CN110321852A - A kind of type of action recognition methods, device, storage medium and computer equipment - Google Patents
A kind of type of action recognition methods, device, storage medium and computer equipment Download PDFInfo
- Publication number
- CN110321852A CN110321852A CN201910604202.XA CN201910604202A CN110321852A CN 110321852 A CN110321852 A CN 110321852A CN 201910604202 A CN201910604202 A CN 201910604202A CN 110321852 A CN110321852 A CN 110321852A
- Authority
- CN
- China
- Prior art keywords
- action
- type
- human
- violation
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Alarm Systems (AREA)
Abstract
A kind of type of action recognition methods, device, storage medium and computer equipment provided in an embodiment of the present invention, can obtain the monitoring objective image of image acquisition device；By human action recognizer, the human action in the monitoring objective image is obtained；According to preset human action typelib, type of action belonging to the human action is determined；When the type of action is violation type, violation information is generated.By using the human action in identification monitoring objective image, the technological means of violation information is generated after determining that the type of action of human body movement is violation type, it overcomes and manually checks that image is difficult to the technical issues of finding the human action of violation type in time, and then reached the technical effect of the human action of timely discovery violation type.
Description
Technical field
The present invention relates to field of image recognition more particularly to a kind of type of action recognition methods, device, storage medium and meters
Calculate machine equipment.
Background technique
It actively is properly completed task in order to guarantee that employee is interior at work, does not violate the regulations and discipline of company
Rule, company need to employee at work in human action be monitored, to find the affiliated violation type of employee
Human action.
Occurs the human action of affiliated violation type within the work hours for employee, each company needs to arrange supervisor
The image for checking image acquisition device, the people by checking the image of image acquisition device, in artificial judgment image
Whether type of action belonging to body movement belongs to violation type.
However, the image due to needing manually to check all image acquisition devices, and judge in image one by one
Whether human action belongs to the human action of violation type, and therefore, it is difficult to find that the human body of the affiliated violation type of employee is dynamic in time
Make.
Summary of the invention
In view of the above problems, the present invention provides a kind of one for overcoming the above problem or at least being partially solved the above problem
Kind type of action recognition methods, device, storage medium and computer equipment, technical solution are as follows:
A kind of type of action recognition methods, comprising:
Obtain the monitoring objective image of image acquisition device；
By human action recognizer, the human action in the monitoring objective image is obtained；
According to preset human action typelib, type of action belonging to the human action is determined；
When the type of action is violation type, violation information is generated.
Optionally, the method also includes:
When the type of action belonging to the human action is violation type, face inspection is carried out to the monitoring objective image
It surveys, obtains the face characteristic of promoter corresponding with the human action；
The face characteristic is matched with preset information bank, is obtained and the matched initiation of the face characteristic
The identity information of person.
Optionally, the violation information includes moving belonging to the identity information and/or the human action of the promoter
Make type.
Optionally, described that human action in the monitoring objective image is obtained by human action recognizer, packet
It includes:
By the human action recognizer, the human body key point in the monitoring objective image is determined；
According to the human body key point, human action corresponding with the human body key point is determined.
Optionally, the method is applied in control equipment, the control equipment and multiple images acquisition device communication link
It connecing, the monitoring objective image is collected by described image acquisition device,
It is described when the type of action be violation type when, generate violation information, comprising:
Determining type of action belonging to the human action in the monitoring objective image of each image acquisition device respectively is
No is violation type；
It is true according to the corresponding default weight of each image collecting device and the corresponding violation type of each image collecting device
Determine as a result, determining whether type of action belonging to the human action in monitoring objective image is violation type, if it is, generating
Violation information.
Optionally, the method also includes:
When the type of action belonging to the human action is violation type, each image acquisition device is determined
Monitoring objective image in the human action the location of promoter；
Face is carried out to the monitoring objective image of image acquisition device corresponding with the location of the promoter
Detection, obtains the face characteristic of the promoter；
The face characteristic is matched with preset information bank, obtains the identity information of the promoter.
A kind of type of action identification device, comprising: image acquiring unit, human action obtaining unit, type of action determine
Unit and violation information generation unit,
Described image obtaining unit, for obtaining the monitoring objective image of image acquisition device；
The human action obtaining unit, for obtaining in the monitoring objective image by human action recognizer
Human action；
The type of action determination unit, for determining the human action institute according to preset human action typelib
The type of action of category；
The violation information generation unit, when the type of action is violation type, for generating violation information.
Optionally, described device further include: face characteristic obtaining unit and identity information obtaining unit,
The face characteristic obtaining unit, when the type of action belonging to the human action is violation type, for pair
The monitoring objective image carries out Face datection, obtains the face characteristic of promoter corresponding with the human action；
The identity information obtaining unit, for the face characteristic to be matched with preset information bank, obtain with
The identity information of the matched promoter of face characteristic.
A kind of storage medium is stored with computer executable instructions in the storage medium, and the computer is executable to be referred to
When order is loaded and executed by processor, above-mentioned type of action recognition methods is realized.
A kind of computer equipment, including processor, memory and be stored on the memory and can be in the processor
The program of upper operation, the processor at least perform the steps of when executing program
Obtain the monitoring objective image of image acquisition device；
By human action recognizer, the human action in the monitoring objective image is obtained；
According to preset human action typelib, type of action belonging to the human action is determined；
When the type of action is violation type, violation information is generated.
By above-mentioned technical proposal, a kind of type of action recognition methods, device, storage medium and calculating provided by the invention
Machine equipment can obtain the monitoring objective image of image acquisition device；By human action recognizer, the prison is obtained
Control the human action in target image；According to preset human action typelib, movement class belonging to the human action is determined
Type；When the type of action is violation type, violation information is generated.By using the human body in identification monitoring objective image
Movement generates the technological means of violation information after determining that the type of action of human body movement is violation type, overcomes artificial
It checks that image is difficult to the technical issues of finding the human action of violation type in time, and then has reached timely discovery violation type
The technical effect of human action.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 shows a kind of flow diagram of type of action recognition methods provided in an embodiment of the present invention；
Fig. 2 shows the flow diagrams of another type of action recognition methods provided in an embodiment of the present invention；
Fig. 3 shows the distribution schematic diagram of human body key point provided in an embodiment of the present invention；
Fig. 4 shows the flow diagram of another type of action recognition methods provided in an embodiment of the present invention；
Fig. 5 shows a kind of pair of human action progress frame provided in an embodiment of the present invention and selects what is marked to illustrate schematic diagram；
What Fig. 6 showed a kind of image collecting device installation site provided in an embodiment of the present invention illustrates schematic diagram；
Fig. 7 shows a kind of structural schematic diagram of type of action identification device provided in an embodiment of the present invention；
Fig. 8 shows the structural schematic diagram of another type of action identification device provided in an embodiment of the present invention；
Fig. 9 shows the structural schematic diagram of another type of action identification device provided in an embodiment of the present invention.
