CN108563982A - Method and apparatus for detection image - Google Patents
Method and apparatus for detection image Download PDFInfo
- Publication number
- CN108563982A CN108563982A CN201810011481.4A CN201810011481A CN108563982A CN 108563982 A CN108563982 A CN 108563982A CN 201810011481 A CN201810011481 A CN 201810011481A CN 108563982 A CN108563982 A CN 108563982A
- Authority
- CN
- China
- Prior art keywords
- topography
- face
- overlapping
- degree
- preset quantity
- 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
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the present application discloses the method and apparatus for detection image.One specific implementation mode of this method includes：From the target location of target image, preset rectangle frame is moved according to preset direction and default displacement distance in the target image at least once, obtain at least two topographies that rectangle frame is confined, wherein the target image presents at least one face；Determine the degree of overlapping of topography of at least one of each face and at least two topography at least one face；According to the sequence of degree of overlapping from big to small, preset quantity topography is chosen from least one topography；Face datection is carried out to selected preset quantity topography.The application improves the recall rate of Face datection by confining at least two topographies and determining degree of overlapping.
Description
Technical field
The invention relates to field of computer technology, and in particular to Internet technical field is more particularly, to examined
The method and apparatus of altimetric image.
Background technology
Face datection, i.e., the image given for any one width use certain strategy to be scanned for it to determine it
In whether contain face, if it is return to the information such as position, size and posture of face.
Invention content
The embodiment of the present application proposes the method and apparatus for detection image.
In a first aspect, the embodiment of the present application provides a kind of method for detection image, including：From the mesh of target image
Cursor position is moved preset rectangle frame according to preset direction and default displacement distance at least once in the target image,
Obtain at least two topographies that rectangle frame is confined, wherein target image presents at least one face；Determine at least one
The degree of overlapping of at least one of each face and at least two topographies in a face topography；According to degree of overlapping from
Small sequence is arrived greatly, and preset quantity topography is chosen from least one topography；To selected preset quantity
Topography carries out Face datection.
In some embodiments, it determines in each face and at least two topographies at least one face at least
The degree of overlapping of one topography, including：For each of at least one face face, selected from least two topographies
Take has overlapping at least one topography with the face；Determine the face and each Local map at least one topography
The size of the intersection of picture and the size of union determine that the ratio of the size of intersection and the size of union is degree of overlapping.
In some embodiments, selected preset quantity topography carries out Face datection, including：Using convolution god
Through network, Face datection is carried out to selected preset quantity topography.
In some embodiments, selected preset quantity topography carries out Face datection, including：Using non-very big
It is worth restrainable algorithms, Face datection is carried out to selected preset quantity topography.
Second aspect, the embodiment of the present application provide a kind of device for detection image, including：Mobile unit, configuration
For from the target location of target image, in the target image according to preset direction and default displacement distance by preset rectangle frame
It is moved at least once, obtains at least two topographies that rectangle frame is confined, wherein target image presents at least one
A face；Determination unit is configured to determine in each face and at least two topographies at least one face extremely
The degree of overlapping of a few topography；Selection unit is configured to the sequence from big to small according to degree of overlapping, from least one office
Preset quantity topography is chosen in portion's image；Detection unit is configured to selected preset quantity topography
Carry out Face datection.
In some embodiments, determination unit is further configured to：For each of at least one face face, from
Chosen at least two topographies has overlapping at least one topography with the face；Determine the face and at least one office
The size of the size and union of the intersection of each topography in portion's image determines the ratio of the size of intersection and the size of union
Value is degree of overlapping.
In some embodiments, detection unit is further configured to：Using convolutional neural networks, selected is preset
Quantity topography carries out Face datection.
In some embodiments, detection unit is further configured to：Using non-maxima suppression algorithm, to selected
Preset quantity topography carries out Face datection.
The third aspect, the embodiment of the present application provide a kind of electronic equipment, including：One or more processors；Storage dress
It sets, for storing one or more programs, when one or more programs are executed by one or more processors so that one or more
A processor is realized such as the method for any embodiment in the method for detection image.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey
Sequence is realized when the program is executed by processor such as the method for any embodiment in the method for detection image.
