CN109033973A - Monitoring and early warning method, apparatus and electronic equipment - Google Patents
Monitoring and early warning method, apparatus and electronic equipment Download PDFInfo
- Publication number
- CN109033973A CN109033973A CN201810675775.7A CN201810675775A CN109033973A CN 109033973 A CN109033973 A CN 109033973A CN 201810675775 A CN201810675775 A CN 201810675775A CN 109033973 A CN109033973 A CN 109033973A
- Authority
- CN
- China
- Prior art keywords
- face
- monitoring image
- region
- monitoring
- feature vector
- 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
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- 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
Abstract
The present invention provides a kind of monitoring and early warning method, apparatus and electronic equipments, are related to monitoring technology field, this method comprises: obtaining monitoring image to be detected；Determine whether the quantity of face in monitoring image is greater than colleague's number threshold value；If the quantity of face is greater than colleague's number threshold value in monitoring image, spatial identification is carried out to the face in monitoring image；If multiple faces in monitoring image belong to identical region, early warning is carried out.Monitoring and early warning method, apparatus provided in an embodiment of the present invention and electronic equipment trigger early warning after localized crowd occurs, monitoring efficiency can be improved, and play the effect that warning and prevention are broken laws and commit crime, significantly reduce incidence of cases.
Description
Technical field
The present invention relates to monitoring technology fields, more particularly, to a kind of monitoring and early warning method, apparatus and electronic equipment.
Background technique
In typical security protection scene, crime personnel can carry out clique's work because of the factors such as affiliation or fellow-villager's relationship
Case causes means of crime and crime source to show certain region.For example, the crime means and case of a certain crime of entering the room
Part time of origin is similar, personnel source place high superposed of committing a crime.
Existing method is to look into detective work progress by screening the conventional screens such as personnel's occupancy information afterwards, even if monitoring device
It is very perfect, such as various picture pick-up devices in security protection scene etc., but variant region is underused in monitoring
Similitude of the crowd in dressing and face carry out anticipation analysis, also do not deployed to ensure effective monitoring and control of illegal activities in time to people at highest risk, high-risk
Crowd cannot trigger warning when appearing in suspicion place, so that the effect of prevention illegal activity is poor.
For above-mentioned monitor mode in the prior art, monitoring efficiency is low and prevents the problem of delinquent effect difference, at present
Not yet put forward effective solutions.
Summary of the invention
In view of this, can be improved the purpose of the present invention is to provide a kind of monitoring and early warning method, apparatus and electronic equipment
Monitoring efficiency plays the effect that warning and prevention are broken laws and commit crime.
In a first aspect, the embodiment of the invention provides a kind of monitoring and early warning methods, comprising: obtain monitoring figure to be detected
Picture；Determine whether the quantity of face in the monitoring image is greater than colleague's number threshold value；If face in the monitoring image
Quantity is greater than colleague's number threshold value, carries out spatial identification to the face in the monitoring image；If the monitoring image
In multiple faces belong to identical region, carry out early warning.
Further, if the quantity of face is greater than colleague's number threshold value in the monitoring image, to the prison
It includes: to carry out recognition of face to the monitoring image, and carry out suspect's face matching that face in control image, which carries out spatial identification,；
If the quantity of face is greater than colleague's number threshold value in successful match and the monitoring image, in the monitoring image
Face carries out spatial identification.
Further, if the quantity of face is greater than colleague's number threshold value in the monitoring image, to the prison
The face controlled in image carries out spatial identification further include: obtains the geographical location information and/or temporal information of the monitoring image；
When judging whether it is Different high risk sites by the geographical location information, and/or judging whether it is high-incidence by the temporal information
Section；If it is Different high risk sites and/or high incidence period, and in the monitoring image, the quantity of face is greater than colleague's number threshold
Value carries out spatial identification to the face in the monitoring image.
Further, the face in the monitoring image carries out spatial identification, comprising: carries out to the monitoring image
Face characteristic extracts, to obtain face feature vector；The region face characteristic obtained according to the face feature vector and in advance
Vector carries out spatial identification.
Further, the method also includes: acquisition belongs to multiple facial images of target area；Pass through face recognition algorithms
Feature extraction is carried out to the facial image respectively, obtains the corresponding initial characteristics vector of each facial image；According to each
A initial characteristics vector determines the region face feature vector of the target area.
Further, the region face feature vector that the target area is determined according to each initial characteristics vector
The step of, comprising: calculate in each described initial characteristics vector and each initial characteristics vector other are described initial
The sum of the distance of feature vector；Using the corresponding initial characteristics vector of the minimum value of the sum as the ground of the target area
Domain face feature vector.
Further, the step of whether quantity of face is greater than colleague's number threshold value in the determination monitoring image, packet
It includes: by Face datection algorithm, extracting effective face in the monitoring image；Judge whether the quantity of the face is greater than together
Pedestrian's number threshold value.
Further, the region face feature vector obtained according to the face feature vector and in advance carries out region knowledge
Other step, comprising: for each of monitoring image face, calculate the face feature vector of each face and pre-
The distance of each region face feature vector first obtained；According to the corresponding region people of minimum range in calculated distance
Face feature vector determines the region of the face.
Further, the corresponding region face feature vector of minimum range according in calculated distance determines institute
The step of stating the region of face, comprising: obtain the corresponding average distance of the target area vector；Calculate the average distance with
The product of coefficient of diminution；Judge whether the face feature vector is less than the product at a distance from the target area vector；
If so, the corresponding region of the target area vector to be determined as to the region of the face.
Further, the method also includes: obtain the geographical location information of the monitoring image；Pass through the geographical location
Information judges whether it is Different high risk sites；If so, improving warning level and carrying out early warning.
Further, the method also includes: obtain the temporal information of the monitoring image；Judged by the temporal information
It whether is high incidence period；If so, improving warning level and carrying out early warning.
Further, the method also includes: whether belong to identical region according to multiple faces in the monitoring image
Result, the monitoring image place whether be the result of Different high risk sites, when whether the time of the monitoring image is high-incidence
Section as a result, determining the alert level of the monitoring image；Early warning is carried out according to the alert level.