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
It is fully disclosed to those skilled in the art.
As shown in Figure 1, a kind of type of action recognition methods provided in an embodiment of the present invention, may include:
S100, the monitoring objective image for obtaining image acquisition device；
Specifically, image collecting device may include panoramic camera, monitor camera and the meter for carrying camera function
Calculation machine and mobile device.Image collecting device can acquire the image of continuous multiple frames as monitoring objective image, can also acquire
A certain frame image is as monitoring objective image.Monitoring objective image may include being in the human body of static movement and in mobile dynamic
The human body of work.It is understood that may include one or more human bodies in the monitoring objective image obtained.
S200, pass through human action recognizer, obtain the human action in the monitoring objective image；
Specifically, human action recognizer may include human body attitude algorithm for estimating, limb action recognizer, human body
Critical point detection algorithm.It is understood that the present embodiment can obtain one or more in same frame monitoring objective image
Human action, and the human action obtained in same frame monitoring objective image can be static human action, can also be with
It is mobile human action.
Optionally, it is based on method shown in FIG. 1, as shown in Fig. 2, another kind type of action provided in an embodiment of the present invention is known
Other method, step S200 may include:
S210, by the human action recognizer, determine the human body key point in the monitoring objective image；
Specifically, one embodiment of the present invention mainly passes through human body critical point detection algorithm, monitoring objective image is determined
In human body key point.Human body critical point detection algorithm can be according to the human body half body or Whole Body in monitoring objective image
Skeletal joint point determines human body key point.As shown in figure 3, white dot is indicated according to the human body in monitoring objective image in figure
Skeletal joint point determined by human body key point, a human body passes through human body critical point detection there are multiple human body key points
Algorithm can determine the human body key point of some or all of human body.It should be noted that the human body of the embodiment of the present invention is crucial
Point detection algorithm can also distinguish between the human body key point of different human body, and if there are two human bodies in Fig. 3, the embodiment of the present invention can be with area
Divide the key point of the two human bodies.
In practical applications, it may appear that the case where more than two human bodies are overlapped in monitoring objective image, this hair
Bright embodiment can mark ID by the human body key point in advance to each human body and/or using the human body key point of different colours
And/or the mode using different human body key point symbols, distinguish the human body key point of different human body.
S220, according to the human body key point, determine human action corresponding with the human body key point.
Specifically, according to the distribution situation of the human body key point of some or all of the same human body, the embodiment of the present invention
It can determine the corresponding human action of human body key point of the human body.On the left of Fig. 3 shown in human body, closed according to the human body of the human body
The distribution situation of key point, the embodiment of the present invention can determine the human action of the human body to stand.On the right side of Fig. 3 shown in human body,
According to the distribution situation of the human body key point of the human body, the embodiment of the present invention can determine the human action of the human body for running.
The embodiment of the present invention can also determine other human actions of the human body, example according to the distribution situation of the human body key point of human body
Such as: human action of smoking, make a phone call and take pictures.The embodiment of the present invention can also be crucial by the overlapping human body of different human bodies
Point determines the human action interacted between the different human body.Such as the human actions such as shake hands, push and shove.
S300, according to preset human action typelib, determine type of action belonging to the human action；
Specifically, preset human action typelib includes the corresponding type of action of a variety of human actions, type of action can
To include normal type, violation type and other types.For example, the human actions such as be seated, stand may belong to normal type, inhale
Cigarette holds mobile phone and claps the human actions such as computer screen and may belong to violation type.Human action and violation for normal type
The human action of type can be determining with application environment according to an embodiment of the present invention, for example, when application environment is gas station, then
The human bodies behavior such as smoke, make a phone call may belong to violation type；When the department that application environment is need for confidentiality, then takes pictures and wait human bodies
Behavior may belong to violation type；When application environment is factory, the human bodies behavior such as run, push and shove may belong to violation type.
The embodiment of the present invention can also count special time period other than specifying certain human actions to belong to violation type
Interior, a certain human body makes number and/or the duration of a certain human action, and time of human body movement is made according to the human body
The several and/or duration determines that the human body makes whether human body movement belongs to violation type.For example, in the place of smoking ban, from
The human action that the smoking of human body is obtained in monitoring objective image, since there is the human action of smoking lift hand to put down to mouth again
The feature of hand, what is obtained when putting down hand may not be the human action of smoking, in order to avoid human body will accidentally be lifted to hand to mouth
The non-smoking human action in portion is determined as the human action of smoking, and therefore, the embodiment of the present invention can be calculated in specific time
The human body makes the number for lifting the human action for the feature that hand puts down hand to mouth again in section, when the number is more than preset times
When, it is determined that the human body makes the human action of smoking, and otherwise, the human action that the human body is made is not confirmed as smoking.Root
Number and/or duration of the human action made according to human body in special time period, determine whether human body movement belongs to
Violation type, it is possible to reduce the erroneous judgement to the human action of violation type improves accuracy rate.
S400, when the type of action be violation type when, generate violation information.
Specifically, violation information can be expressed by forms such as text, screenshot and data.Generate violation information it
Afterwards, the embodiment of the present invention can store violation information into memory, and be believed in violation of rules and regulations in preset exception database this
Breath is recorded, this record includes generation time, place, the device number of image collecting device and the violation class of the violation information
Type.
A kind of type of action recognition methods provided in an embodiment of the present invention, can obtain the monitoring of image acquisition device
Target image；By human action recognizer, the human action in the monitoring objective image is obtained；According to preset human body
Type of action library determines type of action belonging to the human action；When the type of action is violation type, generate in violation of rules and regulations
Information.By using the human action in identification monitoring objective image, determining the type of action of human body movement in violation of rules and regulations
The technological means that violation information is generated after type overcomes and manually checks that image is difficult to find the human action of violation type in time
The technical issues of, and then reached the technical effect of the human action of timely discovery violation type.