Method and apparatus provided by the embodiments of the present application for detection image, first from the target location of target image,
Preset rectangle frame is moved at least once in the target image according to preset direction and default displacement distance, obtains rectangle
At least two topographies that frame is confined, wherein target image presents at least one face.Later, it determines at least one
The degree of overlapping of at least one of each face and at least two topographies in face topography.Then, according to overlapping
The sequence of degree from big to small, chooses preset quantity topography from least one topography.Finally.To selected pre-
If quantity topography carries out Face datection.The application is carried by confining at least two topographies and determining degree of overlapping
The high recall rate of Face datection.
Description of the drawings
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is that this application can be applied to exemplary system architecture figures therein；
Fig. 2 is the flow chart according to one embodiment of the method for detection image of the application；
Fig. 3 is the schematic diagram according to an application scenarios of the method for detection image of the application；
Fig. 4 is the flow chart according to another embodiment of the method for detection image of the application；
Fig. 5 is the flow chart according to another embodiment of the method for detection image of the application；
Fig. 6 is the structural schematic diagram according to one embodiment of the device for detection image of the application；
Fig. 7 is adapted for the structural schematic diagram of the computer system of the electronic equipment for realizing the embodiment of the present application.
Specific implementation mode
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, is illustrated only in attached drawing and invent relevant part with related.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 shows the implementation of the method for detection image or the device for detection image that can apply the application
The exemplary system architecture 100 of example.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 provide communication link medium.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be interacted by network 104 with server 105 with using terminal equipment 101,102,103, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed on terminal device 101,102,103, such as Image Acquisition application,
Image processing application, face recognition application etc..
Terminal device 101,102,103 can be the various electronic equipments with display screen and supported web page browsing, packet
Include but be not limited to smart mobile phone, tablet computer, E-book reader, pocket computer on knee and desktop computer etc..
Server 105 can be to provide the server of various services, such as to being shown on terminal device 101,102,103
Content provides the background server supported.Background server can carry out the data such as the target image that receives analyzing etc.
Reason, and handling result is fed back into terminal device.
It should be noted that the method for detection image that is provided of the embodiment of the present application can by terminal device 101,
102,103 and server 105 execute, correspondingly, the device for detection image is generally positioned at terminal device 101,102,103
In server 105.
It should be understood that the number of the terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the flow of one embodiment of the method for detection image according to the application is shown
200.This is used for the method for detection image, includes the following steps：
Step 201, from the target location of target image, by preset rectangle frame according to preset direction and default displacement distance
It is moved at least once in the target image, obtains at least two topographies that rectangle frame is confined.
In the present embodiment, it is used for electronic equipment (such as the service shown in FIG. 1 of the method operation of detection image thereon
Device) it can be since the target location of target image, by preset rectangle frame according to preset direction and default displacement distance in mesh
It is moved at least once in logo image, obtains at least two topographies that rectangle frame is confined.The size of default rectangle frame
It is pre-set.Target location can be any position set in the target image.Preset direction may include one or
Person two and more than two directions.For example, rectangle frame can move in multiple rows of image, the starting point per a line is default
, it is left end, preset direction is from left to right.In addition it is also possible to be the starting point of the first row in left end, the preset direction of the first row
It is that from left to right, for the starting point of the second row in right end, the preset direction of the second row is from right to left.
In practice, topography may include the topography that rectangle frame is confined in initial position, and include moving
The each topography confined after dynamic.
Step 202, office of at least one of each face and at least two topographies at least one face is determined
The degree of overlapping of portion's image.
In the present embodiment, above-mentioned electronic equipment determines each face at least one face presented in target image
With the degree of overlapping of topography of at least one of above-mentioned at least two topography.At least one topography can be to
It is randomly selected in Shao Liangge topographies, can also be to be chosen according to preset rules.Here degree of overlapping refers to any one
The ratio of the size of intersection between the face that topography and any one image are presented and the size of union.Here big
It is small can also to be calculated using number of pixels etc. with area to calculate.
