CN109035686A - A kind of pre- anti-lost alarm method and device - Google Patents
A kind of pre- anti-lost alarm method and device Download PDFInfo
- Publication number
- CN109035686A CN109035686A CN201810753310.9A CN201810753310A CN109035686A CN 109035686 A CN109035686 A CN 109035686A CN 201810753310 A CN201810753310 A CN 201810753310A CN 109035686 A CN109035686 A CN 109035686A
- Authority
- CN
- China
- Prior art keywords
- image
- character image
- candidate
- target
- character
- 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
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/0202—Child monitoring systems using a transmitter-receiver system carried by the parent and the child
- G08B21/0205—Specific application combined with child monitoring using a transmitter-receiver system
- G08B21/0208—Combination with audio or video communication, e.g. combination with "baby phone" function
-
- 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
- G06V40/168—Feature extraction; Face representation
-
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/0202—Child monitoring systems using a transmitter-receiver system carried by the parent and the child
- G08B21/0261—System arrangements wherein the object is to detect trespassing over a fixed physical boundary, e.g. the end of a garden
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Child & Adolescent Psychology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Human Computer Interaction (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of pre- anti-lost alarm methods and device, which comprises obtains the corresponding target signature of target person image；The corresponding real-time monitoring images of target monitoring equipment are obtained, and identify the character image in the real-time monitoring images, obtain figure map's image set；It is concentrated according to preset personage's candidate feature from the character image and determines candidate character image subset；According to the target signature, determine whether the candidate character image subset includes the target person image；If candidate's character image subset does not include the target person image, prompt messages are sent.Identification can be carried out by age, recognition of face, judge that children under guardianship whether in safety zone, so that it is determined that whether children under guardianship are safe, can effectively improve the accuracy rate of identification.
Description
Technical field
The present embodiments relate to technical field of mobile terminals more particularly to a kind of pre- anti-lost alarm methods and dress
It sets.
Background technique
With the rapid development of mobile terminal, function is also increasingly powerful.In practical applications, it can also use mobile whole
Children are supervised at end, to prevent children from losing.
In the prior art, (Global Position System, the whole world are fixed using GPS by patent CN201510437405.6
Position system) whether/LBS (Location Based Service, based on mobile location-based service) determine children in safety zone.Such as
Fruit does not have, then sends information to parent；And judge whether the expression of children is abnormal by face recognition technology, if abnormal, really
It is in the hole to determine children.
However, GPS/LBS positioning needs hardware device to support, higher cost, and accuracy rate is lower；In addition, age and light
According to etc. environmental informations to will lead to face recognition accuracy rate lower.
Summary of the invention
The present invention provides a kind of pre- anti-lost alarm method and device, to solve the anti-lost above problem of existing children.
According to the first aspect of the invention, a kind of pre- anti-lost alarm method is provided, which comprises
Obtain the corresponding target signature of target person image；
The corresponding real-time monitoring images of target monitoring equipment are obtained, and identify the figure map in the real-time monitoring images
Picture obtains figure map's image set；
It is concentrated according to preset personage's candidate feature from the character image and determines candidate character image subset；
According to the target signature, determine whether the candidate character image subset includes the target person image；
If candidate's character image subset does not include the target person image, prompt messages are sent.
According to the second aspect of the invention, providing a kind of pre- anti-lost warning device includes:
Data obtaining module, for obtaining the corresponding target signature of target person image；
Character image identification module for obtaining the corresponding real-time monitoring images of target monitoring equipment, and identifies the reality
When monitoring image in character image, obtain figure map's image set；
Candidate character image determining module is determined for being concentrated according to preset personage's candidate feature from the character image
Candidate character image subset；
Target person image determining module, for determining that the candidate character image subset is according to the target signature
No includes the target person image；
First warning note module, if not including the target person image for the candidate character image subset,
Send prompt messages.
According to the third aspect of the invention we, a kind of electronic equipment is provided, comprising:
Processor, memory and it is stored in the computer journey that can be run on the memory and on the processor
Sequence, the processor realize the aforementioned pre- anti-lost alarm method when executing described program.
According to the fourth aspect of the invention, provide a kind of readable storage medium storing program for executing, when the instruction in the storage medium by
When the processor of electronic equipment executes, so that electronic equipment is able to carry out the aforementioned pre- anti-lost alarm method.
The embodiment of the invention provides a kind of pre- anti-lost alarm methods and device, which comprises obtains target
The corresponding target signature of character image；The corresponding real-time monitoring images of target monitoring equipment are obtained, and identify the real time monitoring
Character image in image obtains figure map's image set；It is concentrated and is determined from the character image according to preset personage's candidate feature
Candidate character image subset；According to the target signature, determine whether the candidate character image subset includes the target person
Object image；If candidate's character image subset does not include the target person image, prompt messages are sent.It can lead to
Age, recognition of face progress identification are spent, children under guardianship are judged whether in safety zone, so that it is determined that whether children under guardianship pacify
Entirely, it can effectively improve the accuracy rate of identification.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is a kind of specific steps flow chart for pre- anti-lost alarm method that the embodiment of the present invention one provides；
Fig. 2 is a kind of specific steps flow chart of pre- anti-lost alarm method provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of structure chart for pre- anti-lost warning device that the embodiment of the present invention three provides；
Fig. 4 is a kind of structure chart for pre- anti-lost warning device that the embodiment of the present invention four provides.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Embodiment one
Referring to Fig.1, a kind of specific steps of the pre- anti-lost alarm method provided it illustrates the embodiment of the present invention one
Flow chart.