Optionally, it is based on method shown in FIG. 1, as shown in figure 4, another kind type of action provided in an embodiment of the present invention is known
Other method can also include:
When S500, the type of action belonging to the human action are violation type, the monitoring objective image is carried out
Face datection obtains the face characteristic of promoter corresponding with the human action；
Specifically, being carried out by face recognition technology to the monitoring objective image where the human action of affiliated violation type
Recognition of face obtains the facial feature information of the human body of human body movement.It is understood that in determining monitoring objective image
Human action belong to after violation type, it is thus necessary to determine that the human body movement promoter face characteristic.Determining the initiation
After the face characteristic of person, the face characteristic that can intercept the promoter is saved, so that related personnel checks.
S600, the face characteristic is matched with preset information bank, is obtained and the matched hair of the face characteristic
Play the identity information of person.
Specifically, preset information bank can be previously stored in the monitoring objective image including image acquisition device
Some or all of the corresponding promoter of human action including face characteristic.In preset information bank, each storage
Face characteristic match with an identity information.Identity information may include the surname of promoter corresponding with the identity information
Name, department and position.By the face characteristic of acquisition, can be inquired in preset information bank and the face characteristic
The identity information matched, so that it is determined that the identity information of the promoter of some human action.By the initiation for determining human action
The promoter of violation type human action can be accurately positioned in the identity information of person.
In embodiments of the present invention, preset information bank can be connect with preset exception database, preset information bank
The identity information of middle storage can be corresponding with one or more history violation information in preset exception database.For example,
Three there is the human action of violation type three times in this month, then the identity information of the Zhang San stored in preset information bank
Can in preset exception database according to the human action history violation information pair generated of the violation type three times
It answers.When obtaining the identity information of Zhang San, some or all of Zhang San history violation information can be obtained.
Optionally, the violation information may include belonging to the identity information and/or the human action of the promoter
Type of action.
The mode of frame choosing can be used to prison after obtaining the human action in monitoring objective image in the embodiment of the present invention
The human action of each frame in control target image is marked, and has reached the purpose of tracking human body movement.Work as human action
When being shift action, human body movement is selected by frame always in monitoring objective image in the process of moving is marked.The present invention is real
Applying example can be with real-time exhibition type of action belonging to the human action of frame choosing label.As shown in figure 5, in real time to monitoring objective
Each human action in image carries out frame choosing label, and shows type of action belonging to human action around frame choosing label.It can
It is subsequent when checking the monitoring objective image from memory after by the storage to memory of monitoring objective image with understanding,
Type of action belonging to frame choosing label and the human action shown can also be viewed.
It is the human action of violation type for affiliated type of action, the embodiment of the present invention can be around human body movement
Reminder flag is added, the frame choosing label that the human body can also be made to act is shown in red, so as to the human body with affiliated normal type
The frame choosing label of movement distinguishes.
Optionally, another type of action recognition methods provided in an embodiment of the present invention can be applied in control equipment,
The control equipment and multiple images acquisition device communicate to connect, and the monitoring objective image is acquired by described image acquisition device
It obtains, step S400 can be specifically included:
S410, movement class belonging to human action in the monitoring objective image of each image acquisition device is determined respectively
Whether type is violation type；
Each image collecting device of the embodiment of the present invention may be mounted at the different location in a certain region.As shown in fig. 6,
Image collecting device A may be mounted at the intersection of wall and wall, and image collecting device B may be mounted on wall, image collector
Setting C may be mounted in the electronic equipment in this region.It is understood that the installation site of image collecting device is can root
Determining according to actual environment and actual needs, it is not further limited herein.
Specifically, the monitoring objective image due to each image acquisition device is different to the acquisition direction of human action,
Therefore according to the same human action in the monitoring objective image of each image acquisition device, determining type of action can not
Together.For example, the human action of Zhang San can determine in the monitoring objective image of image collecting device A acquisition within the same time
To sit up straight, but the human action of Zhang San can be determined as lying down in the monitoring objective image of image collecting device B acquisition.According to
Human action in the monitoring objective image of each image acquisition device determines type of action belonging to human body movement respectively
It whether is violation type.For example, within the same time, the human body of Li Si in the monitoring objective image of image collecting device A acquisition
Movement is smoking, and the human action in the monitoring objective image of image collecting device B acquisition is to stand, then according to image collector
The monitoring objective image for setting A acquisition determines that the human action of Li Si is the human action of violation type, and according to image collector
The monitoring objective image for setting B acquisition determines that the human action of Li Si is the human action of normal type.
S420, according to the corresponding default weight of each image collecting device and the corresponding violation class of each image collecting device
Type definitive result determines whether type of action belonging to the human action in monitoring objective image is violation type, if it is,
Generate violation information.
The embodiment of the present invention can preset weight for each image collecting device respectively, when each image collecting device acquires respectively
Monitoring objective image in type of action belonging to human body movement to the determination of same human action it is inconsistent when, Ke Yigen
According to the default weight of each image collecting device, type of action belonging to human body movement is determined.
Further, when determining in the monitoring objective image that each image collecting device acquires respectively to belonging to same human action
Type of action when whether being that violation type inconsistent result occurs, can according to the default weight of each image collecting device,
Determine whether type of action belonging to human body movement is violation type.For example, determining the prison of each image acquisition device
It can be 0 or 1 that whether type of action belonging to the human action in control target image, which is the definitive result of violation type, wherein 0
It can be expressed as non-violation type, 1 can be expressed as violation type.The embodiment of the present invention can be to each image acquisition device
Monitoring objective image in human action belonging to type of action whether be that the definitive result of violation type is weighted and averaged
Operation, then the weighted average that weighted mean operation obtains is compared with preset threshold, if the weighted average is greater than
The preset threshold, it is determined that type of action belonging to human body movement is violation type, is somebody's turn to do if the weighted average is not more than
Preset threshold, it is determined that type of action belonging to human body movement is non-violation type.For example, image collecting device A's is default
Weight is 0.3, the default weight of image collecting device B is 0.5 and the default weight of image collecting device C is 0.2, to same people
Body movement: the determination of type of action belonging to the human body movement in the monitoring objective image of image collecting device A acquisition is determined
As a result it is 0, determines that the human body in the monitoring objective image of image collecting device B acquisition acts the determination of affiliated type of action
As a result the determination of type of action belonging to the human action in monitoring objective image acquired for 1 and determining image collecting device C
It as a result is 1, then weighted average is 0.7, if preset threshold is 0.5, type of action belonging to human body movement is violation class
Type.