Step 203, the sequence according to degree of overlapping from big to small chooses preset quantity office from least one topography
Portion's image.
In the present embodiment, sequence of the above-mentioned electronic equipment according to degree of overlapping from big to small, from least one topography
Middle selection preset quantity topography.Specifically, above-mentioned electronic equipment can be according to descending sequence, to each overlapping
Degree is ranked up.Later, from the larger one end of the degree of overlapping of the sequence of sequence gained, preset quantity degree of overlapping is determined, and really
Fixed preset quantity topography corresponding with above-mentioned preset quantity degree of overlapping.
Step 204, Face datection is carried out to selected preset quantity topography.
In the present embodiment, after choosing to preset quantity topography, above-mentioned electronic equipment can be to selected
Topography carry out Face datection, to obtain Face datection result.Face datection result can be face characteristic, for example, people
Face feature can be the location information of each characteristic point, can also be the length etc. between characteristic point.
It is a signal according to the application scenarios of the method for detection image of the present embodiment with continued reference to Fig. 3, Fig. 3
Figure.In the application scenarios of Fig. 3, electronic equipment 301 is from the coordinate (0,0) of image 302, by preset rectangle frame according to default side
To at least 9 movements are carried out in image 302 with default displacement distance, 10 topographies 303 that rectangle frame is confined are obtained,
Wherein, image 302 presents 2 faces.Electronic equipment 301 determines in each face and 10 topographies in 2 faces
6 topographies degree of overlapping 304.According to the sequence of degree of overlapping 304 from big to small, 4 offices are chosen from 6 topographies
Portion's image 305.Face datection 306 is carried out to 4 selected topographies.
The method that above-described embodiment of the application provides, can be from topography by confining at least two topographies
The problem of detecting more faces, avoiding missing inspection, and then improve the recall rate of Face datection.Meanwhile the application is implemented
Example by choosing preset quantity topography, avoid detect the very low topography of degree of overlapping and caused by invalid detection,
Improve detection efficiency.
With further reference to Fig. 4, it illustrates the flows 400 of another embodiment of the method for detection image.The use
In the flow 400 of the method for detection image, include the following steps：
Step 401, from the target location of target image, by preset rectangle frame according to preset direction and default displacement distance
It is moved at least once in the target image, obtains at least two topographies that rectangle frame is confined.
In the present embodiment, server can be since the target location of target image, by preset rectangle frame according to pre-
Set direction and default displacement distance are moved at least once in the target image, are obtained at least two innings that rectangle frame is confined
Portion's image.The size of default rectangle frame is pre-set.Target location can be any position set in the target image.
Preset direction may include one or two and more than two directions.
Step 402, it for each of at least one face face, is chosen and the face from least two topographies
There is at least one topography of overlapping.
In the present embodiment, above-mentioned server is for each of above-mentioned at least one face face, from above-mentioned at least two
Chosen in a topography has overlapping at least one topography with the face.Exist in above-mentioned at least two topography
At least one topography overlaps with the face, and above-mentioned server, which is then selected, to be come.
Step 403, the size and simultaneously of the intersection of each topography in the face and at least one topography is determined
The size of collection determines that the ratio of the size of intersection and the size of union is degree of overlapping.
In the present embodiment, above-mentioned server determines the face and each Local map in above-mentioned at least one topography
The size of the intersection of picture and the size of union.Later, it determines the ratio of the size of intersection and the size of union, and the ratio is made
For degree of overlapping.
Step 404, the sequence according to degree of overlapping from big to small chooses preset quantity office from least one topography
Portion's image.
In the present embodiment, sequence of the above-mentioned server according to degree of overlapping from big to small, from least one topography
Choose preset quantity topography.Specifically, above-mentioned server can according to descending sequence, to each degree of overlapping into
Row sequence.Later, it from the larger one end of degree of overlapping, determines preset quantity degree of overlapping, and determines and above-mentioned preset quantity weight
The folded corresponding preset quantity topography of degree.
Step 405, using convolutional neural networks, Face datection is carried out to selected preset quantity topography.
In the present embodiment, above-mentioned server uses convolutional neural networks, to selected preset quantity topography
Carry out Face datection.Specifically, above-mentioned server can be by each Local map in selected preset quantity topography
As input convolutional neural networks, the face characteristic exported by convolutional neural networks is obtained.In practice, convolutional neural networks
(Convolutional Neural Network, CNN) is a kind of feedforward neural network, its artificial neuron can respond one
Surrounding cells in partial coverage have outstanding performance for image procossing, therefore, it is possible to utilize multilayer convolutional neural networks
Face datection is carried out, face characteristic is obtained.It should be noted that multilayer convolutional neural networks can utilize machine learning method
Existing depth convolutional neural networks (such as ResNet-50) are carried out obtained from Training with training sample.Wherein,
Training sample may include the face characteristic of a large amount of image and the presented face of image.
The present embodiment can realize more accurate Face datection by convolutional neural networks.
With further reference to Fig. 5, it illustrates the flows 500 of another embodiment of the method for detection image.The use
In the flow 500 of the method for detection image, include the following steps：
Step 501, from the target location of target image, by preset rectangle frame according to preset direction and default displacement distance
It is moved at least once in the target image, obtains at least two topographies that rectangle frame is confined.
In the present embodiment, terminal device can since the target location of target image, by preset rectangle frame according to
Preset direction and default displacement distance are moved at least once in the target image, obtain rectangle frame is confined at least two
Topography.The size of default rectangle frame is pre-set.Target location can be the arbitrary position set in the target image
It sets.Preset direction may include one or two and more than two directions.For example, may be used rectangle frame in the picture into
Row multirow moves, and the starting point per a line is preset, is left end, preset direction is from left to right.In addition it is also possible to be the first row
Starting point in left end, the starting point of the second row is in right end etc..
Step 502, office of at least one of each face and at least two topographies at least one face is determined
The degree of overlapping of portion's image.
In the present embodiment, above-mentioned terminal device determines each face at least one face presented in target image
With the degree of overlapping of topography of at least one of above-mentioned at least two topography.At least one topography can be to
It is randomly selected in Shao Liangge topographies, can also be to be chosen according to preset rules.Here degree of overlapping refers to any one
The ratio of the size of intersection between topography and any one face presented and the size of union.Here size can
Can also be calculated using number of pixels etc. with area to calculate.
Step 503, the sequence according to degree of overlapping from big to small chooses preset quantity office from least one topography
Portion's image.
In the present embodiment, sequence of the above-mentioned terminal device according to degree of overlapping from big to small, from least one topography
Middle selection preset quantity topography.Specifically, above-mentioned terminal device can be according to descending sequence, to each overlapping
Degree is ranked up.Later, from the larger one end of degree of overlapping, preset quantity degree of overlapping is determined, and determine and above-mentioned preset quantity
A corresponding preset quantity topography of degree of overlapping.
Step 504, using non-maxima suppression algorithm, face inspection is carried out to selected preset quantity topography
It surveys.
In the present embodiment, above-mentioned terminal device uses non-maxima suppression algorithm, to selected preset quantity office
Portion's image carries out Face datection.Non-maxima suppression algorithm is each face for being presented, it is pair Chong Die with the face simultaneously
And have each topography of overlapping between each other, the highest at least one topography of confidence level therein can be chosen and carried out
Face datection.And Face datection is carried out to each topography not being overlapped between each other.
The present embodiment can realize more accurate Face datection by non-maxima suppression algorithm.
With further reference to Fig. 6, as the realization to method shown in above-mentioned each figure, this application provides one kind for detecting figure
One embodiment of the device of picture, the device embodiment is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in various electronic equipments.