Further, the step of progress early warning includes: that warning information is sent to designated equipment to carry out early warning；It is described
Warning information includes: the geographical location information, described of the monitoring image, the face of the monitoring image, the monitoring image
At least one of the temporal information of monitoring image and the spatial identification result of the monitoring image.
Second aspect, the embodiment of the invention provides a kind of monitoring early-warning devices, comprising: obtain module, for obtain to
The monitoring image of detection；Quantity determining module, for determining whether the quantity of face in the monitoring image is greater than colleague's number
Threshold value；Spatial identification module, if the quantity for face in the monitoring image is greater than colleague's number threshold value, to described
Face in monitoring image carries out spatial identification；Warning module, if multiple faces in the monitoring image belong to phase
Same region carries out early warning.
The third aspect the embodiment of the invention provides a kind of electronic equipment, including memory, processor and is stored in described
On memory and the computer program that can run on the processor, the processor are realized when executing the computer program
The step of method described in above-mentioned first aspect.
Fourth aspect, the embodiment of the invention provides a kind of computer readable storage medium, the computer-readable storage
Computer program is stored on medium, the computer program executes method described in above-mentioned first aspect when being run by processor
The step of.
The embodiment of the invention provides a kind of monitoring and early warning method, apparatus and electronic equipments, can determine in monitoring image
Whether the quantity of face is greater than colleague's number threshold value, carries out spatial identification to above-mentioned face in the case where being greater than threshold value, thus
It determines region belonging to the face in monitoring image, if multiple faces in monitoring image belong to identical region, carries out
Early warning, the above method trigger early warning after localized crowd occurs, monitoring efficiency can be improved, and play warning and the illegal criminal of prevention
The effect of crime, significantly reduces incidence of cases.
Other feature and advantage of the disclosure will illustrate in the following description, alternatively, Partial Feature and advantage can be with
Deduce from specification or unambiguously determine, or by implement the disclosure above-mentioned technology it can be learnt that.
To enable the above objects, features, and advantages of the disclosure to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art be briefly described, it should be apparent that, it is described below
Attached drawing is some embodiments of the present invention, for those of ordinary skill in the art, before not making the creative labor
It puts, is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of structural schematic diagram of processing equipment provided in an embodiment of the present invention；
Fig. 2 is a kind of flow chart of monitoring and early warning method provided in an embodiment of the present invention；
Fig. 3 is a kind of flow chart of the clustering method of region face feature vector provided in an embodiment of the present invention；
Fig. 4 is a kind of flow chart of alert level determination method provided in an embodiment of the present invention；
Fig. 5 is a kind of structural block diagram of monitoring early-warning device provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Technical solution be described, it is clear that described embodiments are some of the embodiments of the present invention, rather than whole implementation
Example.
In view of people is implemented in existing criminal offence, there are localized features, and Video Supervision Technique can not be directed at present
The region of personnel carries out effective early warning, causes the effect for preventing illegal activity poor, and to improve this problem, the present invention is implemented
A kind of monitoring and early warning method, apparatus and electronic equipment that example provides, below describe to the embodiment of the present invention in detail.
Embodiment one:
Firstly, describing the processing equipment 100 for realizing the embodiment of the present invention referring to Fig.1, which can be used
In the method for operation various embodiments of the present invention.
As shown in Figure 1, processing equipment 100 includes one or more processors 102, one or more memories 104, input
Device 106, output device 108 and image acquisition device 110, the company that these components pass through bus system 112 and/or other forms
The interconnection of connection mechanism (not shown).It should be noted that the component and structure of processing equipment 100 shown in FIG. 1 are only exemplary, rather than
Restrictive, as needed, the processing equipment also can have other assemblies and structure.
The processor 102 can use digital signal processor (DSP), field programmable gate array (FPGA), can compile
At least one of journey logic array (PLA) and ASIC (Application Specific Integrated Circuit) are hard
Part form realizes that the processor 102 can be central processing unit (CPU), graphics processing unit (GPU) or have number
According to the processing unit of processing capacity and/or the other forms of instruction execution capability, and can control in the processing equipment 100
Other components to execute desired function.
The memory 104 may include one or more computer program products, and the computer program product can be with
Including various forms of computer readable storage mediums, such as volatile memory and/or nonvolatile memory.It is described volatile
Property memory for example may include random access memory (RAM) and/or cache memory (cache) etc..It is described non-easy
The property lost memory for example may include read-only memory (ROM), hard disk, flash memory etc..On the computer readable storage medium
It can store one or more computer program instructions, processor 102 can run described program instruction, described below to realize
The embodiment of the present invention in the client functionality (realized by processor) and/or other desired functions.In the calculating
Various application programs and various data can also be stored in machine readable storage medium storing program for executing, such as the application program is used and/or produced
Raw various data etc..
The input unit 106 can be the device that user is used to input instruction, and may include keyboard, mouse, wheat
One or more of gram wind and touch screen etc..
The output device 108 can export various information (for example, image or sound) to external (for example, user), and
It and may include one or more of display, loudspeaker etc..
Described image collector 110 is used to be monitored the acquisition of image, for example, image acquisition device can be from monitor video
The monitoring image can also be stored in the memory 104 for other by middle acquisition monitoring image, then, image acquisition device
Component uses.
Illustratively, the processing equipment for realizing monitoring and early warning method according to an embodiment of the present invention may be implemented as
The intelligent terminals such as video camera, capture machine smart phone, tablet computer, computer, server.
Embodiment two:
Referring to a kind of flow chart of monitoring and early warning method shown in Fig. 2, specifically comprise the following steps:
Step S202 obtains monitoring image to be detected.
Before being monitored image and obtaining, the processing equipment in above-described embodiment can be connected with existing monitoring device
It connects, carries out personnel monitoring using existing monitoring device, and obtain monitor video from the monitoring device.Obtaining above-mentioned monitoring
After video, the picture for obtaining single frames in video flowing, i.e. monitoring image can be parsed by video flowing.Wherein it is possible to according to reality
Need to be arranged the time interval or frequency for obtaining single frames.