Optionally, another type of action recognition methods provided in an embodiment of the present invention can also include:
When S700, the type of action belonging to the human action are violation type, described multiple images acquisition dress is determined
Set the location of the promoter of the human action in the monitoring objective image of acquisition；
The embodiment of the present invention can be according to picture position of the promoter of human action in monitoring objective image, and determining should
The physical location of the promoter of human action.The monitoring objective image acquired by multiple images acquisition device determines human body jointly
The physical location of the promoter of movement keeps determining physical location more accurate.
S800, the monitoring objective image of image acquisition device corresponding with the location of the promoter is carried out
Face datection obtains the face characteristic of the promoter；
Specifically, the embodiment of the present invention can pass through the image collecting device of the physical location relative close with the promoter
Face datection is carried out to the promoter in the monitoring objective image of acquisition, obtains the face characteristic of the promoter.The present invention is implemented
Example can also by the monitoring objective image of the image acquisition device of facial institute's direction of promoter to the promoter into
Row Face datection obtains the face characteristic of the promoter.
If it is determined that the promoter of the human action of violation type, on station, the embodiment of the present invention can be called on its station
Image collecting device, obtain the face characteristic of the promoter of the human action of the violation type, such as call and count on its station
Camera on calculation machine obtains the face characteristic of the promoter before the camera.
S900, the face characteristic is matched with preset information bank, obtains the identity information of the promoter.
Corresponding with above method embodiment, the embodiment of the invention also provides a kind of type of action identification devices.
As shown in fig. 7, a kind of type of action identification device provided in an embodiment of the present invention, may include: that image obtains list
Member 100, human action obtaining unit 200, type of action determination unit 300 and violation information generation unit 400,
Described image obtaining unit 100, for obtaining the monitoring objective image of image acquisition device；
Specifically, image collecting device may include panoramic camera, monitor camera and the meter for carrying camera function
Calculation machine and mobile device.Image collecting device can acquire the image of continuous multiple frames as monitoring objective image, can also acquire
A certain frame image is as monitoring objective image.Monitoring objective image may include being in the human body of static movement and in mobile dynamic
The human body of work.It is understood that may include one or more human bodies in the monitoring objective image obtained.
The human action obtaining unit 200, for obtaining the monitoring objective image by human action recognizer
In human action；
Specifically, human action recognizer may include human body attitude algorithm for estimating, limb action recognizer, human body
Critical point detection algorithm.It is understood that the present embodiment can obtain one or more in same frame monitoring objective image
Human action, and the human action obtained in same frame monitoring objective image can be static human action, can also be with
It is mobile human action.
Optionally, it is based on device shown in Fig. 7, as shown in figure 8, another kind type of action provided in an embodiment of the present invention is known
Other device, the human action obtaining unit 200 include human body key point determination unit 210 and human action determination unit 220,
The human body key point determination unit 210, for determining the monitoring mesh by the human action recognizer
Human body key point in logo image；
Specifically, one embodiment of the present invention mainly passes through human body critical point detection algorithm, monitoring objective image is determined
In human body key point.Human body critical point detection algorithm can be according to the human body half body or Whole Body in monitoring objective image
Skeletal joint point determines human body key point.As shown in figure 3, white dot is indicated according to the human body in monitoring objective image in figure
Skeletal joint point determined by human body key point, a human body passes through human body critical point detection there are multiple human body key points
Algorithm can determine the human body key point of some or all of human body.It should be noted that the human body of the embodiment of the present invention is crucial
Point detection algorithm can also distinguish between the human body key point of different human body, and if there are two human bodies in Fig. 3, the embodiment of the present invention can be with area
Divide the key point of the two human bodies.
In practical applications, it may appear that the case where more than two human bodies are overlapped in monitoring objective image, this hair
Bright embodiment can mark ID by the human body key point in advance to each human body and/or using the human body key point of different colours
And/or the mode using different human body key point symbols, distinguish the human body key point of different human body.
The human action determination unit 220, for according to the human body key point, the determining and human body key point pair
The human action answered.
Specifically, according to the distribution situation of the human body key point of some or all of the same human body, the embodiment of the present invention
It can determine the corresponding human action of human body key point of the human body.On the left of Fig. 3 shown in human body, closed according to the human body of the human body
The distribution situation of key point, the embodiment of the present invention can determine the human action of the human body to stand.On the right side of Fig. 3 shown in human body,
According to the distribution situation of the human body key point of the human body, the embodiment of the present invention can determine the human action of the human body for running.
The embodiment of the present invention can also determine other human actions of the human body, example according to the distribution situation of the human body key point of human body
Such as: human action of smoking, make a phone call and take pictures.The embodiment of the present invention can also be crucial by the overlapping human body of different human bodies
Point determines the human action interacted between the different human body.Such as the human actions such as shake hands, push and shove.
The type of action determination unit 300, for determining the human action according to preset human action typelib
Affiliated type of action；
Specifically, preset human action typelib includes the corresponding type of action of a variety of human actions, type of action can
To include normal type, violation type and other types.For example, the human actions such as be seated, stand may belong to normal type, inhale
Cigarette holds mobile phone and claps the human actions such as computer screen and may belong to violation type.Human action and violation for normal type
The human action of type can be determining with application environment according to an embodiment of the present invention, for example, when application environment is gas station, then
The human bodies behavior such as smoke, make a phone call may belong to violation type；When the department that application environment is need for confidentiality, then takes pictures and wait human bodies
Behavior may belong to violation type；When application environment is factory, the human bodies behavior such as run, push and shove may belong to violation type.