As shown in fig. 6, the device 600 for detection image of the present embodiment includes：Mobile unit 601, determination unit
602, selection unit 603 and detection unit 604.Wherein, mobile unit 601 are configured to from the target location of target image, will
Preset rectangle frame is moved at least once in the target image according to preset direction and default displacement distance, obtains rectangle frame
At least two topographies confined, wherein target image presents at least one face；Determination unit 602, is configured to
Determine the degree of overlapping of topography of at least one of each face and at least two topographies at least one face；Choosing
Unit 603 is taken, the sequence from big to small according to degree of overlapping is configured to, preset quantity is chosen from least one topography
Topography；Detection unit 604 is configured to carry out Face datection to selected preset quantity topography.
In the present embodiment, mobile unit 601 can press preset rectangle frame since the target location of target image
It is moved at least once in the target image according to preset direction and default displacement distance, obtains rectangle frame is confined at least two
A topography.The size of default rectangle frame is pre-set.Target location can be set in the target image it is arbitrary
Position.Preset direction may include one or two and more than two directions.
In the present embodiment, determination unit 602 determines each face at least one face presented in target image
With the degree of overlapping of topography of at least one of above-mentioned at least two topography.At least one topography can be to
It is randomly selected in Shao Liangge topographies, can also be to be chosen according to preset rules.Here degree of overlapping refers to any one
The ratio of the size of intersection between the face that topography and any one image are presented and the size of union.Here big
It is small can also to be calculated using number of pixels etc. with area to calculate.
In the present embodiment, sequence of the selection unit 603 according to degree of overlapping from big to small, from least one topography
Choose preset quantity topography.Specifically, selection unit 603 can be according to descending sequence, to each degree of overlapping
It is ranked up.Later, from the larger one end of the degree of overlapping of the sequence of sequence gained, preset quantity degree of overlapping is determined, and determine
Preset quantity topography corresponding with above-mentioned preset quantity degree of overlapping.
In the present embodiment, after choosing to preset quantity topography, detection unit 604 can be to selected
Topography carries out Face datection, to obtain Face datection result.Face datection result can be face characteristic, for example, face
Feature can be the location information of each characteristic point, can also be the length etc. between characteristic point.
In some optional realization methods of the present embodiment, determination unit is further configured to：For at least one
Each of face face, chosen from least two topographies has overlapping at least one topography with the face；Really
The size of the fixed face and the size and union of the intersection of each topography at least one topography, determines intersection
The ratio of the size of size and union is degree of overlapping.
In some optional realization methods of the present embodiment, detection unit is further configured to：Using convolutional Neural
Network carries out Face datection to selected preset quantity topography.
In some optional realization methods of the present embodiment, detection unit is further configured to：Using non-maximum
Restrainable algorithms carry out Face datection to selected preset quantity topography.
Fig. 7 shows the structural schematic diagram of the computer installation of the electronic equipment suitable for being used for realizing the embodiment of the present application.
As shown in fig. 7, computer installation 700 includes central processing unit (CPU) 701, it can be according to being stored in read-only memory
(ROM) it the program in 702 or is executed respectively from the program that storage section 708 is loaded into random access storage device (RAM) 703
Kind action appropriate and processing.In RAM 703, also it is stored with device 700 and operates required various programs and data.CPU
701, ROM 702 and RAM 703 are connected with each other by bus 704.Input/output (I/O) interface 705 is also connected to bus
704。
It is connected to I/O interfaces 705 with lower component：Importation 706 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 707 of spool (CRT), liquid crystal display (LCD) etc. and loud speaker etc.；Storage section 708 including hard disk etc.；
And the communications portion 709 of the network interface card including LAN card, modem etc..Communications portion 709 via such as because
The network of spy's net executes communication process.Driver 710 is also according to needing to be connected to I/O interfaces 705.Detachable media 711, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on driver 710, as needed in order to be read from thereon
Computer program be mounted into storage section 708 as needed.
Particularly, according to an embodiment of the present application, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiments herein includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed by communications portion 709 from network, and/or from detachable media
711 are mounted.When the computer program is executed by central processing unit (CPU) 701, limited in execution the present processes
Above-mentioned function.It should be noted that the computer-readable medium of the application can be computer-readable signal media or calculating
Machine readable storage medium storing program for executing either the two arbitrarily combines.Computer readable storage medium for example can be --- but it is unlimited
In --- electricity, magnetic, optical, electromagnetic, infrared ray or arrangement of semiconductors, device or device, or the arbitrary above combination.It calculates
The more specific example of machine readable storage medium storing program for executing can include but is not limited to：Being electrically connected, be portable with one or more conducting wires
Formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory
(EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or
The above-mentioned any appropriate combination of person.In this application, can be any include computer readable storage medium or storage program
Tangible medium, the program can be commanded executive device, device either device use or it is in connection.And in this Shen
Please in, computer-readable signal media may include in a base band or as the data-signal that a carrier wave part is propagated,
In carry computer-readable program code.Diversified forms may be used in the data-signal of this propagation, including but not limited to
Electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable
Any computer-readable medium other than storage medium, the computer-readable medium can send, propagate or transmit for by
Instruction executing device, device either device use or program in connection.The journey for including on computer-readable medium
Sequence code can transmit with any suitable medium, including but not limited to：Wirelessly, electric wire, optical cable, RF etc. or above-mentioned
Any appropriate combination.
Flow chart in attached drawing and block diagram, it is illustrated that according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part for a part for one module, program segment, or code of table, the module, program segment, or code includes one or more uses
The executable instruction of the logic function as defined in realization.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, this is depended on the functions involved.Also it to note
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit can also be arranged in the processor, for example, can be described as：A kind of processor packet
Include mobile unit, determination unit, selection unit and detection unit.Wherein, the title of these units not structure under certain conditions
The restriction of the pairs of unit itself, for example, determination unit is also described as " determining each face at least one face
With the unit of the degree of overlapping of topography of at least one of at least two topography ".
As on the other hand, present invention also provides a kind of computer-readable medium, which can be
Included in device described in above-described embodiment；Can also be individualism, and without be incorporated the device in.Above-mentioned calculating
Machine readable medium carries one or more program, when said one or multiple programs are executed by the device so that should
Device：From the target location of target image, by preset rectangle frame according to preset direction and default displacement distance in target image
It is middle to be moved at least once, obtain at least two topographies that rectangle frame is confined, wherein target image presents at least
One face；Determine topography of at least one of each face and at least two topographies at least one face
Degree of overlapping；According to the sequence of degree of overlapping from big to small, preset quantity topography is chosen from least one topography；It is right
Selected preset quantity topography carries out Face datection.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.People in the art
Member should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature
Other technical solutions of arbitrary combination and formation.Such as features described above has similar work(with (but not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of method for detection image, including：
From the target location of target image, by preset rectangle frame according to preset direction and default displacement distance in the target figure
It is moved at least once as in, obtains at least two topographies that rectangle frame is confined, wherein the target image is presented
There is at least one face；
Determine at least one of each face and at least two topography at least one face Local map
The degree of overlapping of picture；
According to the sequence of degree of overlapping from big to small, preset quantity topography is chosen from least one topography；
Face datection is carried out to selected preset quantity topography.
2. the method according to claim 1 for detection image, wherein in determination at least one face
The degree of overlapping of at least one of each face and at least two topography topography, including：
For each of at least one face face, chosen from least two topography has weight with the face
Folded at least one topography；Determine the face and the intersection of each topography at least one topography
The size of size and union determines that the ratio of the size of the intersection and the size of the union is degree of overlapping.
3. the method according to claim 1 for detection image, wherein the selected preset quantity Local map
As progress Face datection, including：
Using convolutional neural networks, Face datection is carried out to selected preset quantity topography.
4. the method according to claim 1 for detection image, wherein the selected preset quantity Local map
As progress Face datection, including：
Using non-maxima suppression algorithm, Face datection is carried out to selected preset quantity topography.
5. a kind of device for detection image, including：
Mobile unit is configured to from the target location of target image, by preset rectangle frame according to preset direction and default shifting
Dynamic distance is moved at least once in the target image, obtains at least two topographies that rectangle frame is confined,
In, the target image presents at least one face；
Determination unit is configured to determine in each face and at least two topography at least one face
At least one topography degree of overlapping；
Selection unit is configured to the sequence from big to small according to degree of overlapping, is chosen from least one topography pre-
If quantity topography；
Detection unit is configured to carry out Face datection to selected preset quantity topography.