Step S204, determines whether the quantity of face in above-mentioned monitoring image is greater than colleague's number threshold value.If so, executing
Step S206；If it is not, then terminating process.
Personnel from different geographical have a degree of similitude in face feature, are based on areal relation clique
That commits a crime is multiple, needs to judge whether the several personnel appeared in same monitoring image belong to identical region.Above-mentioned judgement
Precondition be in a monitoring image there are quantity be greater than colleague's number threshold value face.The wherein same pedestrian
Number threshold value can be determined according to the illegal activity type for being intended to early warning, such as certain type of illegal activity criminal activity is four mostly
The above collective's crime of five people, then can set 3 or 4 for colleague's number threshold value, i.e., the quantity 3 of face in monitoring image
Or when 4, just need to carry out subsequent spatial identification.The upper example of act is connect, if the number of face is less than 3 or 4 in monitoring image
When, then it is unsatisfactory for the condition into spatial identification, terminates process.
The above-mentioned process for judging face quantity in monitoring image and whether being greater than threshold value, can be used neural network and extracts face
Mode carry out, such as: by Face datection algorithm, extract effective face in monitoring image, then judge that the quantity of face is
It is no to be greater than colleague's number threshold value.Wherein it is possible to which using existing Face datection convolutional neural networks algorithm, satisfaction can be extracted
The requirement of effective face in image can be used for example Densebox algorithm and carry out the detection of face quantity.
Step S206, if the quantity of face is greater than colleague's number threshold value in above-mentioned monitoring image, in monitoring image
Face carries out spatial identification.
Due to the crowd from different geographical, there is a degree of similitude, i.e., using the phase in face feature
Spatial identification is carried out to face like property.For example, clustering can be carried out to different geographical face in advance, it is poly- according to high dimensional feature
Class and eigencenterization generate the face characteristic of a centralization, and the face characteristic based on the centralization can be to monitoring image
In face carry out spatial identification.The region can be prefecture-level city, county-level city in administrative division etc., be also possible to geographic significance
On region, such as outside the Pass, Huainan or the south of the Five Ridges etc..
Step S208 carries out early warning if multiple faces in above-mentioned monitoring image belong to identical region.
If multiple faces of same monitoring image belong to identical region, need to carry out early warning.Alarm mode
It can be and remind the identical region of the current crowd of monitoring personnel, and can also prompt where its affiliated region was.It needs herein
Bright is, it is not required that the face in same monitoring image belongs to identical region, i.e., the quantity of above-mentioned multiple faces with
The quantity of face is not required for necessary identical in above-mentioned monitoring image, can be less than or equal to the number of face in monitoring image
Amount, as long as such as more than two faces belong to identical region, carry out early warning.Carrying out early warning can execute in the following manner:
Warning information is sent to designated equipment to carry out early warning.The warning information includes: monitoring image, the face of monitoring image, prison
Control at least one of the geographical location information of image, the temporal information of monitoring image and spatial identification result of monitoring image.
In the case where which region the case cradle of all kinds of cases known be, the region that above-mentioned face can be belonged to
It is further identified, determines whether its affiliated region belongs to the case cradle of certain class case, existed if belonging to higher
Incidence of criminal offenses possibility, indicating risk can be carried out.
Above-mentioned monitoring and early warning method provided in an embodiment of the present invention can determine whether the quantity of face in monitoring image is big
In colleague's number threshold value, spatial identification is carried out to above-mentioned face in the case where being greater than threshold value, so that it is determined that in monitoring image
Region described in face carries out early warning, the above method is on ground if multiple faces in monitoring image belong to identical region
Domain crowd triggers early warning after occurring, and monitoring efficiency can be improved, and plays the effect that warning and prevention are broken laws and commit crime, effectively drops
Low incidence of cases.
Spatial identification is carried out to the face in monitoring image, it can be based on extraction face characteristic and characterization face region characteristic
The mode that is compared of feature carry out, such as can execute in the following manner:
(1) face characteristic extraction is carried out to monitoring image, to obtain face feature vector.
Face characteristic, which extracts, can use existing face recognition algorithms, such as recognition of face convolutional neural networks or residual error
Network carries out the calibration of feature point to face and carries out character representation, i.e. feature vector to face characteristic value.Generally change spy
Sign vector is high dimensional feature vector, and each high dimensional feature vector represents a face, European between two high dimensional feature vectors
Distance can indicate the similarity between two different faces.
(2) the region face feature vector obtained according to above-mentioned face feature vector and in advance carries out spatial identification.
When carrying out spatial identification, face feature vector can be carried out with the region face feature vector that in advance obtain pair
Than.It can be understood that the region face feature vector obtained in advance may include it is multiple, respectively represented the people of different geographical
The face feature of member.Above-mentioned spatial identification process can pass through the Europe of calculating face feature vector and region face feature vector
Formula distance carries out, minimum with the Euclidean distance of which region face feature vector, then it represents that the face feature vector belongs to the ground
A possibility that domain face feature vector corresponding region, is maximum.
Above-mentioned monitoring and early warning method provided in an embodiment of the present invention, the ground that face feature vector can be used and obtain in advance
Domain face feature vector carries out spatial identification to above-mentioned face, so that it is determined that region described in the face in monitoring image, if
Multiple faces in monitoring image belong to identical region, then carry out early warning.
In the above-mentioned methods, determine whether that triggering carries out face spatial identification with the quantity of face in monitoring image, then need
Carry out greater number of face spatial identification process, it is contemplated that, can also be upper to the needs that known suspect is monitored
State the identification process increased in method to known suspect.Known suspect is identified in monitoring image and face quantity meets
When condition, then carry out spatial identification process.Above-mentioned steps S206 may comprise steps of: carry out face knowledge to monitoring image
Not, and suspect's face matching is carried out；If the quantity of face is greater than colleague's number threshold value in successful match and monitoring image, to prison
The face controlled in image carries out spatial identification.Wherein, suspicion face can be obtained and be stored in advance, carry out the matching of suspicion face
When, compare the face and suspicion face that recognition of face in monitoring image obtains, obtain whether matched result.
Since in high-risk place or high-risk period, the probability that criminal offence occurs is higher, can also be in above-mentioned side
Increase judgement to high-risk place, high-risk period in method, by high-risk place, the judgement of high-risk period and face quantity collectively as
Triggering carries out the condition of spatial identification.Above-mentioned steps S206 may comprise steps of: obtain the geographical location letter of monitoring image
Breath and/or temporal information；Different high risk sites are judged whether it is by geographical location information, and/or are judged whether by temporal information
For high incidence period；If it is Different high risk sites and/or high incidence period, and in monitoring image, the quantity of face is greater than colleague's number threshold
Value carries out spatial identification to the face in monitoring image.
In the present embodiment, the extraction of the region face feature vector of case cradle can be carried out in advance, such as can be with
It is carried out by way of cluster.A kind of flow chart of the clustering method of region face feature vector shown in Figure 3, this method
It can be applied to above-mentioned processing equipment, this method specifically comprises the following steps:
Step S302, acquisition belong to multiple facial images of target area.
It can be by existing monitoring device or other approach, such as the mode acquired on the spot, to target area personnel
Carry out man face image acquiring.Above-mentioned target area can be common crime personnel source place.
Step S304 carries out feature extraction to facial image by face recognition algorithms respectively, obtains each facial image
Corresponding initial characteristics vector.
The face recognition algorithms used in this step can be identical as the face recognition algorithms in the above method, can also be with
Difference, the present embodiment are without limitation.
Step S306 determines the region face feature vector of target area according to each initial characteristics vector.
After crowd's progress face characteristic extraction of each target area, region cluster is carried out to face characteristic, and count
Calculate current cluster centre point vector.For example, by calculating the distance between every two face characteristic, and to each face characteristic with
All distance summations between other face characteristics, find the minimum value in all summations, the corresponding face characteristic of the minimum value
That is centralization face characteristic (current cluster centre point vector).Such as it can execute in the following manner:
(1) other initial characteristics vectors in each initial characteristics vector and above-mentioned each initial characteristics vector are calculated
The sum of distance.
After the initial characteristics vector for the multiple faces for obtaining certain target area, arbitrarily select an initial characteristics to
Amount, calculates the initial characteristics vector at a distance from other remaining all initial characteristics vectors, and each distance summation is obtained
And value；An initial characteristics vector (different from the initial characteristics vector of aforementioned selection) is arbitrarily selected again, calculates the initial characteristics
Vector obtains the sum value at a distance from other remaining all initial characteristics vectors, and by each distance summation；It obtains each initial
Feature vector at a distance from other initial characteristics vectors and value after, it is more each and value size, determine the smallest and value.
(2) the corresponding initial characteristics vector of the minimum value of sum is determined as center face vector.
For example, the facial image of acquisition shares 100, then can extract to obtain 100 face feature vectors.Calculate face
It at a distance from 99 face feature vectors of feature vector 1 and other, and sums, obtains result 1；Calculate face feature vector 2 and its
The distance of his 99 face feature vectors, and sum, obtain result 2；The rest may be inferred.Again comparison result 1, result 2 ... result
100, the smallest result (being assumed to be result 55) is selected, then changes face characteristic centered on face feature vector 55.
(3) using center face vector as the region face feature vector of target area.
It should be understood that by above-mentioned (2) and (3) only a kind of example described separately.In another example, can directly by
(1) the corresponding initial characteristics vector of the multiple minimum values in being calculated in is as region face feature vector.
After obtaining above-mentioned region face feature vector, the region face feature vector can also be calculated and other are each just
Average distance between beginning feature vector can be obtained with value divided by the number of summation by above-mentioned summation.The average distance
Indicate the average distance for belonging to the face corresponding feature vector and region face feature vector of identical region.By above-mentioned region people
Face feature vector, average distance are corresponding with the title of region to be stored, and can be loaded onto memory Hash table when in use.
For ease of understanding, it present embodiments provides above-mentioned according to face feature vector and region face feature vector progress ground
The concrete mode of domain identification is as follows:
(1) it for each of monitoring image face, calculates the face feature vector of each face and obtains in advance each
The distance of a region face feature vector.Distance herein can be Euclidean distance.Face feature vector and region face characteristic
The distance of vector can indicate the corresponding facial image of face feature vector face figure corresponding with region face feature vector
The similarity degree of picture, it is maximum apart from the smallest similarity degree.
(2) region of face is determined according to the corresponding region face feature vector of minimum range in calculated distance.
In face feature vector minimum at a distance from some region face feature vector of new facial image, can also further sentence
Breaking, whether it really belongs to the region, such as can pass through the distance above-mentioned average distance corresponding with region face feature vector
Relationship determine.During the region face feature vector of above-mentioned determining target area, obtained distance minimum and value and this
With the ratio for the number for being worth corresponding distance, i.e., the ratio that region face feature vector is corresponding and value is with summation apart from quantity,
I.e. above-mentioned average distance.It such as can execute in the following way: obtain the corresponding average distance of target area vector；It calculates flat
The product of equal distance and coefficient of diminution；Judge whether face feature vector is less than product at a distance from the vector of target area；If
It is that vector corresponding region in target area is determined as to the region of face.
For example, some corresponding average distance of region vector is n, when a new face and the region vector distance most
Closely, when and distance of meeting is less than n*m, wherein m is coefficient of diminution and m < 1, it is believed that the face belongs to the region, that is, there is region suspicion
It doubts.Above-mentioned coefficient of diminution configure according to actual use, such as be configurable to 0.8 equal numerical value.
Since certain places or place belong to the high-incidence place of criminal offence, consider to implement people's in the above method
On the basis of localized feature, the geographical location of monitoring image can also be further considered.In one embodiment, above-mentioned side
Method can with the following steps are included:
(1) geographical location information of monitoring image is obtained.Geographical location information can combine and collect the monitoring image
The position of monitoring device determines, such as determines the geographical location information of monitoring image simultaneously in video flowing resolution phase, can make
With GPS (Global Positioning System, global positioning system) information.
(2) Different high risk sites are judged whether it is by geographical location information.Different high risk sites can be by monitoring purpose and monitoring pair
The different flexible settings of elephant, such as the behavior of pilferage, door and window before and after supermarket doorway, cell enclosure wall, market etc. can be determined
For Different high risk sites.
(3) if so, improving warning level and carrying out early warning.If multiple faces belong in the same manner in above-mentioned monitoring image
Domain, and the geographical location of monitoring image is Different high risk sites again, then improves alert level, and carry out early warning.
Since daily certain periods belong to the criminal offence high-incidence time, the ground for implementing people is considered in the above method
On the basis of the feature of domain, the acquisition time factor of monitoring image can also be further considered.In another embodiment, on
The method of stating can with the following steps are included:
(1) temporal information of monitoring image is obtained.
(2) high incidence period is judged whether it is by temporal information.High incidence period can also be by monitoring purpose and monitored object
Different flexible settings, such as be determined as high incidence period for the behavior of pilferage, when can be by morning 3-5.
(3) if so, improving warning level and carrying out early warning.If multiple faces belong in the same manner in above-mentioned monitoring image
Domain, and the physical time of monitoring image is high incidence period again, then improves alert level, and carry out early warning.
It should be noted that can in summary in regional information, location information and temporal information at least two,
To warning content carry out comprehensive judgement after, generate final alert levels and carry out early warning again, can be notified after early warning policeman into
Row security patrol, or warning device of the starting in camera side carry out phonetic warning and expel.
In conclusion monitoring and early warning method provided in this embodiment, can compare monitoring image in real time, and according to
Region, geographical location, time with administrative staff carry out comprehensive alert level and study and judge and alarm, and play warning and crime prevention
Incidence of cases can be effectively reduced in important function.
A kind of flow chart of alert level determination method shown in Figure 4, this method can be applied to above-mentioned processing equipment,
This method specifically comprises the following steps:
Step S402 is monitored video flowing parsing, obtains monitoring image.
Step S404 carries out Face datection to monitoring image.
Step S406 carries out feature extraction to monitoring image when there are multiple faces.It is above-mentioned that face inspection is carried out to image
It surveys and the step of feature extraction, referring to the content of preceding method, details are not described herein.
Step S408 carries out regional ascription judgement according to the feature of extraction.When carrying out regional ascription judgement, what is compared is
Face characteristic in monitoring image and the region face characteristic obtained in advance, which is usually certain
The affiliated region of kind of criminal offence perpetrator, i.e., if regional ascription can be found, then it represents that the personnel belong to it is higher can
The region that can implement above-mentioned behavior, continues to execute step S410；If not finding regional ascription, then it is assumed that danger level is low.
Step S410 carries out colleague's judgement according to above-mentioned regional ascription.The region of face in same monitoring image is returned
Category is compared, and determines whether that the regional ascription in the presence of at least two faces is same region.
Step S412 carries out place judgement according to above-mentioned monitoring image.
Step S414 carries out time judgement according to above-mentioned monitoring image.
Step S416, weighted calculation vigilance coefficient, determines alert level.
Whether can belong to the result of identical region according to multiple faces in monitoring image, the place of monitoring image is
It is no be the results of Different high risk sites, monitoring image time whether be high incidence period as a result, determine the alert level of monitoring image,
And early warning is carried out according to alert level.Such as combine the result COMPREHENSIVE CALCULATING warning that colleague determines, place determines and the time determines
Coefficient.If discovery belongs to identical region, then promotes alert level, such as belongs to crime Different high risk sites, then promote alert level, such as send out
Life then promotes alert level in case high incidence period.It is understood that can determine for colleague, place determines and the time is sentenced
Different weight coefficients is respectively set in fixed result, such as by the above results for when being, value 1 takes when the above results are no
Value is 0, then respectively multiplied by its weight coefficient, then is summed, it can obtains comprehensive vigilance coefficient.It can preset different
The corresponding alert levels of vigilance coefficient and processing mode can correspond to progress early warning and execution obtaining vigilance coefficient.
It should be noted that the embodiment of the present invention to the execution of S412, S414 and S416 sequence without limitation, can use
Arbitrary sequencing executes S412, S414 and S416, or is performed simultaneously.
Embodiment three:
For monitoring and early warning method provided in embodiment two, the embodiment of the invention provides a kind of monitoring and early warning dresses
It sets, a kind of structural block diagram of monitoring early-warning device shown in Figure 5, comprising:
Module 502 is obtained, for obtaining monitoring image to be detected；
Quantity determining module 504, for determining whether the quantity of face in the monitoring image is greater than colleague's number threshold value；
Spatial identification module 506, if the quantity for face in the monitoring image is greater than colleague's number threshold value,
Spatial identification is carried out to the face in the monitoring image；
Warning module 508 carries out early warning if belonging to identical region for multiple faces in the monitoring image.
Above-mentioned monitoring early-warning device provided in an embodiment of the present invention can determine whether the quantity of face in monitoring image is big
In colleague's number threshold value, spatial identification is carried out to above-mentioned face in the case where being greater than threshold value, so that it is determined that in monitoring image
Region described in face carries out early warning if multiple faces in monitoring image belong to identical region, and the above method exists
Localized crowd triggers early warning after occurring, and monitoring efficiency can be improved, and plays the effect that warning and prevention are broken laws and commit crime, effectively
Reduce incidence of cases.
In one embodiment, above-mentioned spatial identification module is also used to: being carried out recognition of face to monitoring image, and is carried out
The matching of suspicion face；If the quantity of face is greater than colleague's number threshold value in successful match and monitoring image, in monitoring image
Face carry out spatial identification.
In another embodiment, above-mentioned spatial identification module is also used to: obtaining the geographical location information of monitoring image
And/or temporal information；Different high risk sites are judged whether it is by geographical location information, and/or are judged whether it is by temporal information
High incidence period；If it is Different high risk sites and/or high incidence period, and in monitoring image, the quantity of face is greater than colleague's number threshold value,
Spatial identification is carried out to the face in monitoring image.
In another embodiment, above-mentioned spatial identification module is also used to: face characteristic extraction is carried out to monitoring image,
To obtain face feature vector；The region face feature vector obtained according to face feature vector and in advance carries out spatial identification.
In one embodiment, above-mentioned apparatus further include: region face feature vector cluster module belongs to for acquiring
Multiple facial images of target area；By face recognition algorithms feature extraction is carried out to facial image respectively, obtains each individual
The corresponding initial characteristics vector of face image；The region face feature vector of target area is determined according to each initial characteristics vector.
In another embodiment, it is initial special to be also used to calculate each for above-mentioned region face feature vector cluster module
Levy vector at a distance from other initial characteristics vectors in each initial characteristics vector and；The minimum value of sum is corresponding initial
Region face feature vector of the feature vector as target area.
In another embodiment, above-mentioned quantity determining module is also used to extract monitoring figure by Face datection algorithm
The effective face as in；Judge whether the quantity of face is greater than colleague's number threshold value.
In another embodiment, above-mentioned spatial identification module is also used to: for each of monitoring image face, meter
The distance of each region face feature vector calculating the face feature vector of each face and obtaining in advance；According to what is be calculated
The corresponding region face feature vector of minimum range in distance determines the region of face.
In another embodiment, above-mentioned spatial identification module is also used to obtain the corresponding average departure of target area vector
From；Calculate the product of average distance and coefficient of diminution；Judge whether face feature vector is less than at a distance from the vector of target area
The product；If so, vector corresponding region in target area to be determined as to the region of face.
In one embodiment, above-mentioned apparatus further include: place determination module, for obtaining the geographical position of monitoring image
Confidence breath；Different high risk sites are judged whether it is by geographical location information；If so, improving warning level and carrying out early warning.
In one embodiment, above-mentioned apparatus further include: time determination module, the time for obtaining monitoring image believe
Breath；High incidence period is judged whether it is by temporal information；If so, improving warning level and carrying out early warning.
In another embodiment, above-mentioned apparatus further include: alert level determining module, for according in monitoring image
Multiple faces whether belong to the result of identical region, monitoring image place whether be Different high risk sites result, monitoring figure
Whether the time of picture is high incidence period as a result, determining the alert level of monitoring image；Early warning is carried out according to alert level.
In another embodiment, above-mentioned warning module is also used to: warning information is sent to designated equipment to carry out
Early warning；Warning information include: monitoring image, the face of monitoring image, the geographical location information of monitoring image, monitoring image when
Between at least one of information and the spatial identification result of monitoring image.Device provided by the present embodiment, realization principle and
The technical effect of generation is identical with previous embodiment, and to briefly describe, Installation practice part does not refer to place, can refer to aforementioned
Corresponding contents in embodiment of the method.
In addition, present embodiments provide a kind of electronic equipment, including memory, processor and storage are on a memory and can
The computer program run on a processor, processor realize side provided by above-described embodiment two when executing computer program
Method.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description
Specific work process, can be with reference to the corresponding process in previous embodiment, and details are not described herein.
Further, a kind of computer readable storage medium is present embodiments provided, is deposited on the computer readable storage medium
The step of containing computer program, method provided by above-described embodiment two executed when computer program is run by processor.
The present embodiment additionally provides a kind of computer program, which can store beyond the clouds or local storage
On medium.When the computer program is run by computer or processor for executing method provided by above-described embodiment two
Corresponding steps.
The computer program product of a kind of monitoring and early warning method, apparatus and processing equipment provided by the embodiment of the present invention,
Computer readable storage medium including storing program code, the instruction that said program code includes can be used for executing front side
Method method as described in the examples, specific implementation can be found in embodiment of the method, and details are not described herein.If the function is with software
The form of functional unit is realized and when sold or used as an independent product, can store in a computer-readable storage
In medium.Based on this understanding, technical solution of the present invention substantially in other words the part that contributes to existing technology or
The part of person's technical solution can be embodied in the form of software products, which is stored in a storage
In medium, including some instructions are used so that a computer equipment (can be personal computer, server or network are set
It is standby etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.And storage medium above-mentioned includes: USB flash disk, moves
Dynamic hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access
Memory), the various media that can store program code such as magnetic or disk.
Finally, it should be noted that embodiment described above, only a specific embodiment of the invention, to illustrate the present invention
Technical solution, rather than its limitations, scope of protection of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, those skilled in the art should understand that: anyone skilled in the art
In the technical scope disclosed by the present invention, it can still modify to technical solution documented by previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features；And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of technical solution of the embodiment of the present invention, should all cover in protection of the invention
Within the scope of.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (16)
1. a kind of monitoring and early warning method characterized by comprising
Obtain monitoring image to be detected；
Determine whether the quantity of face in the monitoring image is greater than colleague's number threshold value；
If in the monitoring image quantity of face be greater than colleague's number threshold value, to the face in the monitoring image into
Row spatial identification；
If multiple faces in the monitoring image belong to identical region, early warning is carried out.
2. if the method according to claim 1, wherein the quantity of face is greater than in the monitoring image
Colleague's number threshold value, carrying out spatial identification to the face in the monitoring image includes:
Recognition of face is carried out to the monitoring image, and carries out suspect's face matching；
If the quantity of face is greater than colleague's number threshold value in successful match and the monitoring image, to the monitoring image
In face carry out spatial identification.
3. if the method according to claim 1, wherein the quantity of face is greater than in the monitoring image
Colleague's number threshold value carries out spatial identification to the face in the monitoring image further include:
Obtain the geographical location information and/or temporal information of the monitoring image；
Different high risk sites are judged whether it is by the geographical location information, and/or judge whether it is high by the temporal information
Send out the period；
If it is Different high risk sites and/or high incidence period, and in the monitoring image, the quantity of face is greater than colleague's number threshold
Value carries out spatial identification to the face in the monitoring image.
4. according to the method in any one of claims 1 to 3, which is characterized in that the people in the monitoring image
Face carries out spatial identification, comprising:
Face characteristic extraction is carried out to the monitoring image, to obtain face feature vector；
The region face feature vector obtained according to the face feature vector and in advance carries out spatial identification.
5. according to the method described in claim 4, it is characterized in that, the method also includes:
Acquisition belongs to multiple facial images of target area；
By face recognition algorithms feature extraction is carried out to the facial image respectively, it is corresponding to obtain each facial image
Initial characteristics vector；
The region face feature vector of the target area is determined according to each initial characteristics vector.
6. according to the method described in claim 5, it is characterized in that, described according to each initial characteristics vector determination
The step of region face feature vector of target area, comprising:
Calculate other described initial characteristics vectors in each described initial characteristics vector and each initial characteristics vector
Distance sum；
Using the corresponding initial characteristics vector of the minimum value of the sum as the region face feature vector of the target area.
7. method according to claim 1-6, which is characterized in that face in the determination monitoring image
Whether quantity is greater than the step of colleague's number threshold value, comprising:
By Face datection algorithm, effective face in the monitoring image is extracted；
Judge whether the quantity of the face is greater than colleague's number threshold value.
8. method according to claim 1-7, which is characterized in that described according to the face feature vector and pre-
The region face feature vector first obtained carries out the step of spatial identification, comprising:
For each of monitoring image face, calculates the face feature vector of each face and obtain in advance each
The distance of a region face feature vector；
The region of the face is determined according to the corresponding region face feature vector of minimum range in calculated distance.
9. according to the method described in claim 8, it is characterized in that, the minimum range pair according in calculated distance
The step of region face feature vector answered determines the region of the face, comprising:
Obtain the corresponding average distance of the target area vector；
Calculate the product of the average distance and coefficient of diminution；
Judge whether the face feature vector is less than the product at a distance from the target area vector；
If so, the corresponding region of the target area vector to be determined as to the region of the face.
10. -9 described in any item methods according to claim 1, which is characterized in that the method also includes:
Obtain the geographical location information of the monitoring image；
Different high risk sites are judged whether it is by the geographical location information；
If so, improving warning level and carrying out early warning.
11. -10 described in any item methods according to claim 1, which is characterized in that the method also includes:
Obtain the temporal information of the monitoring image；
High incidence period is judged whether it is by the temporal information；
If so, improving warning level and carrying out early warning.
12. -11 described in any item methods according to claim 1, which is characterized in that the method also includes:
Whether belong to the result of identical region according to multiple faces in the monitoring image, the place of the monitoring image is
No is the result of Different high risk sites, whether the time of the monitoring image is high incidence period as a result, determining the monitoring image
Alert level；
Early warning is carried out according to the alert level.
13. -12 described in any item methods according to claim 1, which is characterized in that the step of progress early warning includes:
Warning information is sent to designated equipment to carry out early warning；The warning information includes: the monitoring image, the monitoring
The face of image, the geographical location information of the monitoring image, the temporal information of the monitoring image and the monitoring image
At least one of spatial identification result.
14. a kind of monitoring early-warning device characterized by comprising
Module is obtained, for obtaining monitoring image to be detected；
Quantity determining module, for determining whether the quantity of face in the monitoring image is greater than colleague's number threshold value；
Spatial identification module, if the quantity for face in the monitoring image is greater than colleague's number threshold value, to described
Face in monitoring image carries out spatial identification；
Warning module carries out early warning if belonging to identical region for multiple faces in the monitoring image.
15. a kind of electronic equipment, including memory, processor and it is stored on the memory and can transports on the processor
Capable computer program, which is characterized in that the processor is realized in claim 1 to 13 when executing the computer program appoints
The step of method described in one.
16. a kind of computer readable storage medium, computer program, feature are stored on the computer readable storage medium
It is, when the computer program is run by processor the step of 1 to 13 described in any item methods of perform claim requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810675775.7A CN109033973B (en) | 2018-06-26 | 2018-06-26 | Monitoring and early warning method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810675775.7A CN109033973B (en) | 2018-06-26 | 2018-06-26 | Monitoring and early warning method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109033973A true CN109033973A (en) | 2018-12-18 |
| CN109033973B CN109033973B (en) | 2020-11-03 |
Family
ID=64610955
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810675775.7A Active CN109033973B (en) | 2018-06-26 | 2018-06-26 | Monitoring and early warning method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109033973B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783540A (en) * | 2019-01-08 | 2019-05-21 | 武汉烽火众智数字技术有限责任公司 | It is a kind of based on condition code to the analysis method and system of special group personnel |
| CN109858365A (en) * | 2018-12-28 | 2019-06-07 | 深圳云天励飞技术有限公司 | A kind of special population Assembling Behavior analysis method, device and electronic equipment |
| CN110263633A (en) * | 2019-05-13 | 2019-09-20 | 广州烽火众智数字技术有限公司 | The personnel that are involved in drug traffic based on space time correlation detect method for early warning, system and storage medium |
| CN110633915A (en) * | 2019-09-24 | 2019-12-31 | 北京明略软件系统有限公司 | High-risk place identification method and device |
| CN111382626A (en) * | 2018-12-28 | 2020-07-07 | 广州市百果园信息技术有限公司 | Method, device and equipment for detecting illegal image in video and storage medium |
| CN111770279A (en) * | 2020-08-03 | 2020-10-13 | 维沃移动通信有限公司 | Shooting method and electronic equipment |
| CN113702833A (en) * | 2021-06-30 | 2021-11-26 | 中国电信集团工会上海市委员会 | Corridor battery car monitoring system and method |
| CN114612959A (en) * | 2022-01-28 | 2022-06-10 | 北京深睿博联科技有限责任公司 | Face recognition system and method for assisting blind person in interpersonal communication |
| CN113702833B (en) * | 2021-06-30 | 2024-04-19 | 中国电信集团工会上海市委员会 | Corridor storage battery car monitoring system and method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101883258A (en) * | 2009-05-08 | 2010-11-10 | 上海弘视通信技术有限公司 | Violent crime detection system and detection method thereof |
| CN103246869A (en) * | 2013-04-19 | 2013-08-14 | 福建亿榕信息技术有限公司 | Crime monitoring method based on face recognition technology and behavior and sound recognition |
| CN106503622A (en) * | 2016-09-26 | 2017-03-15 | 北京格灵深瞳信息技术有限公司 | A kind of vehicle antitracking method and device |
| KR101745598B1 (en) * | 2017-01-17 | 2017-06-09 | 주식회사 경림이앤지 | Security system and method for detecting abnormal behavior |
| CN107358158A (en) * | 2017-06-07 | 2017-11-17 | 浙江大华技术股份有限公司 | A kind of clique's crime method for early warning and device |
| CN107480246A (en) * | 2017-08-10 | 2017-12-15 | 北京中航安通科技有限公司 | A kind of recognition methods of associate people and device |
-
2018
- 2018-06-26 CN CN201810675775.7A patent/CN109033973B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101883258A (en) * | 2009-05-08 | 2010-11-10 | 上海弘视通信技术有限公司 | Violent crime detection system and detection method thereof |
| CN103246869A (en) * | 2013-04-19 | 2013-08-14 | 福建亿榕信息技术有限公司 | Crime monitoring method based on face recognition technology and behavior and sound recognition |
| CN106503622A (en) * | 2016-09-26 | 2017-03-15 | 北京格灵深瞳信息技术有限公司 | A kind of vehicle antitracking method and device |
| KR101745598B1 (en) * | 2017-01-17 | 2017-06-09 | 주식회사 경림이앤지 | Security system and method for detecting abnormal behavior |
| CN107358158A (en) * | 2017-06-07 | 2017-11-17 | 浙江大华技术股份有限公司 | A kind of clique's crime method for early warning and device |
| CN107480246A (en) * | 2017-08-10 | 2017-12-15 | 北京中航安通科技有限公司 | A kind of recognition methods of associate people and device |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111382626B (en) * | 2018-12-28 | 2023-04-18 | 广州市百果园信息技术有限公司 | Method, device and equipment for detecting illegal image in video and storage medium |
| CN109858365A (en) * | 2018-12-28 | 2019-06-07 | 深圳云天励飞技术有限公司 | A kind of special population Assembling Behavior analysis method, device and electronic equipment |
| CN111382626A (en) * | 2018-12-28 | 2020-07-07 | 广州市百果园信息技术有限公司 | Method, device and equipment for detecting illegal image in video and storage medium |
| CN109858365B (en) * | 2018-12-28 | 2021-03-05 | 深圳云天励飞技术有限公司 | Special crowd gathering behavior analysis method and device and electronic equipment |
| CN109783540A (en) * | 2019-01-08 | 2019-05-21 | 武汉烽火众智数字技术有限责任公司 | It is a kind of based on condition code to the analysis method and system of special group personnel |
| CN110263633A (en) * | 2019-05-13 | 2019-09-20 | 广州烽火众智数字技术有限公司 | The personnel that are involved in drug traffic based on space time correlation detect method for early warning, system and storage medium |
| CN110263633B (en) * | 2019-05-13 | 2023-08-04 | 广州烽火众智数字技术有限公司 | Method, system and storage medium for detecting and early warning of toxic person based on space-time correlation |
| CN110633915A (en) * | 2019-09-24 | 2019-12-31 | 北京明略软件系统有限公司 | High-risk place identification method and device |
| CN111770279B (en) * | 2020-08-03 | 2022-04-08 | 维沃移动通信有限公司 | Shooting method and electronic equipment |
| CN111770279A (en) * | 2020-08-03 | 2020-10-13 | 维沃移动通信有限公司 | Shooting method and electronic equipment |
| CN113702833A (en) * | 2021-06-30 | 2021-11-26 | 中国电信集团工会上海市委员会 | Corridor battery car monitoring system and method |
| CN113702833B (en) * | 2021-06-30 | 2024-04-19 | 中国电信集团工会上海市委员会 | Corridor storage battery car monitoring system and method |
| CN114612959A (en) * | 2022-01-28 | 2022-06-10 | 北京深睿博联科技有限责任公司 | Face recognition system and method for assisting blind person in interpersonal communication |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109033973B (en) | 2020-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109033973A (en) | Monitoring and early warning method, apparatus and electronic equipment | |
| CN105376255B (en) | A kind of Android platform intrusion detection method based on K-means cluster | |
| RU2393544C2 (en) | Method and device to detect flame | |
| US9852342B2 (en) | Surveillance system | |
| CN103839346B (en) | A kind of intelligent door and window anti-intrusion device and system, intelligent access control system | |
| CN109299646B (en) | Crowd abnormal event detection method, device, system and storage medium | |
| CN108229335A (en) | It is associated with face identification method and device, electronic equipment, storage medium, program | |
| US20180278894A1 (en) | Surveillance system | |
| CN110895861B (en) | Abnormal behavior early warning method and device, monitoring equipment and storage medium | |
| US20090195382A1 (en) | Video sensor and alarm system and method with object and event classification | |
| US20050036658A1 (en) | Non-motion detection | |
| CN108965826A (en) | Monitoring method, device, processing equipment and storage medium | |
| CN106716382A (en) | Methods and systems for aggregated multi-application behavioral analysis of mobile device behaviors | |
| Lee et al. | ArchCam: Real time expert system for suspicious behaviour detection in ATM site | |
| WO2019089441A1 (en) | Exclusion zone in video analytics | |
| CN110189355A (en) | Safe escape channel occupies detection method, device, electronic equipment and storage medium | |
| CN107330414A (en) | Act of violence monitoring method | |
| CN112153373A (en) | Fault identification method and device for bright kitchen range equipment and storage medium | |
| CN109815839A (en) | Hover personal identification method and Related product under micro services framework | |
| KR20220000172A (en) | An apparatus and a system for providing a security surveillance service based on edge computing and a method for operating them | |
| WO2018210039A1 (en) | Data processing method, data processing device, and storage medium | |
| CN112927178B (en) | Occlusion detection method, occlusion detection device, electronic device, and storage medium | |
| CN111091047B (en) | Living body detection method and device, server and face recognition equipment | |
| CN105447998B (en) | A kind of method and device of definite paper money mouth foreign matter | |
| KR20220000216A (en) | An apparatus for providing a security surveillance service based on deep learning distributed processing |
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 |