The embodiment of the present invention can also count special time period other than specifying certain human actions to belong to violation type
Interior, a certain human body makes number and/or the duration of a certain human action, and time of human body movement is made according to the human body
The several and/or duration determines that the human body makes whether human body movement belongs to violation type.For example, in the place of smoking ban, from
The human action that the smoking of human body is obtained in monitoring objective image, since there is the human action of smoking lift hand to put down to mouth again
The feature of hand, what is obtained when putting down hand may not be the human action of smoking, in order to avoid human body will accidentally be lifted to hand to mouth
The non-smoking human action in portion is determined as the human action of smoking, and therefore, the embodiment of the present invention can be calculated in specific time
The human body makes the number for lifting the human action for the feature that hand puts down hand to mouth again in section, when the number is more than preset times
When, it is determined that the human body makes the human action of smoking, and otherwise, the human action that the human body is made is not confirmed as smoking.Root
Number and/or duration of the human action made according to human body in special time period, determine whether human body movement belongs to
Violation type, it is possible to reduce the erroneous judgement to the human action of violation type improves accuracy rate.
The violation information generation unit 400, when the type of action is violation type, for generating violation information.
Specifically, violation information can be expressed by forms such as text, screenshot and data.Generate violation information it
Afterwards, the embodiment of the present invention can store violation information into memory, and be believed in violation of rules and regulations in preset exception database this
Breath is recorded, this record includes generation time, place, the device number of image collecting device and the violation class of the violation information
Type.
A kind of type of action identification device provided in an embodiment of the present invention, can obtain the monitoring of image acquisition device
Target image；By human action recognizer, the human action in the monitoring objective image is obtained；According to preset human body
Type of action library determines type of action belonging to the human action；When the type of action is violation type, generate in violation of rules and regulations
Information.By using the human action in identification monitoring objective image, determining the type of action of human body movement in violation of rules and regulations
The technological means that violation information is generated after type overcomes and manually checks that image is difficult to find the human action of violation type in time
The technical issues of, and then reached the technical effect of the human action of timely discovery violation type.
Optionally, it is based on device shown in Fig. 8, as shown in figure 9, another kind type of action provided in an embodiment of the present invention is known
Other device, can also include: face characteristic obtaining unit 500 and identity information obtaining unit 600,
The face characteristic obtaining unit 500 is used for when the type of action belonging to the human action is violation type
Face datection is carried out to the monitoring objective image, obtains the face characteristic of promoter corresponding with the human action；
Specifically, being carried out by face recognition technology to the monitoring objective image where the human action of affiliated violation type
Recognition of face obtains the facial feature information of the human body of human body movement.It is understood that in determining monitoring objective image
Human action belong to after violation type, it is thus necessary to determine that the human body movement promoter face characteristic.Determining the initiation
After the face characteristic of person, the face characteristic that can intercept the promoter is saved, so that related personnel checks.
The identity information obtaining unit 600 is obtained for matching the face characteristic with preset information bank
With the identity information of the matched promoter of the face characteristic.
Specifically, preset information bank can be previously stored in the monitoring objective image including image acquisition device
Some or all of the corresponding promoter of human action including face characteristic.In preset information bank, each storage
Face characteristic match with an identity information.Identity information may include the surname of promoter corresponding with the identity information
Name, department and position.By the face characteristic of acquisition, can be inquired in preset information bank and the face characteristic
The identity information matched, so that it is determined that the identity information of the promoter of some human action.By the initiation for determining human action
The promoter of violation type human action can be accurately positioned in the identity information of person.
In embodiments of the present invention, preset information bank can be connect with preset exception database, preset information bank
The identity information of middle storage can be corresponding with one or more history violation information in preset exception database.For example,
Three there is the human action of violation type three times in this month, then the identity information of the Zhang San stored in preset information bank
Can in preset exception database according to the human action history violation information pair generated of the violation type three times
It answers.When obtaining the identity information of Zhang San, some or all of Zhang San history violation information can be obtained.
Optionally, the violation information may include belonging to the identity information and/or the human action of the promoter
Type of action.
The mode of frame choosing can be used to prison after obtaining the human action in monitoring objective image in the embodiment of the present invention
The human action of each frame in control target image is marked, and has reached the purpose of tracking human body movement.Work as human action
When being shift action, human body movement is selected by frame always in monitoring objective image in the process of moving is marked.The present invention is real
Applying example can be with real-time exhibition type of action belonging to the human action of frame choosing label.As shown in figure 5, in real time to monitoring objective
Each human action in image carries out frame choosing label, and shows type of action belonging to human action around frame choosing label.It can
It is subsequent when checking the monitoring objective image from memory after by the storage to memory of monitoring objective image with understanding,
Type of action belonging to frame choosing label and the human action shown can also be viewed.
It is the human action of violation type for affiliated type of action, the embodiment of the present invention can be around human body movement
Reminder flag is added, the frame choosing label that the human body can also be made to act is shown in red, so as to the human body with affiliated normal type
The frame choosing label of movement distinguishes.
Optionally, the type of action identification device can acquire for control equipment, the control equipment and multiple images
Device communication connection, the monitoring objective image are collected by described image acquisition device,
The violation information generation unit 400 can be specifically used for: determine the monitoring of each image acquisition device respectively
Whether type of action belonging to the human action in target image is violation type；It is corresponding according to each image collecting device
Default weight and the corresponding violation type definitive result of each image collecting device, determine the human action institute in monitoring objective image
Whether the type of action of category is violation type, if it is, generating violation information.
Each image collecting device of the embodiment of the present invention may be mounted at the different location in a certain region.As shown in fig. 6,
Image collecting device A may be mounted at the intersection of wall and wall, and image collecting device B may be mounted on wall, image collector
Setting C may be mounted in the electronic equipment in this region.It is understood that the installation site of image collecting device is can root
Determining according to actual environment and actual needs, it is not further limited herein.
Specifically, the monitoring objective image due to each image acquisition device is different to the acquisition direction of human action,
Therefore according to the same human action in the monitoring objective image of each image acquisition device, determining type of action can not
Together.For example, the human action of Zhang San can determine in the monitoring objective image of image collecting device A acquisition within the same time
To sit up straight, but the human action of Zhang San can be determined as lying down in the monitoring objective image of image collecting device B acquisition.According to
Human action in the monitoring objective image of each image acquisition device determines type of action belonging to human body movement respectively
It whether is violation type.For example, within the same time, the human body of Li Si in the monitoring objective image of image collecting device A acquisition
Movement is smoking, and the human action in the monitoring objective image of image collecting device B acquisition is to stand, then according to image collector
The monitoring objective image for setting A acquisition determines that the human action of Li Si is the human action of violation type, and according to image collector
The monitoring objective image for setting B acquisition determines that the human action of Li Si is the human action of normal type.
The embodiment of the present invention can preset weight for each image collecting device respectively, when each image collecting device acquires respectively
Monitoring objective image in type of action belonging to human body movement to the determination of same human action it is inconsistent when, Ke Yigen
According to the default weight of each image collecting device, type of action belonging to human body movement is determined.Further, when determining that each image adopts
It whether is that violation type occurs to type of action belonging to same human action in the monitoring objective image that acquisition means acquire respectively
When inconsistent result, movement class belonging to human body movement can be determined according to the default weight of each image collecting device
Whether type is violation type.
Optionally, the type of action identification device can also include: position determination unit, face characteristic obtaining unit
500 and identity information obtaining unit 600,
When the position determination unit for the type of action belonging to the human action is violation type, institute is determined
State the location of the promoter of the human action in the monitoring objective image of each image acquisition device；
The embodiment of the present invention can be according to picture position of the promoter of human action in monitoring objective image, and determining should
The physical location of the promoter of human action.The monitoring objective image acquired by multiple images acquisition device determines human body jointly
The physical location of the promoter of movement keeps determining physical location more accurate.
The face characteristic obtaining unit 500, for image collector corresponding with the location of the promoter
The monitoring objective image for setting acquisition carries out Face datection, obtains the face characteristic of the promoter；
Specifically, the embodiment of the present invention can pass through the image collecting device of the physical location relative close with the promoter
Face datection is carried out to the promoter in the monitoring objective image of acquisition, obtains the face characteristic of the promoter.The present invention is implemented
Example can also by the monitoring objective image of the image acquisition device of facial institute's direction of promoter to the promoter into
Row Face datection obtains the face characteristic of the promoter.
If it is determined that the promoter of the human action of violation type, on station, the embodiment of the present invention can be called on its station
Image collecting device, obtain the face characteristic of the promoter of the human action of the violation type, such as call and count on its station
Camera on calculation machine obtains the face characteristic of the promoter before the camera.
The identity information obtaining unit 600 is obtained for matching the face characteristic with preset information bank
The identity information of the promoter.
A kind of storage medium provided in an embodiment of the present invention is stored with computer executable instructions in the storage medium,
When the computer executable instructions are loaded and executed by processor, above-mentioned type of action recognition methods is realized.
A kind of computer equipment provided in an embodiment of the present invention, including processor, memory and it is stored in the memory
Program that is upper and can running on the processor, the processor at least perform the steps of when executing program
Obtain the monitoring objective image of image acquisition device；
By human action recognizer, the human action in the monitoring objective image is obtained；
According to preset human action typelib, type of action belonging to the human action is determined；
When the type of action is violation type, violation information is generated.
A kind of type of action identification device includes processor and memory, and above-mentioned image acquiring unit 100, human body are dynamic
Make obtaining unit 200, type of action determination unit 300 and violation information generation unit 400 etc. to be stored in as program unit
In memory, above procedure unit stored in memory is executed by processor to realize corresponding function.
Include kernel in processor, is gone in memory to transfer corresponding program unit by kernel.Kernel can be set one
Or more, to find the human action of violation type in time by adjusting kernel parameter.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/
Or the forms such as Nonvolatile memory, if read-only memory (ROM) or flash memory (flash RAM), memory include that at least one is deposited
Store up chip.
The embodiment of the invention provides a kind of storage mediums, are stored thereon with program, real when which is executed by processor
The existing type of action recognition methods.
The embodiment of the invention provides a kind of processor, the processor is for running program, wherein described program operation
Type of action recognition methods described in Shi Zhihang.
The embodiment of the invention provides a kind of equipment, equipment include processor, memory and storage on a memory and can
The program run on a processor, processor perform the steps of when executing program
Obtain the monitoring objective image of image acquisition device；
By human action recognizer, the human action in the monitoring objective image is obtained；
According to preset human action typelib, type of action belonging to the human action is determined；
When the type of action is violation type, violation information is generated.
Optionally, can also include:
When the type of action belonging to the human action is violation type, face inspection is carried out to the monitoring objective image
It surveys, obtains the face characteristic of promoter corresponding with the human action；
The face characteristic is matched with preset information bank, is obtained and the matched initiation of the face characteristic
The identity information of person.
Optionally, described that human action in the monitoring objective image is obtained by human action recognizer, packet
It includes:
By the human action recognizer, the human body key point in the monitoring objective image is determined；
According to the human body key point, human action corresponding with the human body key point is determined.
Optionally, the method is applied in control equipment, the control equipment and multiple images acquisition device communication link
It connecing, the monitoring objective image is collected by described image acquisition device,
It is described when the type of action be violation type when, generate violation information, comprising:
Determining type of action belonging to the human action in the monitoring objective image of each image acquisition device respectively is
No is violation type；
It is true according to the corresponding default weight of each image collecting device and the corresponding violation type of each image collecting device
Determine as a result, determining whether type of action belonging to the human action in monitoring objective image is violation type, if it is, generating
Violation information.
Optionally, can also include:
When the type of action belonging to the human action is violation type, each image acquisition device is determined
Monitoring objective image in the human action the location of promoter；
Face is carried out to the monitoring objective image of image acquisition device corresponding with the location of the promoter
Detection, obtains the face characteristic of the promoter；
The face characteristic is matched with preset information bank, obtains the identity information of the promoter.
Equipment herein can be server, PC, PAD, mobile phone etc..
Present invention also provides a kind of computer program products, when executing on data processing equipment, are adapted for carrying out just
The program of beginningization there are as below methods step:
Obtain the monitoring objective image of image acquisition device；
By human action recognizer, the human action in the monitoring objective image is obtained；
According to preset human action typelib, type of action belonging to the human action is determined；
When the type of action is violation type, violation information is generated.
Optionally, can also include:
When the type of action belonging to the human action is violation type, face inspection is carried out to the monitoring objective image
It surveys, obtains the face characteristic of promoter corresponding with the human action；
The face characteristic is matched with preset information bank, is obtained and the matched initiation of the face characteristic
The identity information of person.
Optionally, described that human action in the monitoring objective image is obtained by human action recognizer, packet
It includes:
By the human action recognizer, the human body key point in the monitoring objective image is determined；
According to the human body key point, human action corresponding with the human body key point is determined.