6. the device according to claim 5 for detection image, wherein the determination unit is further configured to：
For each of at least one face face, chosen from least two topography has weight with the face
Folded at least one topography；Determine the face and the intersection of each topography at least one topography
The size of size and union determines that the ratio of the size of the intersection and the size of the union is degree of overlapping.
7. the device according to claim 5 for detection image, wherein the detection unit is further configured to：
Using convolutional neural networks, Face datection is carried out to selected preset quantity topography.
8. the device according to claim 5 for detection image, wherein the detection unit is further configured to：
Using non-maxima suppression algorithm, Face datection is carried out to selected preset quantity topography.
9. a kind of electronic equipment, including：
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors so that one or more of processors are real
The now method as described in any in claim 1-4.
10. a kind of computer readable storage medium, is stored thereon with computer program, wherein when the program is executed by processor
Realize the method as described in any in claim 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810011481.4A CN108563982B (en) | 2018-01-05 | 2018-01-05 | Method and apparatus for detecting image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810011481.4A CN108563982B (en) | 2018-01-05 | 2018-01-05 | Method and apparatus for detecting image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108563982A true CN108563982A (en) | 2018-09-21 |
| CN108563982B CN108563982B (en) | 2020-01-17 |
Family
ID=63530200
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810011481.4A Active CN108563982B (en) | 2018-01-05 | 2018-01-05 | Method and apparatus for detecting image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108563982B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242801A (en) * | 2018-09-26 | 2019-01-18 | 北京字节跳动网络技术有限公司 | Image processing method and device |
| CN111353473A (en) * | 2020-03-30 | 2020-06-30 | 浙江大华技术股份有限公司 | Face detection method and device, electronic equipment and storage medium |
| WO2021169126A1 (en) * | 2020-02-25 | 2021-09-02 | 平安科技（深圳）有限公司 | Lesion classification model training method and apparatus, computer device, and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6445819B1 (en) * | 1998-09-10 | 2002-09-03 | Fuji Photo Film Co., Ltd. | Image processing method, image processing device, and recording medium |
| US20060133654A1 (en) * | 2003-01-31 | 2006-06-22 | Toshiaki Nakanishi | Image processing device and image processing method, and imaging device |
| CN102750929A (en) * | 2011-04-20 | 2012-10-24 | 索尼公司 | Image display control device, image display system, image display control method and computer program |
| CN103400391A (en) * | 2013-08-09 | 2013-11-20 | 北京博思廷科技有限公司 | Multiple-target tracking method and device based on improved random forest |
| CN103996186A (en) * | 2014-04-29 | 2014-08-20 | 小米科技有限责任公司 | Image cutting method and image cutting 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 |
| CN106372624A (en) * | 2016-10-15 | 2017-02-01 | 杭州艾米机器人有限公司 | Human face recognition method and human face recognition system |
| CN106529398A (en) * | 2016-09-23 | 2017-03-22 | 深圳云天励飞技术有限公司 | Quick and accurate face detection method based on cascade structure |
| CN106803263A (en) * | 2016-11-29 | 2017-06-06 | 深圳云天励飞技术有限公司 | A kind of method for tracking target and device |
| CN107066943A (en) * | 2017-03-06 | 2017-08-18 | 中国科学院信息工程研究所 | A kind of method for detecting human face and device |
| CN107220618A (en) * | 2017-05-25 | 2017-09-29 | 中国科学院自动化研究所 | Method for detecting human face and device, computer-readable recording medium, equipment |
-
2018
- 2018-01-05 CN CN201810011481.4A patent/CN108563982B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6445819B1 (en) * | 1998-09-10 | 2002-09-03 | Fuji Photo Film Co., Ltd. | Image processing method, image processing device, and recording medium |
| US20060133654A1 (en) * | 2003-01-31 | 2006-06-22 | Toshiaki Nakanishi | Image processing device and image processing method, and imaging device |
| CN102750929A (en) * | 2011-04-20 | 2012-10-24 | 索尼公司 | Image display control device, image display system, image display control method and computer program |
| CN103400391A (en) * | 2013-08-09 | 2013-11-20 | 北京博思廷科技有限公司 | Multiple-target tracking method and device based on improved random forest |
| CN103996186A (en) * | 2014-04-29 | 2014-08-20 | 小米科技有限责任公司 | Image cutting method and image cutting 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 |
| CN106529398A (en) * | 2016-09-23 | 2017-03-22 | 深圳云天励飞技术有限公司 | Quick and accurate face detection method based on cascade structure |
| CN106372624A (en) * | 2016-10-15 | 2017-02-01 | 杭州艾米机器人有限公司 | Human face recognition method and human face recognition system |
| CN106803263A (en) * | 2016-11-29 | 2017-06-06 | 深圳云天励飞技术有限公司 | A kind of method for tracking target and device |
| CN107066943A (en) * | 2017-03-06 | 2017-08-18 | 中国科学院信息工程研究所 | A kind of method for detecting human face and device |
| CN107220618A (en) * | 2017-05-25 | 2017-09-29 | 中国科学院自动化研究所 | Method for detecting human face and device, computer-readable recording medium, equipment |
Non-Patent Citations (2)
| Title |
|---|
| GILLIAN RHODES等: "Is the Fusiform Face Area Specialized for Faces,Individuation, or Expert Individuation?", 《JOURNAL OF COGNITIVE NEUROSCIENCE》 * |
| 龚小彪: "基于TLD框架的目标跟踪算法研究", 《中国优秀硕士学位论文全文数据库（信息科技辑）》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242801A (en) * | 2018-09-26 | 2019-01-18 | 北京字节跳动网络技术有限公司 | Image processing method and device |
| WO2021169126A1 (en) * | 2020-02-25 | 2021-09-02 | 平安科技（深圳）有限公司 | Lesion classification model training method and apparatus, computer device, and storage medium |
| CN111353473A (en) * | 2020-03-30 | 2020-06-30 | 浙江大华技术股份有限公司 | Face detection method and device, electronic equipment and storage medium |
| CN111353473B (en) * | 2020-03-30 | 2023-04-14 | 浙江大华技术股份有限公司 | Face detection method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108563982B (en) | 2020-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109409517B (en) | The training method and device of object detection network | |
| CN109308681A (en) | Image processing method and device | |
| CN109308469B (en) | Method and apparatus for generating information | |
| CN108446698A (en) | Method, apparatus, medium and the electronic equipment of text are detected in the picture | |
| CN109472264A (en) | Method and apparatus for generating object detection model | |
| CN109242801A (en) | Image processing method and device | |
| CN102682099B (en) | Method and device for providing target pictures in target areas | |
| CN108595628A (en) | Method and apparatus for pushed information | |
| CN109389640A (en) | Image processing method and device | |
| CN109344762A (en) | Image processing method and device | |
| CN106910210A (en) | Method and apparatus for generating image information | |
| CN109255337A (en) | Face critical point detection method and apparatus | |
| CN108563982A (en) | Method and apparatus for detection image | |
| CN109711508B (en) | Image processing method and device | |
| CN109272050B (en) | Image processing method and device | |
| CN110097063A (en) | Data processing method, medium, device and the calculating equipment of electronic equipment | |
| CN109118456A (en) | Image processing method and device | |
| CN108021666A (en) | Page adaptation method and apparatus | |
| CN109255767A (en) | Image processing method and device | |
| CN109345580A (en) | Method and apparatus for handling image | |
| CN108154468A (en) | Lung neoplasm detection method, application server and computer readable storage medium | |
| CN108597034A (en) | Method and apparatus for generating information | |
| CN109597956A (en) | Page display method and device | |
| CN109086828A (en) | Method and apparatus for detecting battery pole piece | |
| CN109064464A (en) | Method and apparatus for detecting battery pole piece burr |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |