establish unit, wherein to establish unit specific for the recognition of face deep learning model
For：
Acquisition module, for gathering the different types of face picture for training；
Extraction module, extracted for carrying out multilayer feature using multistage convolutional neural networks algorithm to the face picture
To training data；
Training module, for training to obtain the recognition of face depth to the training data using deep learning method
Practise model.
On the basis of above-described embodiment, acquiring unit 10, determining unit 11, judging unit 12 in the present embodiment with it is upper
The performance and structure of unit in individual embodiment are identical, will not be repeated here.
Wherein, described device also establishes unit including recognition of face deep learning model, and specifically, collection model collection is used
In the different types of face picture of training, extraction model is passed through to these face pictures using multistage convolutional neural networks algorithm
Multilayer feature is carried out to extract to obtain training data, the training that training pattern is extracted to obtain using deep learning method to extraction model
Data are trained, and obtain the recognition of face deep learning model.
Rapid classification sorter provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Alternatively, the determining unit 11 is specially：Face datection is carried out to an at least image, will be met certain pre-
If the facial image of rule, the target facial image being defined as in described image.
On the basis of above-described embodiment, determining unit 11 is entered after an at least image is received to the face in image
Row detection, it is a kind of it might be that at a time, during to being taken pictures by the personnel of detector gate, may also it photograph just
By personnel bystander, so a just more than facial image in photo in this case；Alternatively possible situation is,
It is from difference so for one to be ready the personnel that pass through because the diverse location on detector gate is installed with collecting device
Angle shot, the display of face is just different in the photo of each angle shot.
It is above-mentioned it is possible in the case of, determining unit 11 will carry out Face datection to face different in image, according to
One default rule, determining unit 11 are judged described face, if meeting this default rule, decide that this
Individual face is target facial image, multi-faceted face can be shot, detected, so as to improve accuracy.
Alternatively, the certain predetermined rule is the maximum facial image of the weighted value of the facial image.
On the basis of above-described embodiment, server detects according to certain preset rules to the face in image,
If meet certain predetermined rule, it is determined that be target facial image, wherein described certain predetermined rule is the facial image
The maximum facial image of weighted value.
For example, the weight of the facial image, the size of the area occupied of face, or face can be referred to
Definition of image etc..If had in the image for having multiple faces, the weighted value of at least two faces is identical, and server can
One is determined as target facial image with random.
Rapid classification sorter provided in an embodiment of the present invention, applied to safe examination system, and by face recognition technology
Field of safety check is incorporated into, passes through the face recognition result to the target facial image and the recognition of face depth pre-established
Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device
Show which kind of security action used so as to remind security staff to be adapted to, improve safety check efficiency, reduce safety check labor intensity, lifting peace
Examine the precision and efficiency of management.
Device and system embodiment described above is only schematical, wherein described be used as separating component explanation
Unit can be or may not be physically separate, can be as the part that unit is shown or may not be
Physical location, you can with positioned at a place, or can also be distributed on multiple NEs.Can be according to the actual needs
Some or all of module therein is selected to realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying
In the case of performing creative labour, you can to understand and implement.
Claims (10)
- A kind of 1. rapid classification sorting technique, applied to safe examination system, it is characterised in that including：Obtain at least image that collecting device uploads；Face datection is carried out to an at least image, determines the target facial image in described image；The face recognition result of the target facial image and the recognition of face deep learning model pre-established will be carried out Contrast, and the comparing result of acquisition is classified, according to classification results, control the display of display device.
- 2. according to the method for claim 1, it is characterised in that the recognition of face deep learning model uses following step It is rapid to establish：Gather the different types of face picture for training；Multilayer feature is carried out to the face picture using multistage convolutional neural networks algorithm to extract to obtain training data；The training data is trained using deep learning method to obtain the recognition of face deep learning model.
- 3. according to the method for claim 1, it is characterised in that it is described that Face datection is carried out to an at least image, really The target facial image made in described image is specially：Face datection is carried out to an at least image, the facial image of certain predetermined rule will be met, be defined as described image In target facial image.
- 4. according to the method for claim 3, it is characterised in that the certain predetermined rule is the weight of the facial image It is worth maximum facial image.
- 5. according to the method for claim 1, it is characterised in that the recognition of face deep learning model pre-established is： The model of suspect or the mould of a suspect after the model of the face picture foundation of user's collection and public security system networking Type.
- 6. according to the method for claim 1, it is characterised in that it is described according to the classification results, control display device It has been shown that, it is specially：If the face recognition result of the target facial image meets in the recognition of face deep learning model pre-established With public security system network after suspect model, then the display device shown with first state；If the face recognition result of the target facial image meets in the recognition of face deep learning model pre-established The model of a suspect, then the display device shown with first state；If the face recognition result of the target facial image meets in the recognition of face deep learning model pre-established User collection face picture establish model, then the display device shown with the second state；If the face recognition result of the target facial image does not meet the recognition of face deep learning model pre-established, The display device is shown with the third state.
- A kind of 7. rapid classification sorter, applied to safe examination system, it is characterised in that including：Acquiring unit, for obtaining an at least image for collecting device upload；Determining unit, for carrying out Face datection to an at least image, determine the target facial image in described image；Judging unit, for by the face recognition result of the target facial image and the recognition of face depth pre-established Learning model is contrasted, and the comparing result of acquisition is classified, and according to classification results, controls the aobvious of display device Show.
- 8. device according to claim 7, it is characterised in that also established including the recognition of face deep learning model single Member, specifically include：Acquisition module, for gathering the different types of face picture for training；Extraction module, instructed for being extracted to the face picture using multistage convolutional neural networks algorithm progress multilayer feature Practice data；Training module, for training to obtain the recognition of face deep learning mould to the training data using deep learning method Type.
- 9. device according to claim 7, it is characterised in that the determining unit is specially：Face datection is carried out to an at least image, the facial image of certain predetermined rule will be met, be defined as described image In target facial image.
- 10. device according to claim 9, it is characterised in that the certain predetermined rule is the power of the facial image The maximum facial image of weight values.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710618685.XA CN107403155A (en) | 2017-07-26 | 2017-07-26 | A kind of rapid classification sorting technique and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710618685.XA CN107403155A (en) | 2017-07-26 | 2017-07-26 | A kind of rapid classification sorting technique and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107403155A true CN107403155A (en) | 2017-11-28 |
Family
ID=60401523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710618685.XA Pending CN107403155A (en) | 2017-07-26 | 2017-07-26 | A kind of rapid classification sorting technique and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107403155A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108109348A (en) * | 2017-12-11 | 2018-06-01 | 日立楼宇技术（广州）有限公司 | Classifying alarm method and device |
| CN108171123A (en) * | 2017-12-11 | 2018-06-15 | 日立楼宇技术（广州）有限公司 | Dangerous user's identification and the method and apparatus of alarm |
| CN108197616A (en) * | 2018-03-09 | 2018-06-22 | 广州佳都数据服务有限公司 | By can real name registration complete subway classify safety check method and system |
| CN109100806A (en) * | 2018-07-31 | 2018-12-28 | 国政通科技有限公司 | A kind of hierarchical detection method and device |
| CN109896374A (en) * | 2017-12-11 | 2019-06-18 | 日立楼宇技术（广州）有限公司 | Elevator screen methods of exhibiting and system |
| CN110244370A (en) * | 2018-08-03 | 2019-09-17 | 浙江大华技术股份有限公司 | A kind of safety check control method, apparatus and system |
| CN110850493A (en) * | 2019-10-24 | 2020-02-28 | 武汉艾崴科技有限公司 | Linear array high-speed security inspection machine for visual image judgment of scanning imaging machine |
| CN111103629A (en) * | 2018-10-25 | 2020-05-05 | 杭州海康威视数字技术股份有限公司 | Target detection method and device, NVR (network video recorder) equipment and security check system |
| CN112485842A (en) * | 2019-09-12 | 2021-03-12 | 杭州海康威视数字技术股份有限公司 | Security inspection system |
| CN117218081A (en) * | 2023-09-13 | 2023-12-12 | 北京声迅电子股份有限公司 | Express item security inspection method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226664A (en) * | 2008-02-02 | 2008-07-23 | 北京海鑫科金高科技股份有限公司 | Intelligent monitoring system and method for self-help bank and ATM |
| CN101534425A (en) * | 2009-04-10 | 2009-09-16 | 上海伯里春贸易有限公司 | System for acquiring, comparing, identifying and monitoring portrait at station, airdrome and metro, as well as method thereof |
| CN102201061A (en) * | 2011-06-24 | 2011-09-28 | 常州锐驰电子科技有限公司 | Intelligent safety monitoring system and method based on multilevel filtering face recognition |
| CN103150553A (en) * | 2013-02-06 | 2013-06-12 | 北京中科虹霸科技有限公司 | Mobile terminal and method for realizing multi-mode identity characteristic recognition |
| CN106355170A (en) * | 2016-11-22 | 2017-01-25 | Tcl集团股份有限公司 | Photo classifying method and device |
| CN106407369A (en) * | 2016-09-09 | 2017-02-15 | 华南理工大学 | Photo management method and system based on deep learning face recognition |
| CN106682650A (en) * | 2017-01-26 | 2017-05-17 | 北京中科神探科技有限公司 | Mobile terminal face recognition method and system based on technology of embedded deep learning |
-
2017
- 2017-07-26 CN CN201710618685.XA patent/CN107403155A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226664A (en) * | 2008-02-02 | 2008-07-23 | 北京海鑫科金高科技股份有限公司 | Intelligent monitoring system and method for self-help bank and ATM |
| CN101534425A (en) * | 2009-04-10 | 2009-09-16 | 上海伯里春贸易有限公司 | System for acquiring, comparing, identifying and monitoring portrait at station, airdrome and metro, as well as method thereof |
| CN102201061A (en) * | 2011-06-24 | 2011-09-28 | 常州锐驰电子科技有限公司 | Intelligent safety monitoring system and method based on multilevel filtering face recognition |
| CN103150553A (en) * | 2013-02-06 | 2013-06-12 | 北京中科虹霸科技有限公司 | Mobile terminal and method for realizing multi-mode identity characteristic recognition |
| CN106407369A (en) * | 2016-09-09 | 2017-02-15 | 华南理工大学 | Photo management method and system based on deep learning face recognition |
| CN106355170A (en) * | 2016-11-22 | 2017-01-25 | Tcl集团股份有限公司 | Photo classifying method and device |
| CN106682650A (en) * | 2017-01-26 | 2017-05-17 | 北京中科神探科技有限公司 | Mobile terminal face recognition method and system based on technology of embedded deep learning |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108109348A (en) * | 2017-12-11 | 2018-06-01 | 日立楼宇技术（广州）有限公司 | Classifying alarm method and device |
| CN108171123A (en) * | 2017-12-11 | 2018-06-15 | 日立楼宇技术（广州）有限公司 | Dangerous user's identification and the method and apparatus of alarm |
| CN109896374A (en) * | 2017-12-11 | 2019-06-18 | 日立楼宇技术（广州）有限公司 | Elevator screen methods of exhibiting and system |
| CN109896374B (en) * | 2017-12-11 | 2021-08-03 | 日立楼宇技术（广州）有限公司 | Elevator screen display method and system |
| CN108197616A (en) * | 2018-03-09 | 2018-06-22 | 广州佳都数据服务有限公司 | By can real name registration complete subway classify safety check method and system |
| CN109100806A (en) * | 2018-07-31 | 2018-12-28 | 国政通科技有限公司 | A kind of hierarchical detection method and device |
| CN110244370A (en) * | 2018-08-03 | 2019-09-17 | 浙江大华技术股份有限公司 | A kind of safety check control method, apparatus and system |
| CN111103629A (en) * | 2018-10-25 | 2020-05-05 | 杭州海康威视数字技术股份有限公司 | Target detection method and device, NVR (network video recorder) equipment and security check system |
| CN112485842A (en) * | 2019-09-12 | 2021-03-12 | 杭州海康威视数字技术股份有限公司 | Security inspection system |
| CN110850493A (en) * | 2019-10-24 | 2020-02-28 | 武汉艾崴科技有限公司 | Linear array high-speed security inspection machine for visual image judgment of scanning imaging machine |
| CN117218081A (en) * | 2023-09-13 | 2023-12-12 | 北京声迅电子股份有限公司 | Express item security inspection method |
| CN117218081B (en) * | 2023-09-13 | 2024-04-09 | 北京声迅电子股份有限公司 | Express item security inspection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107403155A (en) | A kind of rapid classification sorting technique and device | |
| CN105975959B (en) | Face characteristic neural network based extracts modeling, face identification method and device | |
| CN108154134A (en) | Internet live streaming pornographic image detection method based on depth convolutional neural networks | |
| CN105303193B (en) | A kind of passenger number statistical system based on single-frame images processing | |
| CN107007257B (en) | The automatic measure grading method and apparatus of the unnatural degree of face | |
| CN106529571B (en) | Multi-layer image feature extraction modeling, image-recognizing method and device based on neutral net | |
| CN109558840A (en) | A kind of biopsy method of Fusion Features | |
| CN107832802A (en) | Quality of human face image evaluation method and device based on face alignment | |
| CN107194396A (en) | Method for early warning is recognized based on the specific architecture against regulations in land resources video monitoring system | |
| CN105869085A (en) | Transcript inputting system and method for processing images | |
| CN110322438A (en) | The training method and automatic checkout system of the automatic detection model of mycobacterium tuberculosis | |
| CN106667506A (en) | Method and device for detecting lies on basis of electrodermal response and pupil change | |
| CN108564577A (en) | Solar cell segment grid defect inspection method based on convolutional neural networks | |
| CN207636791U (en) | A kind of large passenger flow quick security check door | |
| CN106485217B (en) | A kind of method and system of identification visit scenic spot stream of people saturation degree | |
| CN108710822A (en) | Personnel falling detection system based on infrared array sensor | |
| CN106709438A (en) | Method for collecting statistics of number of people based on video conference | |
| CN107330370A (en) | A kind of brow furrows motion detection method and device and vivo identification method and system | |
| Zhafran et al. | Computer vision system based for personal protective equipment detection, by using convolutional neural network | |
| CN109389146A (en) | Equipment state feedback method and system based on neural network image sorting algorithm | |
| CN110084987A (en) | A kind of foreign matter inspecting system and method towards rail traffic | |
| CN107798276A (en) | A kind of facial aging method and device of facial image | |
| CN107832730A (en) | Improve the method and face identification system of face recognition accuracy rate | |
| CN108363944A (en) | Recognition of face terminal is double to take the photograph method for anti-counterfeit, apparatus and system | |
| CN109034038A (en) | A kind of fire identification device based on multi-feature fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171128 |