Optionally, the method is applied in control equipment, the control equipment and multiple images acquisition device communication link
It connecing, the monitoring objective image is collected by described image acquisition device,
It is described when the type of action be violation type when, generate violation information, comprising:
Determining type of action belonging to the human action in the monitoring objective image of each image acquisition device respectively is
No is violation type；
It is true according to the corresponding default weight of each image collecting device and the corresponding violation type of each image collecting device
Determine as a result, determining whether type of action belonging to the human action in monitoring objective image is violation type, if it is, generating
Violation information.
Optionally, can also include:
When the type of action belonging to the human action is violation type, each image acquisition device is determined
Monitoring objective image in the human action the location of promoter；
Face is carried out to the monitoring objective image of image acquisition device corresponding with the location of the promoter
Detection, obtains the face characteristic of the promoter；
The face characteristic is matched with preset information bank, obtains the identity information of the promoter.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/
Or the forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable Jie
The example of matter.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including element
There is also other identical elements in process, method, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can provide as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application
Form.It is deposited moreover, the application can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The above is only embodiments herein, are not intended to limit this application.To those skilled in the art,
Various changes and changes are possible in this application.It is all within the spirit and principles of the present application made by any modification, equivalent replacement,
Improve etc., it should be included within the scope of the claims of this application.
Claims (10)
1. a kind of type of action recognition methods characterized by comprising
Obtain the monitoring objective image of image acquisition device；
By human action recognizer, the human action in the monitoring objective image is obtained；
According to preset human action typelib, type of action belonging to the human action is determined；
When the type of action is violation type, violation information is generated.
2. the method according to claim 1, wherein further include:
When the type of action belonging to the human action is violation type, Face datection is carried out to the monitoring objective image,
Obtain the face characteristic of promoter corresponding with the human action；
The face characteristic is matched with preset information bank, is obtained with the matched promoter's of the face characteristic
Identity information.
3. according to the method described in claim 2, it is characterized in that, the violation information includes the identity information of the promoter
And/or type of action belonging to the human action.
4. obtaining the prison the method according to claim 1, wherein described by human action recognizer
Control the human action in target image, comprising:
By the human action recognizer, the human body key point in the monitoring objective image is determined；
According to the human body key point, human action corresponding with the human body key point is determined.
5. the control is set the method according to claim 1, wherein the method is applied in control equipment
Standby to communicate to connect with multiple images acquisition device, the monitoring objective image is collected by described image acquisition device,
It is described when the type of action be violation type when, generate violation information, comprising:
Determine respectively type of action belonging to the human action in the monitoring objective image of each image acquisition device whether be
Violation type；
Knot is determined according to the corresponding default weight of each image collecting device and the corresponding violation type of each image collecting device
Fruit determines whether type of action belonging to the human action in monitoring objective image is violation type, if it is, generating in violation of rules and regulations
Information.
6. according to the method described in claim 5, it is characterized by further comprising:
When the type of action belonging to the human action is violation type, the prison of each image acquisition device is determined
Control the location of the promoter of the human action in target image；
Face datection is carried out to the monitoring objective image of image acquisition device corresponding with the location of the promoter,
Obtain the face characteristic of the promoter；
The face characteristic is matched with preset information bank, obtains the identity information of the promoter.
7. a kind of type of action identification device characterized by comprising image acquiring unit, human action obtaining unit, movement
Type determining units and violation information generation unit,
Described image obtaining unit, for obtaining the monitoring objective image of image acquisition device；
The human action obtaining unit, for obtaining the people in the monitoring objective image by human action recognizer
Body movement；
The type of action determination unit, for determining belonging to the human action according to preset human action typelib
Type of action；
The violation information generation unit, when the type of action is violation type, for generating violation information.
8. device according to claim 7, which is characterized in that further include: face characteristic obtaining unit and identity information obtain
Unit is obtained,
The face characteristic obtaining unit, when the type of action belonging to the human action is violation type, for described
Monitoring objective image carries out Face datection, obtains the face characteristic of promoter corresponding with the human action；
The identity information obtaining unit, for the face characteristic to be matched with preset information bank, obtain with it is described
The identity information of the matched promoter of face characteristic.
9. a kind of storage medium, which is characterized in that be stored with computer executable instructions, the computer in the storage medium
When executable instruction is loaded and executed by processor, as above type of action identification side as claimed in any one of claims 1 to 6 is realized
Method.
10. a kind of computer equipment, which is characterized in that including processor, memory and be stored on the memory and can be
The program run on the processor, the processor at least perform the steps of when executing program
Obtain the monitoring objective image of image acquisition device；
By human action recognizer, the human action in the monitoring objective image is obtained；
According to preset human action typelib, type of action belonging to the human action is determined；
When the type of action is violation type, violation information is generated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910604202.XA CN110321852B (en) | 2019-07-05 | 2019-07-05 | Action type identification method and device, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910604202.XA CN110321852B (en) | 2019-07-05 | 2019-07-05 | Action type identification method and device, storage medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110321852A true CN110321852A (en) | 2019-10-11 |
| CN110321852B CN110321852B (en) | 2022-06-14 |
Family
ID=68122765
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910604202.XA Active CN110321852B (en) | 2019-07-05 | 2019-07-05 | Action type identification method and device, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110321852B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110775930A (en) * | 2019-10-18 | 2020-02-11 | 北京地平线机器人技术研发有限公司 | Security protection device, method, computer-readable storage medium, and electronic device |
| CN111063144A (en) * | 2019-12-03 | 2020-04-24 | 联通物联网有限责任公司 | Abnormal behavior monitoring method, device, equipment and computer readable storage medium |
| CN112580543A (en) * | 2020-12-24 | 2021-03-30 | 四川云从天府人工智能科技有限公司 | Behavior recognition method, system and device |
| CN113627330A (en) * | 2021-08-10 | 2021-11-09 | 北京百度网讯科技有限公司 | Method and device for identifying target type dynamic image and electronic equipment |
| CN114596530A (en) * | 2022-03-23 | 2022-06-07 | 中国航空油料有限责任公司浙江分公司 | Airplane refueling intelligent management method and device based on non-contact optical AI |
| WO2023116863A1 (en) * | 2021-12-23 | 2023-06-29 | 广州明珞装备股份有限公司 | Optimization method and device for production line, and storage medium |
| CN113627330B (en) * | 2021-08-10 | 2024-05-14 | 北京百度网讯科技有限公司 | Method and device for identifying target type dynamic image and electronic equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109272221A (en) * | 2018-09-06 | 2019-01-25 | 济南市市中区综合行政执法局 | City management method, apparatus and electronic equipment |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855641A (en) * | 2012-08-10 | 2013-01-02 | 上海电机学院 | Fruit level classification system based on external quality |
| CN102867411B (en) * | 2012-09-21 | 2014-07-09 | 博康智能网络科技股份有限公司 | Taxi dispatching method and taxi dispatching system on basis of video monitoring system |
| JP6447424B2 (en) * | 2015-09-03 | 2019-01-09 | オムロン株式会社 | Violation suppression device and violation suppression system provided with the same |
| CN109145868A (en) * | 2018-09-11 | 2019-01-04 | 广州杰赛科技股份有限公司 | A kind of Activity recognition method and apparatus assisting running training |
| CN109726663A (en) * | 2018-12-24 | 2019-05-07 | 广东德诚科教有限公司 | Online testing monitoring method, device, computer equipment and storage medium |
| CN109447597B (en) * | 2018-12-27 | 2022-04-12 | 深圳市沃特沃德信息有限公司 | Method and device for attendance checking by multiple persons and face recognition system |
| CN109670490A (en) * | 2019-02-21 | 2019-04-23 | 无锡加视诚智能科技有限公司 | A kind of dynamic human face identification comparison method |
| CN109961037A (en) * | 2019-03-20 | 2019-07-02 | 中共中央办公厅电子科技学院(北京电子科技学院) | A kind of examination hall video monitoring abnormal behavior recognition methods |
-
2019
- 2019-07-05 CN CN201910604202.XA patent/CN110321852B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109272221A (en) * | 2018-09-06 | 2019-01-25 | 济南市市中区综合行政执法局 | City management method, apparatus and electronic equipment |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110775930A (en) * | 2019-10-18 | 2020-02-11 | 北京地平线机器人技术研发有限公司 | Security protection device, method, computer-readable storage medium, and electronic device |
| CN110775930B (en) * | 2019-10-18 | 2021-07-06 | 北京地平线机器人技术研发有限公司 | Security protection device, method, computer-readable storage medium, and electronic device |
| CN111063144A (en) * | 2019-12-03 | 2020-04-24 | 联通物联网有限责任公司 | Abnormal behavior monitoring method, device, equipment and computer readable storage medium |
| CN112580543A (en) * | 2020-12-24 | 2021-03-30 | 四川云从天府人工智能科技有限公司 | Behavior recognition method, system and device |
| CN112580543B (en) * | 2020-12-24 | 2024-04-16 | 四川云从天府人工智能科技有限公司 | Behavior recognition method, system and device |
| CN113627330A (en) * | 2021-08-10 | 2021-11-09 | 北京百度网讯科技有限公司 | Method and device for identifying target type dynamic image and electronic equipment |
| CN113627330B (en) * | 2021-08-10 | 2024-05-14 | 北京百度网讯科技有限公司 | Method and device for identifying target type dynamic image and electronic equipment |
| WO2023116863A1 (en) * | 2021-12-23 | 2023-06-29 | 广州明珞装备股份有限公司 | Optimization method and device for production line, and storage medium |
| CN114596530A (en) * | 2022-03-23 | 2022-06-07 | 中国航空油料有限责任公司浙江分公司 | Airplane refueling intelligent management method and device based on non-contact optical AI |
| CN114596530B (en) * | 2022-03-23 | 2022-11-18 | 中国航空油料有限责任公司浙江分公司 | Airplane refueling intelligent management method and device based on non-contact optical AI |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110321852B (en) | 2022-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110321852A (en) | A kind of type of action recognition methods, device, storage medium and computer equipment | |
| US9524418B2 (en) | Systems and methods for detecting, identifying and tracking objects and events over time | |
| CN104766059B (en) | Quick accurate human-eye positioning method and the gaze estimation method based on human eye positioning | |
| US10810442B2 (en) | People flow estimation device, people flow estimation method, and recording medium | |
| CN108288027A (en) | A kind of detection method of picture quality, device and equipment | |
| CN107949298A (en) | Determine the body part that user is currently nursed | |
| CN108256431A (en) | A kind of hand position identification method and device | |
| CN107341436A (en) | Gestures detection network training, gestures detection and control method, system and terminal | |
| CN111563396A (en) | Method and device for online identifying abnormal behavior, electronic equipment and readable storage medium | |
| CN110009614A (en) | Method and apparatus for output information | |
| WO2020134102A1 (en) | Article recognition method and device, vending system, and storage medium | |
| CN109508583A (en) | A kind of acquisition methods and device of distribution trend | |
| CN109840503A (en) | A kind of method and device of determining information | |
| CN110826610A (en) | Method and system for intelligently detecting whether dressed clothes of personnel are standard | |
| CN109145127A (en) | Image processing method and device, electronic equipment and storage medium | |
| CN109409962A (en) | Image processing method, device, electronic equipment, computer readable storage medium | |
| CN109189878A (en) | A kind of crowd's thermodynamic chart preparation method and device | |
| CN103946887A (en) | Gaze position estimation system, control method for gaze position estimation system, gaze position estimation device, control method for gaze position estimation device, program, and information storage medium | |
| CN110458615A (en) | Pass through the method and device of internet information assessment customer satisfaction | |
| CN108229291A (en) | Characteristic point detection, network training method, device, electronic equipment and storage medium | |
| Rusdi et al. | Student attendance using face recognition technology | |
| CN110287878A (en) | The determination method and device of marketing strategy, storage medium, electronic device | |
| Kuwahara et al. | Blink detection using image processing to predict eye fatigue | |
| US20210004573A1 (en) | Sentiment analysis | |
| WO2022024274A1 (en) | Image processing device, image processing method, and recording medium |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201130 Address after: Room 011, first floor, no.2429, Xingang East Road, Haizhu District, Guangzhou City, Guangdong Province (office only) Applicant after: CHUANGYOU digital technology (Guangdong) Co., Ltd Address before: 519000 -41072, 105 room 6, Baohua Road, Hengqin New District, Zhuhai, Guangdong (centralized office area) Applicant before: MINISO (HENGQIN) ENTERPRISE MANAGEMENT Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |