CN107292293A - The method and device of Face datection - Google Patents
The method and device of Face datection Download PDFInfo
- Publication number
- CN107292293A CN107292293A CN201710618497.7A CN201710618497A CN107292293A CN 107292293 A CN107292293 A CN 107292293A CN 201710618497 A CN201710618497 A CN 201710618497A CN 107292293 A CN107292293 A CN 107292293A
- Authority
- CN
- China
- Prior art keywords
- face
- detector
- region
- candidate region
- image
- 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
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)
Abstract
The disclosure is directed to a kind of method and device of Face datection.The method of the Face datection includes：More than one face candidate region in image is obtained, more than one face candidate region is merged, combined region is obtained, convolutional calculation is carried out to combined region, convolution results are obtained, Face datection is carried out according to convolution results.The method and device of the Face datection of the disclosure, can reduce the area for the image-region for carrying out convolutional calculation, reduce the complexity and operand of convolutional calculation, improve the speed of convolutional calculation, accelerate Face datection.
Description
Technical field
This disclosure relates to technical field of image processing, more particularly to Face datection method and device.
Background technology
Face datection is the first step that face is analyzed and handled, and thus Face datection turns into image processing field
Hot issue, and the intelligent terminal for being equipped with Face datection function will significantly improve the interactive experience and usage experience of user.Phase
In the technology of pass, Face datection algorithm of the Face datection algorithm including traditional Face datection algorithm and based on deep learning.Compare
In traditional Face datection algorithm, the Face datection algorithm based on deep learning has more preferably Detection results, in various harshnesses
Testing conditions under can access preferable accuracy of detection.
Face datection algorithm based on deep learning uses convolution operation.Convolutional layer is a weight in deep neural network
The layer wanted, convolutional layer can realize local receptor field, and the number of parameter can be effectively reduced by local receptor field.Fig. 1 is
The schematic diagram of convolution operation in Caffe in correlation technique.As shown in figure 1, in Caffe, convolution operation is by convolution kernel
Matrix A is transformed to, input picture is transformed to matrix B, matrix A is multiplied with matrix B obtains Matrix C.In matrix A, M represents volume
The number of product core, K=k × k, k represents convolution kernel size.Often row i.e. in matrix A represents that a convolution kernel vector (is by two dimension
Convolution kernel be converted into it is one-dimensional), M rows indicate M convolution kernel.In matrix B, N=((image_h+2 × pad_h-
Kernel_h)/stride_h+1) × ((image_w+2 × pad_w-kernel_w)/stride_w+1), wherein, image_h
The height of input picture is represented, image_w represents the width of input picture, and pad_h represents the short transverse two in input picture
Side respectively increases pad_h unit length, and pad_w represents respectively to increase pad_w unit on the width both sides of input picture long
Degree, kernel_h represents the height of convolution kernel, and kernel_w represents the width of convolution kernel, and stride_h represents the cunning of short transverse
Dynamic step-length, stride_w represents the sliding step of width.Therefore, N represents the length and width product of output image size, is also volume
Product core slides the maximum characteristic that can be intercepted over an input image.K=k × k, represents in input to scheme using the frame of convolution kernel size
It is big as convolution kernel size as the intercepted size of data of upper slip.In Matrix C, Matrix C is that matrix A is multiplied with matrix B
Result, Matrix C is M × N matrix, wherein, often row represents an output image, and having M output image, (output is schemed
As number is equal to convolution kernel number).
In correlation technique, in the Face datection algorithm based on deep learning, convolution kernel is in the form of scanning window, in figure
Each position of picture calculate the characteristic value in current window weighted average and.Because each position of image will participate in meter
Calculate, so the computation complexity of the Face datection algorithm based on deep learning is too high, amount of calculation is larger.
The content of the invention
To overcome problem present in correlation technique, the disclosure provides a kind of method and device of Face datection.
According to the first aspect of the embodiment of the present disclosure there is provided a kind of method of Face datection, including：
Obtain more than one face candidate region in image；
More than one face candidate region is merged, combined region is obtained；
Convolutional calculation is carried out to the combined region, convolution results are obtained；
Face datection is carried out according to the convolution results.
In a kind of possible implementation, more than one face candidate region is merged, combined region is obtained, wrapped
Include：
More than one face candidate region is spliced into combined region.
In a kind of possible implementation, more than one face candidate region in image is obtained, including：
Waited using more than one face the detector acquisition image of the face under multiple different conditions suitable for detection
Favored area.
In a kind of possible implementation, the detector of the multiple face being applied under detection different condition includes
Below at least two：
Suitable for detecting the detector of positive face, the detector suitable for detecting side face, suitable for detecting the face being blocked
Detector, suitable for detection image brightness be more than first threshold under conditions of face detector and suitable for detection image
Brightness be less than Second Threshold under conditions of face detector, wherein, the Second Threshold be less than the first threshold.
According to the second aspect of the embodiment of the present disclosure there is provided a kind of device of Face datection, including：
Acquisition module, for obtaining more than one face candidate region in image；
Merging module, for more than one face candidate region to be merged, obtains combined region；
Computing module, for carrying out convolutional calculation to the combined region, obtains convolution results；
Detection module, for carrying out Face datection according to the convolution results.
In a kind of possible implementation, the merging module includes：
Concatenation module, for more than one face candidate region to be spliced into combined region.
In a kind of possible implementation, the acquisition module includes：
Face candidate region acquisition module, is obtained for the detector using multiple faces being applied under detection different condition
Take more than one face candidate region in image.
In a kind of possible implementation, the detector of the multiple face being applied under detection different condition includes
Below at least two：
Suitable for detecting the detector of positive face, the detector suitable for detecting side face, suitable for detecting the face being blocked
Detector, suitable for detection image brightness be more than first threshold under conditions of face detector and suitable for detection image
Brightness be less than Second Threshold under conditions of face detector, wherein, the Second Threshold be less than the first threshold.
According to the third aspect of the embodiment of the present disclosure there is provided a kind of device of Face datection, including：Processor；For depositing
Store up the memory of processor-executable instruction；Wherein, the processor is configured as realizing above-mentioned method when performing.
According to the fourth aspect of the embodiment of the present disclosure there is provided a kind of non-volatile computer readable storage medium storing program for executing, deposit thereon
Computer program instructions are contained, the computer program instructions realize above-mentioned method when being executed by processor.
The technical scheme provided by this disclosed embodiment can include the following benefits：The side of the Face datection of the disclosure
Method and device, by obtaining more than one face candidate region in image, more than one face candidate region is merged,
Combined region is obtained, convolutional calculation is carried out to combined region, convolution results are obtained, Face datection is carried out according to convolution results, by
This can reduce the area for the image-region for carrying out convolutional calculation, reduce the complexity and operand of convolutional calculation, improve convolution
The speed of calculating, accelerates Face datection.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary and explanatory, not
The disclosure can be limited.
Brief description of the drawings
Accompanying drawing herein is merged in specification and constitutes the part of this specification, shows the implementation for meeting the disclosure
Example, and be used to together with specification to explain the principle of the disclosure.
Fig. 1 is the schematic diagram of convolution operation in Caffe in correlation technique.
Fig. 2 is a kind of flow chart of the method for Face datection according to an exemplary embodiment.
Fig. 3 is the schematic diagram in the face candidate region according to an exemplary embodiment.
Fig. 4 is an a kind of schematical flow chart of the method for Face datection according to an exemplary embodiment.
Fig. 5 is a kind of block diagram of the device of Face datection according to an exemplary embodiment.
Fig. 6 is an a kind of schematical block diagram of the device of Face datection according to an exemplary embodiment.
Fig. 7 is a kind of block diagram of the device 800 of Face datection according to an exemplary embodiment.
Embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the disclosure.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects be described in detail in claims, the disclosure.
Fig. 2 is a kind of flow chart of the method for Face datection according to an exemplary embodiment.The Face datection
Method can apply to such as smart mobile phone, tablet personal computer, PC (Personal Computer, computer) or wearable device
Deng intelligent terminal, the disclosure is not restricted to this.As shown in Fig. 2 the method for the Face datection comprises the following steps.
In step s 201, more than one face candidate region in image is obtained.
Wherein, face candidate region can refer to one or more regions obtained from the face in detection image.
As an example of the implementation, in a face in detecting image every time, rectangle can be used
Frame chooses face, and the region that the rectangle frame is chosen is used as a face candidate region.Fig. 3 is according to an exemplary embodiment
The schematic diagram in the face candidate region shown.As shown in figure 3, three face candidate regions in image can be obtained.
It should be noted that although the region chosen using rectangle frame describes face candidate region as above as example,
It will be appreciated by those skilled in the art that the disclosure answers not limited to this.Those skilled in the art can be according to practical application scene spirit
Setting face candidate region living, the region for for example choosing circular frame is used as face candidate region.
In a kind of possible implementation, more than one the face candidate region (step S101) obtained in image can be with
For：Using more than one face candidate region in one or more detector acquisition images.
As an example of the implementation, figure can be obtained using the detector that detection speed is fast and recall rate is high
More than one face candidate region as in.It is for instance possible to use the AdaBoost detectors of detection speed.Wherein,
The detection speed of AdaBoost detectors can reach 5ms/frame.For another example can be by reducing the detection threshold value of detector
The recall rate for improving detector is realized with accuracy of detection.
It should be noted that those skilled in the art are it should be understood that in more than one face candidate region of acquisition
In, some face candidate regions include real face, and some face candidate regions do not include real face.It can be directed to
More than one the face candidate region obtained is merged, and convolutional calculation is carried out after obtaining a new region, determines that this is new
Human face region in region, improves the accuracy of Face datection.Calculated using the Face datection based on deep learning of convolutional calculation
Method has more preferably Detection results, and preferable accuracy of detection can be accessed under various harsh testing conditions.For example, illumination
Acute variation (strong light or half-light etc.), human face posture is abnormal (side face or wry face etc.), serious shielding (wear masks, cap or sunglasses
Deng), human face expression is abnormal (laugh or cry bitterly).
In step S202, more than one face candidate region is merged, combined region is obtained.
, can be by the case of a face candidate region in obtaining image as an example of the present embodiment
One face candidate region is used as combined region.
, can be by the case of multiple face candidate regions in obtaining image as an example of the present embodiment
The plurality of face candidate region is merged, and obtains combined region.For example, can be taken simultaneously to the plurality of face candidate region
Collection operation, obtains combined region.
In a kind of possible implementation, more than one face candidate region is merged, combined region (step is obtained
Rapid S102) can be：More than one face candidate region is spliced into combined region.
, can be with the case of multiple face candidate regions in obtaining image as an example of the implementation
The plurality of face candidate region is spliced into polygonal region, and the polygonal region that the splicing is obtained is used as combined region.
It should be noted that those skilled in the art are it should be understood that there is various ways to realize in correlation technique
The plurality of face candidate region is spliced into polygonal region.For the multiple face candidate regions of identical, using different spellings
The mode of connecing can obtain polygonal region of different shapes, and the disclosure is not restricted to this.
In step S203, convolutional calculation is carried out to the combined region, convolution results are obtained.
In a kind of possible implementation, convolutional calculation is carried out to the combined region, convolution results (step is obtained
S103) can be：Convolutional calculation only is carried out to the combined region, convolution results are obtained.
Wherein, the combined region can be new region, and the area of the combined region is less than the area of image.
In a kind of possible implementation, if the area of the combined region is the 1/n of the area of image, the combined region is rolled up
The time that product is calculated can be the 1/n for the time that convolutional calculation is carried out to the image, thus, it is possible to greatly reduce convolutional calculation
Time, accelerate Face datection.
In step S104, Face datection is carried out according to the convolution results.
As an example of the present embodiment, using detector scan image, more than one face obtained in image is waited
Favored area, for example, obtain five face candidate regions in image.Five face candidate regions are carried out taking union operation, obtained
To combined region.Convolutional calculation is carried out to the combined region, convolution results are obtained.Face datection is carried out according to the convolution results,
For example obtain four human face regions in the combined region.
The method of the Face datection of the disclosure can reduce the area for the image-region for carrying out convolutional calculation, reduce convolution meter
The complexity and operand of calculation, improve the speed of convolutional calculation, accelerate Face datection.
Fig. 4 is an a kind of schematical flow chart of the method for Face datection according to an exemplary embodiment.Such as
Shown in Fig. 4, the method for the Face datection comprises the following steps.
In step S401, using one the detector acquisition image of the face under multiple different conditions suitable for detection
Individual above face candidate region.
In a kind of possible implementation, the detector of multiple faces being applied under detection different condition is including following
At least two：Suitable for detecting the detector of positive face, the detector suitable for detecting side face, suitable for detecting the face being blocked
Detector, suitable for detection image brightness be more than first threshold under conditions of face detector and suitable for detection image
Brightness be less than Second Threshold under conditions of face detector, wherein, Second Threshold be less than first threshold.
Wherein, first threshold and Second Threshold can be the numerical value pre-set, and the disclosure is not restricted to this.Image is bright
Degree can be the face under intense light conditions more than the face under conditions of first threshold, and brightness of image is less than the condition of Second Threshold
Under face can be subdued light conditions under face.
In the implementation, it is adaptable to detect the detector of positive face, the detector suitable for detecting side face, suitable for inspection
Survey be blocked face detector, suitable for detection image brightness be more than first threshold under conditions of face detector and
The detector of face being less than suitable for detection image brightness under conditions of Second Threshold can be training in advance
AdaBoost detectors.
As an example of the implementation, four AdaBoost detectors can be trained.Wherein, the first AdaBoost
Detector can be the detector suitable for detecting positive face.2nd AdaBoost detectors can be abnormal suitable for test pose
Or the detector of the abnormal face of expression, such as side face, wry face, laugh or wail.3rd AdaBoost detectors can be
Suitable for detecting the detector of face being blocked, such as wearing masks, cap or sunglasses.4th AdaBoost detectors can be with
For suitable for detector of the detection light according to the face under acute variation, such as strong light or half-light.
In step S402, more than one face candidate region is merged, combined region is obtained.
Description for the step may refer to step S202.
In step S403, convolutional calculation is carried out to the combined region, convolution results are obtained.
Description for the step may refer to step S203.
In step s 404, Face datection is carried out according to the convolution results.
Description for the step may refer to step S204.
The method of the Face datection of the disclosure, by using the detection of multiple faces being applied under detection different condition
Device, the high recall rate thus, it is possible to ensure more than one face candidate region in acquisition image, enabling detect difference
Under the conditions of face, face candidate region is not lost, so as to improve the accuracy of Face datection.
Fig. 5 is a kind of block diagram of the device of Face datection according to an exemplary embodiment.Reference picture 5, the device
Including acquisition module 11, merging module 12, computing module 13 and detection module 14.
Wherein, the acquisition module 11 is configured as obtaining more than one face candidate region in image；The merging module
12 are configured as merging in more than one face candidate region, obtain combined region；The computing module 13 is configured as pair
The combined region carries out convolutional calculation, obtains convolution results；The detection module 14 is configured as being entered according to the convolution results
Row Face datection.
Fig. 6 is an a kind of schematical block diagram of the device of Face datection according to an exemplary embodiment.Reference
Fig. 6：
In a kind of possible implementation, the merging module 12 includes concatenation module 121.The concatenation module 121 by with
It is set to and more than one face candidate region is spliced into combined region.
In a kind of possible implementation, the acquisition module 11 includes face candidate region acquisition module 111.The face
Candidate region acquisition module 111 is configured as the detector acquisition image using multiple faces being applied under detection different condition
In more than one face candidate region.
In a kind of possible implementation, the detector of the multiple face being applied under detection different condition includes
Below at least two：Suitable for detecting the detector of positive face, the detector suitable for detecting side face, suitable for detecting what is be blocked
The detector of face, suitable for detection image brightness be more than first threshold under conditions of face detector and suitable for detection
Brightness of image be less than Second Threshold under conditions of face detector, wherein, the Second Threshold be less than the first threshold.
On the device in above-described embodiment, wherein modules perform the concrete mode of operation in relevant this method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
The device of the Face datection of the disclosure can reduce the area for the image-region for carrying out convolutional calculation, reduce convolution meter
The complexity and operand of calculation, improve the speed of convolutional calculation, accelerate Face datection.
Fig. 7 is a kind of block diagram of the device 800 of Face datection according to an exemplary embodiment.For example, device 800
Can be mobile phone, computer, digital broadcast terminal, messaging devices, game console, tablet device, Medical Devices,
Body-building equipment, personal digital assistant etc..
Reference picture 7, device 800 can include following one or more assemblies：Processing assembly 802, memory 804, power supply
Component 806, multimedia groupware 808, audio-frequency assembly 810, the interface 812 of input/output (I/O), sensor cluster 814, and
Communication component 816.
The integrated operation of the usual control device 800 of processing assembly 802, such as with display, call, data communication, phase
Machine operates the operation associated with record operation.Processing assembly 802 can refer to including one or more processors 820 to perform
Order, to complete all or part of step of above-mentioned method.In addition, processing assembly 802 can include one or more modules, just
Interaction between processing assembly 802 and other assemblies.For example, processing assembly 802 can include multi-media module, it is many to facilitate
Interaction between media component 808 and processing assembly 802.
Memory 804 is configured as storing various types of data supporting the operation in device 800.These data are shown
Example includes the instruction of any application program or method for being operated on device 800, and contact data, telephone book data disappears
Breath, picture, video etc..Memory 804 can be by any kind of volatibility or non-volatile memory device or their group
Close and realize, such as static RAM (SRAM), Electrically Erasable Read Only Memory (EEPROM) is erasable to compile
Journey read-only storage (EPROM), programmable read only memory (PROM), read-only storage (ROM), magnetic memory, flash
Device, disk or CD.
Power supply module 806 provides electric power for the various assemblies of device 800.Power supply module 806 can include power management system
System, one or more power supplys, and other components associated with generating, managing and distributing electric power for device 800.
Multimedia groupware 808 is included in the screen of one output interface of offer between described device 800 and user.One
In a little embodiments, screen can include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Curtain may be implemented as touch-screen, to receive the input signal from user.Touch panel includes one or more touch sensings
Device is with the gesture on sensing touch, slip and touch panel.The touch sensor can not only sensing touch or sliding action
Border, but also detection touches or slide related duration and pressure with described.In certain embodiments, many matchmakers
Body component 808 includes a front camera and/or rear camera.When device 800 be in operator scheme, such as screening-mode or
During video mode, front camera and/or rear camera can receive the multi-medium data of outside.Each front camera and
Rear camera can be a fixed optical lens system or with focusing and optical zoom capabilities.
Audio-frequency assembly 810 is configured as output and/or input audio signal.For example, audio-frequency assembly 810 includes a Mike
Wind (MIC), when device 800 be in operator scheme, when such as call model, logging mode and speech recognition mode, microphone by with
It is set to reception external audio signal.The audio signal received can be further stored in memory 804 or via communication set
Part 816 is sent.In certain embodiments, audio-frequency assembly 810 also includes a loudspeaker, for exports audio signal.
I/O interfaces 812 is provide interface between processing assembly 802 and peripheral interface module, above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include but be not limited to：Home button, volume button, start button and lock
Determine button.
Sensor cluster 814 includes one or more sensors, and the state for providing various aspects for device 800 is commented
Estimate.For example, sensor cluster 814 can detect opening/closed mode of device 800, the relative positioning of component is for example described
Component is the display and keypad of device 800, and sensor cluster 814 can be with 800 1 components of detection means 800 or device
Position change, the existence or non-existence that user contacts with device 800, the orientation of device 800 or acceleration/deceleration and device 800
Temperature change.Sensor cluster 814 can include proximity transducer, be configured to detect in not any physical contact
The presence of neighbouring object.Sensor cluster 814 can also include optical sensor, such as CMOS or ccd image sensor, for into
As being used in application.In certain embodiments, the sensor cluster 814 can also include acceleration transducer, gyro sensors
Device, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 816 is configured to facilitate the communication of wired or wireless way between device 800 and other equipment.Device
800 can access the wireless network based on communication standard, such as WiFi, 2G or 3G, or combinations thereof.In an exemplary implementation
In example, communication component 816 receives broadcast singal or broadcast related information from external broadcasting management system via broadcast channel.
In one exemplary embodiment, the communication component 816 also includes near-field communication (NFC) module, to promote junction service.Example
Such as, NFC module can be based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) technology,
Bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, device 800 can be believed by one or more application specific integrated circuits (ASIC), numeral
Number processor (DSP), digital signal processing appts (DSPD), PLD (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for performing the above method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instructing, example are additionally provided
Such as include the memory 804 of instruction, above-mentioned instruction can be performed to complete the above method by the processor 820 of device 800.For example,
The non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk
With optical data storage devices etc..
Those skilled in the art will readily occur to its of the disclosure after considering specification and putting into practice invention disclosed herein
Its embodiment.The application is intended to any modification, purposes or the adaptations of the disclosure, these modifications, purposes or
Person's adaptations follow the general principle of the disclosure and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.Description and embodiments are considered only as exemplary, and the true scope of the disclosure and spirit are by following
Claim is pointed out.
It should be appreciated that the precision architecture that the disclosure is not limited to be described above and is shown in the drawings, and
And various modifications and changes can be being carried out without departing from the scope.The scope of the present disclosure is only limited by appended claim.
Claims (10)
1. a kind of method of Face datection, it is characterised in that including：
Obtain more than one face candidate region in image；
More than one face candidate region is merged, combined region is obtained；
Convolutional calculation is carried out to the combined region, convolution results are obtained；
Face datection is carried out according to the convolution results.
2. according to the method described in claim 1, it is characterised in that more than one face candidate region is merged, obtained
Combined region, including：
More than one face candidate region is spliced into combined region.
3. according to the method described in claim 1, it is characterised in that obtain more than one face candidate region in image, bag
Include：
Using more than one face candidate area the detector acquisition image of the face under multiple different conditions suitable for detection
Domain.
4. method according to claim 3, it is characterised in that the multiple face being applied under detection different condition
Detector includes following at least two：
Suitable for detecting the detector of positive face, the detector suitable for detecting side face, the inspection suitable for detecting the face being blocked
Survey device, suitable for detection image brightness be more than first threshold under conditions of face detector and suitable for detection image brightness
Less than the detector of the face under conditions of Second Threshold, wherein, the Second Threshold is less than the first threshold.
5. a kind of device of Face datection, it is characterised in that including：
Acquisition module, for obtaining more than one face candidate region in image；
Merging module, for more than one face candidate region to be merged, obtains combined region；
Computing module, for carrying out convolutional calculation to the combined region, obtains convolution results；
Detection module, for carrying out Face datection according to the convolution results.
6. device according to claim 5, it is characterised in that the merging module includes：
Concatenation module, for more than one face candidate region to be spliced into combined region.
7. device according to claim 5, it is characterised in that the acquisition module includes：
Face candidate region acquisition module, for the detector acquisition figure using multiple faces being applied under detection different condition
More than one face candidate region as in.
8. device according to claim 7, it is characterised in that the multiple face being applied under detection different condition
Detector includes following at least two：
Suitable for detecting the detector of positive face, the detector suitable for detecting side face, the inspection suitable for detecting the face being blocked
Survey device, suitable for detection image brightness be more than first threshold under conditions of face detector and suitable for detection image brightness
Less than the detector of the face under conditions of Second Threshold, wherein, the Second Threshold is less than the first threshold.
9. a kind of device of Face datection, it is characterised in that including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
Obtain more than one face candidate region in image；
More than one face candidate region is merged, combined region is obtained；
Convolutional calculation is carried out to the combined region, convolution results are obtained；
Face datection is carried out according to the convolution results.
10. a kind of non-volatile computer readable storage medium storing program for executing, is stored thereon with computer program instructions, it is characterised in that institute
State and method in Claims 1-4 described in any one is realized when computer program instructions are executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710618497.7A CN107292293A (en) | 2017-07-26 | 2017-07-26 | The method and device of Face datection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710618497.7A CN107292293A (en) | 2017-07-26 | 2017-07-26 | The method and device of Face datection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107292293A true CN107292293A (en) | 2017-10-24 |
Family
ID=60102804
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710618497.7A Pending CN107292293A (en) | 2017-07-26 | 2017-07-26 | The method and device of Face datection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107292293A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509894A (en) * | 2018-03-28 | 2018-09-07 | 北京市商汤科技开发有限公司 | Method for detecting human face and device |
| CN109558864A (en) * | 2019-01-16 | 2019-04-02 | 苏州科达科技股份有限公司 | Face critical point detection method, apparatus and storage medium |
| CN110210474A (en) * | 2019-04-30 | 2019-09-06 | 北京市商汤科技开发有限公司 | Object detection method and device, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101571912A (en) * | 2008-04-30 | 2009-11-04 | 中国科学院半导体研究所 | Computer face-positioning method based on human visual simulation |
| CN104573715A (en) * | 2014-12-30 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Recognition method and device for image main region |
| CN105528078A (en) * | 2015-12-15 | 2016-04-27 | 小米科技有限责任公司 | Method and device controlling electronic equipment |
| CN105718868A (en) * | 2016-01-18 | 2016-06-29 | 中国科学院计算技术研究所 | Face detection system and method for multi-pose faces |
| CN105844248A (en) * | 2016-03-29 | 2016-08-10 | 北京京东尚科信息技术有限公司 | Human face detection method and human face detection device |
| CN106295515A (en) * | 2016-07-28 | 2017-01-04 | 北京小米移动软件有限公司 | Determine the method and device of human face region in image |
| CN106845338A (en) * | 2016-12-13 | 2017-06-13 | 深圳市智美达科技股份有限公司 | Pedestrian detection method and system in video flowing |
-
2017
- 2017-07-26 CN CN201710618497.7A patent/CN107292293A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101571912A (en) * | 2008-04-30 | 2009-11-04 | 中国科学院半导体研究所 | Computer face-positioning method based on human visual simulation |
| CN104573715A (en) * | 2014-12-30 | 2015-04-29 | 百度在线网络技术（北京）有限公司 | Recognition method and device for image main region |
| CN105528078A (en) * | 2015-12-15 | 2016-04-27 | 小米科技有限责任公司 | Method and device controlling electronic equipment |
| CN105718868A (en) * | 2016-01-18 | 2016-06-29 | 中国科学院计算技术研究所 | Face detection system and method for multi-pose faces |
| CN105844248A (en) * | 2016-03-29 | 2016-08-10 | 北京京东尚科信息技术有限公司 | Human face detection method and human face detection device |
| CN106295515A (en) * | 2016-07-28 | 2017-01-04 | 北京小米移动软件有限公司 | Determine the method and device of human face region in image |
| CN106845338A (en) * | 2016-12-13 | 2017-06-13 | 深圳市智美达科技股份有限公司 | Pedestrian detection method and system in video flowing |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509894A (en) * | 2018-03-28 | 2018-09-07 | 北京市商汤科技开发有限公司 | Method for detecting human face and device |
| CN109558864A (en) * | 2019-01-16 | 2019-04-02 | 苏州科达科技股份有限公司 | Face critical point detection method, apparatus and storage medium |
| CN110210474A (en) * | 2019-04-30 | 2019-09-06 | 北京市商汤科技开发有限公司 | Object detection method and device, equipment and storage medium |
| CN110210474B (en) * | 2019-04-30 | 2021-06-01 | 北京市商汤科技开发有限公司 | Target detection method and device, equipment and storage medium |
| US11151358B2 (en) | 2019-04-30 | 2021-10-19 | Beijing Sensetime Technology Development Co., Ltd. | Target detection method and apparatus, device, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104036240B (en) | The localization method and device of human face characteristic point | |
| CN106791893A (en) | Net cast method and device | |
| CN107239535A (en) | Similar pictures search method and device | |
| CN106339680A (en) | Human face key point positioning method and device | |
| CN106548468B (en) | The method of discrimination and device of image definition | |
| CN108848313B (en) | Multi-person photographing method, terminal and storage medium | |
| CN104598076B (en) | Touch information screen method and device | |
| CN106651955A (en) | Method and device for positioning object in picture | |
| CN107818180A (en) | Video correlating method, image display method, device and storage medium | |
| CN106951884A (en) | Gather method, device and the electronic equipment of fingerprint | |
| CN106778531A (en) | Face detection method and device | |
| CN108037863A (en) | A kind of method and apparatus for showing image | |
| US10248855B2 (en) | Method and apparatus for identifying gesture | |
| CN107582028A (en) | Sleep monitor method and device | |
| US20210248363A1 (en) | Posture detection method, apparatus and device, and storage medium | |
| CN106231419A (en) | Operation performs method and device | |
| CN106020671A (en) | Adjustment method and device for response sensitivity of fingerprint sensor | |
| CN106682736A (en) | Image identification method and apparatus | |
| CN105975961B (en) | The method, apparatus and terminal of recognition of face | |
| CN106228158A (en) | The method and apparatus of picture detection | |
| CN106055707A (en) | Bullet screen display method and device | |
| CN107463903A (en) | Face key independent positioning method and device | |
| CN107563994A (en) | The conspicuousness detection method and device of image | |
| CN106339695A (en) | Face similarity detection method, device and terminal | |
| CN107832746A (en) | Expression recognition method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171024 |