Step 101, the corresponding target signature of target person image is obtained.
Wherein, target person image is the target person of monitoring.In embodiments of the present invention, target person image is to be supervised
The character image of children is protected, it can also be for by the object that other need to guard such as monitoring old man.In addition, target person image can be with
It is selected by guardian.
Target signature may include facial characteristics.
For facial characteristics, can be extracted from target person image.The embodiment of the present invention does not limit extraction algorithm
System.
Further, target signature may also include but be not limited to: height, clothes color, garment type, the one kind at age or
It is a variety of.For height, clothes color, garment type, age etc., can be extracted from target person image, it can also be specified
It concurrently sets, and is stored into system database when target person.
It is appreciated that the above process can be carried out on the mobile terminal of oneself by guardian, it can also be by guardian in place
It is selected in the equipment of offer.For example, example 1: guardian's downloading monitoring application on mobile terminals first；Then monitoring is opened
The set interface of application, the information such as height, age, dress and the color of setting monitoring children, and interface is identified in facial characteristics
Facial characteristics etc. is obtained by scanning face-image；Example 2: guardian carries out selecting in the equipment provided by field and typing is related
Information.
Certainly, above-mentioned height, age, dress and color, facial characteristics can be modified.
Step 102, the corresponding real-time monitoring images of target monitoring equipment are obtained, and are identified in the real-time monitoring images
Character image obtains figure map's image set.
Wherein, target monitoring equipment can be specified by guardian.In practical applications, the position that can be selected according to user
The current location set or positioned shows the camera on the position periphery, for selection；It can also be arranged from all target monitoring equipment
It is selected in table.The embodiment of the present invention is without restriction to the selection mode of target monitoring equipment.It is appreciated that target monitoring equipment
For the monitoring device of user's concern, the as monitoring device of the safety zone of monitoring monitoring children.
It is presented as in real time and periodically obtains monitoring image.It is appreciated that the period for obtaining monitoring image is shorter, then obtain
Monitoring image it is more, can more find in time children whether safety, but image procossing complexity is higher；Obtain the week of monitoring image
Phase is longer, then the monitoring image obtained is fewer, whether safe cannot find children in time, but image procossing complexity is lower.From
And need to set the reasonable period according to practical application scene, it can find children whether safely and as far as possible at reduction in time
Manage complexity.
Specifically, figure map's image set can be obtained by face recognition technology.
In embodiments of the present invention, children can be guarded according to common monitoring equipment, prevents children from losing.In practical application
In, the image data of common monitoring monitoring of tools can be imported in third party database, or directly from common monitoring equipment
Original database in read.
It is appreciated that the present embodiments relate to third party database and monitoring applications, monitoring device and its corresponding
Database, mobile terminal, wherein third party database is that corresponding database is applied in monitoring, and mobile terminal is that monitoring is supported to answer
With the terminal of operation.Specifically, firstly, children's information and mesh that user passes through the specified monitoring of monitoring application on mobile terminal
Mark monitoring device；Then, monitoring data of the monitoring application according to target monitoring equipment in third party database, the real-time judge youngster
It is virgin whether in a safe condition；If in the hole, it is arranged to the mobile terminal or guardian by the mobile terminal
Target terminal sends prompt messages.
Step 103, it is concentrated according to preset personage's candidate feature from the character image and determines candidate character image subset.
In embodiments of the present invention, in order to reduce the interference of computation complexity and other factors to identification, in body
It is concentrated before part identification from character image and rejects ineligible character image.For example, weeding out the age not and being the people of children
Object image, so that candidate feature is to determine the feature at age.It is, of course, also possible to which weeding out other does not obviously meet specified requirements
Character image, so that candidate feature is to determine the feature of specified requirements.
Specifically, preferred, for each of character image collection object image, the personage for extracting the task image is candidate special
Sign, as reference candidate feature；Then, this is compared with reference to candidate feature with personage's candidate feature, if unanimously, the people
Object image is candidate character image；If inconsistent, which is not candidate character image.
Step 104, according to the target signature, determine whether the candidate character image subset includes the target person
Image.
Specifically, firstly, for the candidate character image of each of candidate character image subset, its fixed reference feature is extracted；
Then, which is compared with target signature；If consistent, represent in candidate character image subset comprising target person
Object image, and stop judging other candidate character images；If inconsistent, continue the reference for judging next candidate character image
Whether feature is consistent with target signature.
If the fixed reference feature for all persons' image being appreciated that in candidate character image subset is different with target signature
It causes, then represents candidate character image subset not comprising target person image；If at least one personage in candidate character image subset
The fixed reference feature of image is consistent with target signature, then representing candidate character image subset includes target person image.
In practical applications, if target signature includes height, for height, then allow height difference in very a small range
When, it is consistent to represent height, thus judgement inaccuracy caused by avoiding error.In addition, it is more accurate in order to judge, only work as reference
Feature respectively with corresponding multiple target signatures it is consistent when, judge that representing candidate character image subset includes target person figure
Picture；Otherwise do not include.For example, only when height, age, dress and color, facial characteristics are consistent, it is determined that candidate personage
Image subset includes target person image；Otherwise, if one of them are inconsistent, do not include.
Step 105, if candidate's character image subset does not include the target person image, warning note letter is sent
Breath.
In practical applications, candidate character image subset does not include target person image, then represents monitoring children and do not referring to
Region is determined, to need warning note；Candidate character image subset includes target person image, then represents monitoring children specified
Region, without warning note.
Wherein, prompt messages include but is not limited to: word alarm, audible alarm, vibration alarm etc..The present invention is implemented
Example is without restriction to the concrete mode of prompt messages.
In practical applications, if the candidate character image subset in subsequent time includes target person image, stop reporting
Alert prompt information.
In conclusion the embodiment of the invention provides a kind of pre- anti-lost alarm methods, which comprises obtain mesh
Mark the corresponding target signature of character image；The corresponding real-time monitoring images of target monitoring equipment are obtained, and identify the real-time prison
The character image in image is controlled, figure map's image set is obtained；It is concentrated really according to preset personage's candidate feature from the character image
Fixed candidate's character image subset；According to the target signature, determine whether the candidate character image subset includes the target
Character image；If candidate's character image subset does not include the target person image, prompt messages are sent.It can be with
By the age, recognition of face carry out identification, judge children under guardianship whether in safety zone, so that it is determined that children under guardianship whether
Safety, can effectively improve the accuracy rate of identification.
Embodiment two
Referring to Fig. 2, it illustrates a kind of specific steps of pre- anti-lost alarm method provided by Embodiment 2 of the present invention
Flow chart.
Step 201, the corresponding monitoring image of target monitoring equipment is obtained, and the personage in the monitoring image is returned
Class obtains initial figure map's image set.
Wherein, target monitoring equipment is referred to the detailed description of step 102, and details are not described herein.
Specifically, the personage in monitoring image is sorted out, it may be assumed that the character image in identification monitoring image, and will be same
One people's different angle, different conditions image sort out to the personage.Subsumption algorithm comparative maturity, the embodiment of the present invention
It is not described in detail.
Optionally, in another embodiment of the invention, step 201 includes sub-step 2011:
Sub-step 2011 sorts out the personage in the monitoring image using unsupervised clustering.
Unsupervised clustering can be used in subsumption algorithm, clusters for Fast Learning, without advance demand flag data.
Step 202, it receives and concentrates the target person image of selection from the initial character image, and from the target person
Extracting target from images feature.
The embodiment of the present invention can show each character image that character image is concentrated on mobile terminal, for guardian
Select one of them for target person image.
Optionally, in another embodiment of the invention, the target signature not only includes face characteristic, further includes body
High feature and/or clothes color characteristic.
It is appreciated that the extraction algorithm of target signature is related to the concrete form of target signature.For example, when target signature is
When height, the head of personage and the position of foot are identified from target person image, determine reference system so as to be calculated and appoint
The height of business image；It, can be in conjunction with height and the comprehensive determination of facial characteristics when target signature is the age；At present identification dress and
Color has been the technology of comparative maturity, and not in this to go forth for the embodiment of the present invention.
It is appreciated that step 201 specifies target person image to 202 for guardian, and extract the process of target signature.
Step 203, the confirmation target person request that mobile terminal is sent is obtained.In practical applications, guardian can be with
The image on mobile terminal is uploaded as target person image.To which guardian can shoot image for target person, or, from phase
Selection image uploads in volume.
It is appreciated that carrying the target person image of upload in target person request.
Step 204, the target signature of the corresponding target person image of the confirmation target person request is obtained.
The algorithm for extracting target signature is referred to the detailed description in step 101, and details are not described herein.
It should be noted that the target spy that step 201 respectively obtains target person image to 204 to 202, step 203
The two methods of sign, in practical applications, optional one of which.
Step 205, the corresponding real-time monitoring images of target monitoring equipment are obtained, and are identified in the real-time monitoring images
Character image obtains figure map's image set.
The step is referred to the detailed description of step 102, and details are not described herein.
Step 206, each character image concentrated for the character image, determines the characteristic point of the character image, and
The pixel value for extracting each characteristic point, obtains the first candidate feature.
Wherein, pixel value can be indicated by (Red Green Blue, RGB) three chrominance channels RGB.It is, of course, also possible to
It is indicated by extended formatting, for example, CMYK (Cyan Magenta Yellow Black, Cyan Magenta Yellow blacK).The embodiment of the present invention
It is without restriction to the color mode for indicating pixel value.
In practical applications, in order to reduce computation complexity, character image can be converted first, reduce dimension,
Pixel value is extracted again.For example, the image of 512*128 is converted to 256*64 first, then pixel value is extracted, so that number of pixels drops
It is low, computation complexity can be effectively reduced.
Step 207, each character image concentrated for the character image calculates every two characteristic point and forms the remaining of angle
String value obtains the second candidate feature.
Wherein, characteristic point includes but is not limited to: eyes, nose, ear, mouth, eyebrow, face contour composition point.
Identify that the algorithm of face is quite mature, the embodiment of the present invention is without restriction to the algorithm of use.
Specifically, any two feature pixel PiAnd PjCorresponding second candidate feature SVi,jSpecific formula for calculation such as
Under:
SVi,j=cos (θi,j) (1)
Wherein, θi,jIt is characterized pixel PiAnd PjAngle between the line of reference point respectively.
Step 208, each character image concentrated for the character image, extracts the local binarization of the character image
Information obtains third candidate feature.
Wherein, the brightness value of each pixel of image is divided into white or black by binaryzation according to threshold value, i.e. brightness value
It is set as 0 or 255.For example, the brightness value by 0 to 126 is divided into 0,255 are divided by 127 to 255.
It is appreciated that the algorithm for extracting local binarization information can be pixel value method, mean value method, histogram method etc..
Wherein, the pixel that pixel value is 0 to 126 is set as 0 black by pixel value method, and the pixel that pixel value is 127 to 255 is set as
255；Mean value method calculates the average value of the pixel value of all pixels point in image first, and pixel value is less than or equal to averagely
The pixel of value is set as 0, sets 255 for the pixel that pixel value is greater than average value；Histogram method determines in image first
Maximum two positions of pixel value, then take two the smallest pixel values of position median pixel value as threshold value, and by pixel value
Pixel less than or equal to the threshold value is set as 0, sets 255 for the pixel that pixel value is greater than the threshold value.
It is appreciated that the value of the corresponding binaryzation of each pixel of character image.
Step 209, each character image concentrated for the character image, the direction gradient for extracting the character image are straight
Square figure information, obtains the 4th candidate feature.
Wherein, histograms of oriented gradients (Histogram of Oriented Gradients, HOG) is for describing image
Local textural feature.
Specifically, firstly, dividing an image into equal-sized zonule, for example, the zonule of 20*20；Then, respectively
The gradient orientation histogram of these zonules is calculated, then forms large area by a certain number of zonules, for example, an area Ge great
Domain is made of 2*2 zonule, finally, being made of the side of entire image the feature vector of the histograms of oriented gradients in big region again
To the feature vector of histogram of gradients.To which this feature vector can uniquely describe the image.
Step 210, each character image concentrated for the character image, by first candidate feature, the second candidate
Feature, third candidate feature and the 4th candidate feature are input in the age identification model that training obtains in advance, obtain the age point
Class.
Wherein, character classification by age includes but is not limited to: 1. adult and minor, 2. the elderly and non-senile etc. are a variety of
Mode classification.
Age identification model can determine whether personage is minor/the elderly by candidate feature.It is appreciated that defeated
Enter for the first candidate feature, the second candidate feature, third candidate feature and the 4th candidate feature, exports as character classification by age.
In embodiments of the present invention, age identification model can be trained by being labelled with the image pattern of character classification by age in advance
It obtains.
Optionally, in another embodiment of the invention, training obtains above-mentioned age identification model as follows:
Sub-step A1 obtains character image sample set, each character image sample mark in the character image sample set
Has age classification.
Wherein, character image sample set can be the image that monitoring device monitors, or the network of collection
Piece etc., the embodiment of the present invention is without restriction to its.
Specifically, sample is noted as positive sample and negative sample, such as: minor and non-minor.
Sub-step A2 determines the character image for each character image sample in the character image sample set
Characteristic point, and the pixel value of each characteristic point is extracted, obtain the first candidate feature sample.
The step is referred to the detailed description of step 206, and details are not described herein.
Sub-step A3 calculates every two feature dot for each character image sample in the character image sample set
Angled cosine value obtains the second candidate feature sample.
The step is referred to the detailed description of step 207, and details are not described herein.
It is decent to extract the figure map for each character image sample in the character image sample set by sub-step A4
This local binarization information, obtains third candidate feature sample.
The step is referred to the detailed description of step 208, and details are not described herein.
It is decent to extract the figure map for each character image sample in the character image sample set by sub-step A5
This histograms of oriented gradients information, obtains the 4th candidate feature sample.
The step is referred to the detailed description of step 209, and details are not described herein.
Sub-step A6, according to the first candidate feature sample of character image sample each in the character image sample set,
Two candidate feature samples, third candidate feature sample, the 4th candidate feature sample and character classification by age are instructed using default training pattern
It gets to age identification model.
Wherein, training pattern can for RNN (Recurrent Neural Networks, Recognition with Recurrent Neural Network) model,
CNN (Convolutional Neural Network, convolutional neural networks) etc..
Specifically, firstly, for each character image sample, by the first candidate feature sample, the second candidate feature sample,
Third candidate feature sample, the 4th candidate feature sample are input in Cyclic Operation Network, the character classification by age estimated；So
Afterwards, the character classification by age of the character classification by age of character image each in character image sample set estimation and mark is compared, if more than
Or it is consistent with the character classification by age of mark equal to the character classification by age of a certain proportion of character image estimation, then training terminates, then at this time
Training pattern be age identification model；If more than or equal to a certain proportion of character image estimation character classification by age and mark
Character classification by age is inconsistent, then the parameter of adjusting training model, continues to train, until being greater than or equal to a certain proportion of character image
The character classification by age of estimation and the character classification by age of mark are consistent.
Step 211, the character image that the character classification by age is target age classification is added to candidate character image subset
In.
Wherein, target age classification is different and different according to the mode of character classification by age, for minor/non-teenage
People, target age are classified as minor, and for the elderly/non-senile, target age is classified as the elderly.
It is appreciated that sequencing is not present in the character image in candidate character image subset.
It in practical applications, can also be by character image that character classification by age is non-minor/non-senile from figure map
It is rejected in image set, so that remaining character image forms candidate tasks image subset.
Step 212, according to the target signature, determine whether the candidate character image subset includes the target person
Image.
The step is referred to the detailed description of step 104, and details are not described herein.
Step 213, if candidate's character image subset does not include the target person image, to the mobile terminal
Or the target terminal of the mobile terminal setting sends prompt messages.
Specifically, warning message can be sent to pre-set target terminal, target terminal can be mobile phone, Intelligent bracelet
Etc. equipment, warning message can also be sent to the corresponding mobile phone of preset cell-phone number.
Step 214, corresponding second real-time monitoring images of the second target monitoring equipment are obtained.
The embodiment of the present invention can determine that target person image is from the image that other target monitoring monitoring of tools obtain
It is no in the monitoring area of other target monitoring equipment.
Step 215, if second real-time monitoring images include the target person image, the second warning note is sent
Information.
Specifically, the second prompt messages can include: the second target monitoring facility information, corresponding area information, mesh
Time that mark personage occurs etc. is one or more, for prompting guardian's target person to be currently located region, helps guardian fast
Speed positioning target person.In conclusion the embodiment of the invention provides a kind of pre- anti-lost alarm method, the method packet
It includes: obtaining the corresponding target signature of target person image；The corresponding real-time monitoring images of target monitoring equipment are obtained, and identify institute
The character image in real-time monitoring images is stated, figure map's image set is obtained；According to preset personage's candidate feature from the figure map
Candidate's character image subset is determined in image set；According to the target signature, determine the candidate character image subset whether include
The target person image；If candidate's character image subset does not include the target person image, warning note is sent
Information.Identification can be carried out by age, dress and children/old man's recognition of face, judge whether children/old man is pacifying
In entire area, so that it is determined that whether children/old man is safe, the accuracy rate of identification can effectively improve.
Embodiment three
Referring to Fig. 3, it illustrates a kind of structure chart for pre- anti-lost warning device that the embodiment of the present invention three provides, tools
Body is as follows.
Data obtaining module 301, for obtaining the corresponding target signature of target person image.
Character image identification module 302, for obtaining the corresponding real-time monitoring images of target monitoring equipment, and described in identification
Character image in real-time monitoring images obtains figure map's image set.
Candidate character image determining module 303, for being concentrated according to preset personage's candidate feature from the character image
Determine candidate's character image subset.
Target person image determining module 304, for determining the candidate character image subset according to the target signature
It whether include the target person image.
First warning note module 305, if not including the target person image for the candidate character image subset,
Then send prompt messages.
In conclusion the embodiment of the invention provides a kind of pre- anti-lost warning device, described device includes: that information obtains
Modulus block, for obtaining the corresponding target signature of target person image；Character image identification module is set for obtaining target monitoring
Standby corresponding real-time monitoring images, and identify the character image in the real-time monitoring images, obtain figure map's image set；Candidate
Object image determining module determines candidate character image for concentrating according to preset personage's candidate feature from the character image
Collection；Target person image determining module, for according to the target signature, determine the candidate character image subset whether include
The target person image；First warning note module, if not including the target person for the candidate character image subset
Object image then sends prompt messages.Identification can be carried out by age, recognition of face, whether judge children under guardianship
In safety zone, so that it is determined that whether children under guardianship are safe, the accuracy rate of identification can effectively improve.
Example IV
Referring to Fig. 4, it illustrates a kind of structure chart for pre- anti-lost warning device that the embodiment of the present invention four provides, tools
Body is as follows.
Data obtaining module 401, for obtaining the corresponding target signature of target person image.Optionally, of the invention real
It applies in example, above- mentioned information obtain module 401, comprising:
Figure map's image set generates submodule 4011, for obtaining the corresponding monitoring image of target monitoring equipment, and to described
Personage in monitoring image sorts out, and obtains initial figure map's image set.
First object feature extraction submodule 4012, for receiving the target person for concentrating selection from the initial character image
Object image, and from the target person extracting target from images feature.
Confirm that target person requests submodule 4013, for obtaining the confirmation target person request of mobile terminal transmission.
Second target's feature-extraction submodule 4014, for obtaining the corresponding target person of the confirmation target person request
The target signature of object image.
Optionally, in embodiments of the present invention, above-mentioned target signature not only includes face characteristic, further includes height feature
And/or clothes color characteristic.
Character image identification module 402, for obtaining the corresponding real-time monitoring images of target monitoring equipment, and described in identification
Character image in real-time monitoring images obtains figure map's image set.
Candidate character image determining module 403, for being concentrated according to preset personage's candidate feature from the character image
Determine candidate's character image subset.Optionally, in embodiments of the present invention, above-mentioned candidate character image determining module 403, packet
It includes:
First candidate feature extracting sub-module 4031, each character image for concentrating for the character image determine
The characteristic point of the character image, and the pixel value of each characteristic point is extracted, obtain the first candidate feature.
Second candidate feature computational submodule 4032, each character image for concentrating for the character image calculate
Every two characteristic point forms the cosine value of angle, obtains the second candidate feature.
Third candidate feature extracting sub-module 4033, each character image for concentrating for the character image extract
The local binarization information of the character image, obtains third candidate feature.
4th candidate feature extracting sub-module 4034, each character image for concentrating for the character image extract
The histograms of oriented gradients information of the character image, obtains the 4th candidate feature.
Character classification by age submodule 4035, each character image for concentrating for the character image, described first is waited
Feature, the second candidate feature, third candidate feature and the 4th candidate feature is selected to be input to the age identification mould that training obtains in advance
In type, character classification by age is obtained.
Candidate character image adds submodule 4036, for being the character image of target age classification by the character classification by age
It is added in candidate character image subset.
Target person image determining module 404, for determining the candidate character image subset according to the target signature
It whether include the target person image.
First warning note module 405, if not including the target person image for the candidate character image subset,
Then send prompt messages.Optionally, above-mentioned first warning note module 405 includes:
First warning note submodule 4051, the target terminal for being arranged to the mobile terminal or the mobile terminal
Send prompt messages.
Second real-time monitoring images obtain module 406, for obtaining the prison in real time of the second target monitoring equipment corresponding second
Control image.
Second warning note module 407, if including the target person image for second real-time monitoring images,
Send the second prompt messages.
Optionally, in another embodiment of the invention, above-mentioned figure map's image set generates submodule 4011, comprising:
Sort out unit, for sorting out to the personage in the monitoring image using unsupervised clustering.
In conclusion the embodiment of the invention provides a kind of pre- anti-lost warning device, described device includes: that information obtains
Modulus block, for obtaining the corresponding target signature of target person image；Character image identification module is set for obtaining target monitoring
Standby corresponding real-time monitoring images, and identify the character image in the real-time monitoring images, obtain figure map's image set；Candidate
Object image determining module determines candidate character image for concentrating according to preset personage's candidate feature from the character image
Collection；Target person image determining module, for according to the target signature, determine the candidate character image subset whether include
The target person image；First warning note module, if not including the target person for the candidate character image subset
Object image then sends prompt messages.Identification can be carried out by age, dress and children's recognition of face, thus really
Whether safe determine children/the elderly, can effectively improve the accuracy rate of identification.
The embodiment of the invention also provides a kind of electronic equipment, comprising: processor, memory and is stored in the storage
On device and the computer program that can run on the processor, the processor are realized aforementioned pre- anti-lost when executing described program
The alarm method of mistake.
The embodiment of the invention also provides a kind of readable storage medium storing program for executing, when the instruction in the storage medium is by electronic equipment
Processor execute when so that electronic equipment is able to carry out aforementioned pre- anti-lost alarm method.
For device embodiment, since it is basically similar to the method embodiment, related so being described relatively simple
Place illustrates referring to the part of embodiment of the method.
Algorithm and display are not inherently related to any particular computer, virtual system, or other device provided herein.
Various general-purpose systems can also be used together with teachings based herein.As described above, it constructs required by this kind of system
Structure be obvious.In addition, the present invention is also not directed to any particular programming language.It should be understood that can use various
Programming language realizes summary of the invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this specification.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of the various inventive aspects,
Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the disclosed method should not be interpreted as reflecting the following intention: i.e. required to protect
Shield the present invention claims features more more than feature expressly recited in each claim.More precisely, as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, wherein each claim itself
All as a separate embodiment of the present invention.
Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment
Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any
Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed
All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
Various component embodiments of the invention can be implemented in hardware, or to run on one or more processors
Software module realize, or be implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice
Microprocessor or digital signal processor (DSP) are realized in pre- anti-lost warning device according to an embodiment of the present invention
The some or all functions of some or all components.The present invention is also implemented as executing method as described herein
Some or all device or device programs.It is such to realize that program of the invention can store computer-readable
On medium, or it may be in the form of one or more signals.Such signal can be downloaded from an internet website
It arrives, is perhaps provided on the carrier signal or is provided in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability
Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch
To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame
Claim.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (12)
1. a kind of pre- anti-lost alarm method, which is characterized in that the described method includes:
Obtain the corresponding target signature of target person image；
The corresponding real-time monitoring images of target monitoring equipment are obtained, and identify the character image in the real-time monitoring images, are obtained
To figure map's image set；
It is concentrated according to preset personage's candidate feature from the character image and determines candidate character image subset；
According to the target signature, determine whether the candidate character image subset includes the target person image；
If candidate's character image subset does not include the target person image, prompt messages are sent.
2. the method according to claim 1, wherein the target person image corresponding target signature of obtaining
Step, comprising:
The corresponding monitoring image of target monitoring equipment is obtained, and the personage in the monitoring image is sorted out, is obtained initial
Figure map's image set；
The target person image for concentrating selection from the initial character image is received, and extracts mesh from the target person image
Mark feature.
3. according to the method described in claim 2, it is characterized in that, what the personage in the monitoring image was sorted out
Step, comprising:
Personage in the monitoring image is sorted out using unsupervised clustering.
4. the method according to claim 1, wherein personage's candidate feature includes the first candidate feature, the
Two candidate features, third candidate feature and the 4th candidate feature, it is described according to preset personage's candidate feature from the figure map
The step of candidate's character image subset is determined in image set, comprising:
For each character image that the character image is concentrated, the characteristic point of the character image is determined, and extract each spy
The pixel value for levying point, obtains the first candidate feature；
For each character image that the character image is concentrated, the cosine value that every two characteristic point forms angle is calculated, obtains the
Two candidate features；
For each character image that the character image is concentrated, the local binarization information of the character image is extracted, obtains the
Three candidate features；
For each character image that the character image is concentrated, the histograms of oriented gradients information of the character image is extracted, is obtained
To the 4th candidate feature；
It is for each character image that the character image is concentrated, first candidate feature, the second candidate feature, third is candidate
Feature and the 4th candidate feature are input in the age identification model that training obtains in advance, obtain character classification by age；
The character image that the character classification by age is target age classification is added in candidate character image subset.
5. further including the method according to claim 1, wherein the target signature not only includes face characteristic
Height feature and/or clothes color characteristic.
6. the method according to claim 1, wherein the target person image corresponding target signature of obtaining
Step, comprising: obtain the confirmation target person request that mobile terminal is sent；
Obtain the target signature of the corresponding target person image of the confirmation target person request；Then, the transmission alarm mentions
The step of showing information, comprising:
The target terminal being arranged to the mobile terminal or the mobile terminal sends prompt messages.
7. according to the method described in claim 6, it is characterized in that, the method also includes:
Obtain corresponding second real-time monitoring images of the second target monitoring equipment；
If second real-time monitoring images include the target person image, the second prompt messages are sent.
8. a kind of pre- anti-lost warning device, which is characterized in that described device includes:
Data obtaining module, for obtaining the corresponding target signature of target person image；
Character image identification module for obtaining the corresponding real-time monitoring images of target monitoring equipment, and identifies the real-time prison
The character image in image is controlled, figure map's image set is obtained；
Candidate character image determining module determines candidate for concentrating according to preset personage's candidate feature from the character image
Character image subset；
Target person image determining module, for determining whether the candidate character image subset wraps according to the target signature
Containing the target person image；
First warning note module is sent if not including the target person image for the candidate character image subset
Prompt messages.
9. device according to claim 8, which is characterized in that the data obtaining module, comprising:
Figure map's image set generates submodule, for obtaining the corresponding monitoring image of target monitoring equipment, and to the monitoring image
In personage sort out, obtain initial figure map's image set；
Target's feature-extraction submodule, for receiving the target person image for concentrating selection from the initial character image, and from
The target person extracting target from images feature.
10. device according to claim 8, which is characterized in that personage's candidate feature includes the first candidate feature, the
Two candidate features, third candidate feature and the 4th candidate feature, candidate's character image determining module, comprising:
First candidate feature extracting sub-module, each character image for concentrating for the character image, determines the personage
The characteristic point of image, and the pixel value of each characteristic point is extracted, obtain the first candidate feature；
It is special to calculate every two for second candidate feature computational submodule, each character image for concentrating for the character image
The angled cosine value of dot is levied, the second candidate feature is obtained；
Third candidate feature extracting sub-module, each character image for concentrating for the character image, extracts the personage
The local binarization information of image, obtains third candidate feature；
4th candidate feature extracting sub-module, each character image for concentrating for the character image, extracts the personage
The histograms of oriented gradients information of image, obtains the 4th candidate feature；
Character classification by age submodule, each character image for being concentrated for the character image, by first candidate feature,
Two candidate features, third candidate feature and the 4th candidate feature are input in the age identification model that training obtains in advance, are obtained
Character classification by age；
Candidate character image adds submodule, for the character image that the character classification by age is target age classification to be added to time
In object image subset of choosing.
11. a kind of electronic equipment characterized by comprising
Processor, memory and it is stored in the computer program that can be run on the memory and on the processor,
It is characterized in that, the processor is realized pre- anti-lost as described in one or more in claim 1-7 when executing described program
Alarm method.
12. a kind of readable storage medium storing program for executing, which is characterized in that when the instruction in the storage medium is held by the processor of electronic equipment
When row, so that electronic equipment is able to carry out the pre- anti-lost alarm side as described in one or more in claim to a method 1-7
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810753310.9A CN109035686B (en) | 2018-07-10 | 2018-07-10 | Loss prevention alarm method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810753310.9A CN109035686B (en) | 2018-07-10 | 2018-07-10 | Loss prevention alarm method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109035686A true CN109035686A (en) | 2018-12-18 |
| CN109035686B CN109035686B (en) | 2020-11-03 |
Family
ID=64640821
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810753310.9A Active CN109035686B (en) | 2018-07-10 | 2018-07-10 | Loss prevention alarm method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109035686B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886101A (en) * | 2018-12-29 | 2019-06-14 | 江苏云天励飞技术有限公司 | Posture identification method and relevant apparatus |
| CN109887234A (en) * | 2019-03-07 | 2019-06-14 | 百度在线网络技术（北京）有限公司 | A kind of children loss prevention method, apparatus, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046876A (en) * | 2015-07-23 | 2015-11-11 | 中山大学深圳研究院 | Child safety monitoring system based on image identification |
| KR101617342B1 (en) * | 2015-06-26 | 2016-05-02 | 배상윤 | Application for caring children and method for operating the same |
| CN107239744A (en) * | 2017-05-15 | 2017-10-10 | 深圳奥比中光科技有限公司 | Monitoring method, system and the storage device of human body incidence relation |
| CN107423674A (en) * | 2017-05-15 | 2017-12-01 | 广东数相智能科技有限公司 | A kind of looking-for-person method based on recognition of face, electronic equipment and storage medium |
| CN107845234A (en) * | 2017-11-27 | 2018-03-27 | 浙江卓锐科技股份有限公司 | A kind of anti-anti- method of wandering away of system and scenic spot of wandering away in scenic spot |
-
2018
- 2018-07-10 CN CN201810753310.9A patent/CN109035686B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101617342B1 (en) * | 2015-06-26 | 2016-05-02 | 배상윤 | Application for caring children and method for operating the same |
| CN105046876A (en) * | 2015-07-23 | 2015-11-11 | 中山大学深圳研究院 | Child safety monitoring system based on image identification |
| CN107239744A (en) * | 2017-05-15 | 2017-10-10 | 深圳奥比中光科技有限公司 | Monitoring method, system and the storage device of human body incidence relation |
| CN107423674A (en) * | 2017-05-15 | 2017-12-01 | 广东数相智能科技有限公司 | A kind of looking-for-person method based on recognition of face, electronic equipment and storage medium |
| CN107845234A (en) * | 2017-11-27 | 2018-03-27 | 浙江卓锐科技股份有限公司 | A kind of anti-anti- method of wandering away of system and scenic spot of wandering away in scenic spot |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886101A (en) * | 2018-12-29 | 2019-06-14 | 江苏云天励飞技术有限公司 | Posture identification method and relevant apparatus |
| CN109887234A (en) * | 2019-03-07 | 2019-06-14 | 百度在线网络技术（北京）有限公司 | A kind of children loss prevention method, apparatus, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109035686B (en) | 2020-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11783589B2 (en) | License plate detection and recognition system | |
| CN105404860B (en) | Method and apparatus for managing personal information of wandering away | |
| CN109558783A (en) | A kind of violation detection method, system and equipment for construction site | |
| CN109309808A (en) | A kind of monitoring system and method based on recognition of face | |
| CN106295567A (en) | The localization method of a kind of key point and terminal | |
| CN106022317A (en) | Face identification method and apparatus | |
| CN106372662A (en) | Helmet wearing detection method and device, camera, and server | |
| CN108986245A (en) | Work attendance method and terminal based on recognition of face | |
| US11804071B2 (en) | Method for selecting images in video of faces in the wild | |
| CN106897659A (en) | The recognition methods of blink motion and device | |
| CN109672863A (en) | A kind of construction personnel's safety equipment intelligent monitoring method based on image recognition | |
| CN111914812A (en) | Image processing model training method, device, equipment and storage medium | |
| CN110688893A (en) | Detection method for wearing safety helmet, model training method and related device | |
| WO2005024707A1 (en) | Apparatus and method for feature recognition | |
| CN110139067A (en) | A kind of wild animal monitoring data management information system | |
| JP2022552754A (en) | IMAGE DETECTION METHOD AND RELATED DEVICE, DEVICE, STORAGE MEDIUM, AND COMPUTER PROGRAM | |
| CN107909683A (en) | Realize method, terminal device and the computer-readable recording medium of boarding | |
| CN110659391A (en) | Video detection method and device | |
| CN105844245A (en) | Fake face detecting method and system for realizing same | |
| US8311271B2 (en) | Image processing apparatus and image processing method | |
| CN109670517A (en) | Object detection method, device, electronic equipment and target detection model | |
| CN208689267U (en) | A kind of intelligent glasses Focus tracking device and intelligent glasses | |
| CN109035686A (en) | A kind of pre- anti-lost alarm method and device | |
| CN114612813A (en) | Identity recognition method, model training method, device, equipment and storage medium | |
| CN109637123A (en) | A kind of complexity traffic environment downlink people living things feature recognition and traffic control system |
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 |