CN104318248B - Action identification method and action recognition device - Google Patents
Action identification method and action recognition device Download PDFInfo
- Publication number
- CN104318248B CN104318248B CN201410563395.6A CN201410563395A CN104318248B CN 104318248 B CN104318248 B CN 104318248B CN 201410563395 A CN201410563395 A CN 201410563395A CN 104318248 B CN104318248 B CN 104318248B
- Authority
- CN
- China
- Prior art keywords
- action
- destination object
- change information
- humidity change
- humidity
- 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.)
- Active
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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
- G06V10/955—Hardware or software architectures specially adapted for image or video understanding using specific electronic processors
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The embodiment of the present application provides a kind of action identification method and action recognition device.Methods described includes：The humidity change information at least in a seclected time section of an object space at least position where acquisition destination object；According at least to the humidity change information at least in a seclected time section, identify the destination object at least one action at least performed in a seclected time section.The method and device of each embodiment of the application identifies the action of destination object by the humidity change information in space where destination object, requires less to the environmental condition of object space, simple and easy applied widely.
Description
Technical field
Each embodiment of the application is related to action recognition technical field, more particularly to a kind of action identification method and action recognition
Device.
Background technology
It is increasing in the work and life of people to be set dependent on intelligence with advances in technology with the development in market
It is standby, for example, game/amusement equipment, monitoring device, various smart machines etc..For this kind equipment, it can accurately analyze, manage
The activity of user is solved, to observe, monitor user or be very important with user mutual, and thus, it is possible to be used for user
Such equipment belt carrys out convenience.Identification to user action will undoubtedly be with a wide range of applications, applicable scene
Including：The input of somatic sensation television game, fall detection, identification, the control of smart machine, etc..Therefore, it is necessary to which a kind of be applicable letter
Single easy and accurate action recognition scheme.
The content of the invention
In view of this, a purpose of the embodiment of the present application is to provide a kind of action recognition scheme.
To achieve the above object, according to the first aspect of the embodiment of the present application, there is provided a kind of action identification method, the side
Method includes：
The change letter of the humidity at least in a seclected time section of an object space at least position where acquisition destination object
Breath；
According at least to the humidity change information at least in a seclected time section, identify the destination object it is described extremely
At least one action performed in a few seclected time section.
According to the second aspect of the embodiment of the present application, there is provided a kind of action recognition device, described device include：
One first acquisition module, at least one for obtaining the object space at least position where destination object it is selected when
Between humidity change information in section；
One identification module, for according at least to the humidity change information at least in a seclected time section, described in identification
Destination object is at least one action at least performed in a seclected time section.
The method and device of the embodiment of the present application identifies target pair by the moisture distribution information in space where destination object
The action of elephant, less is required to the environmental condition of object space, it is simple and easy applied widely.
Brief description of the drawings
Fig. 1 is the flow chart of the action identification method of the embodiment of the present application；
Fig. 2 (a) to Fig. 2 (f) is the structure chart of the action recognition device of the embodiment of the present application；
Fig. 3 is the structure chart of the embodiment of the present application another kind action recognition device.
Embodiment
With reference to the accompanying drawings and examples, the embodiment of the present invention is described in further detail.Implement below
Example is used to illustrate the present invention, but is not limited to the scope of the present invention.
It will be understood by those skilled in the art that the term such as " first ", " second " in the application is only used for distinguishing difference and set
Standby, module or parameter etc., any particular technology implication is neither represented, also do not indicate that the inevitable logical order between them.
In each embodiment of the application, the destination object can be any movable object, for example, people or other dynamic
Thing.Object space is the whole physical space where destination object, for example, the physics where room or destination object
A certain region in space, for example, the pocket in room, for example, correspond to will in the embodiment of the present application for its size
The region of the described at least sensing range of a dew cell.The action feeling the pulse with the finger-tip mark subject's body is at least partly (for example, limb
Body) motion that occurs in certain time period, for example, destination object is waved, kicked.Object space can be because present in it
The motion of destination object and disturbance is produced in moisture distribution, different motions will corresponding different humidity changing pattern.This Shen
Please each embodiment be based on this, there is provided a kind of humidity changing pattern by space where destination object identifies that destination object moves
The scheme of work.
As shown in figure 1, the action identification method of the embodiment of the present application includes：
S120. obtain destination object where object space in an at least position the humidity at least in a seclected time section
Change information.
In the ideal case, because destination object athletic meeting produces disturbance to the moisture distribution of object space, different is dynamic
Work will produce different disturbances to object space moisture distribution, and in response to different actions, in object space at diverse location
The humidity that can be obtained is on time dimension also by the change with respective AD HOC.In order to identify the action of destination object,
Humidity change information of at least position at least in a seclected time section in object space is obtained in the step s 120.Wherein, institute
It can be any number of point or the region in object space to state an at least position, and can preferably be collected in optional object space
To the position of humidity change, for example, close to the point of destination object or the region of face destination object, such position can be obvious
Ground detects that the motion of destination object disturbs to caused by the moisture distribution of object space.Correspondingly, the humidity change information
Can be in the humidity change information to a little less time dimension being collected into；Or at least it can be collected into a region
The humidity change information of time dimension, or the humidity change information of time-space dimension.Seclected time section is also with any
A period of time of duration, it is corresponding specific with specifically action that the duration should meet that corresponding temperature change granularity identifies enough
Changing pattern.
S140. according at least to the humidity change information at least in a seclected time section, identify that the destination object exists
At least one action performed in an at least seclected time section.
As described above, spy is changed in the different humidity acted that destination object performs according to an at least position response
Sign, caused spy when destination object performs specific action can be matched in the humidity change information acquired in step S120
The temperature change of mould-fixed, and then realize the action that destination object is identified according to obtained humidity change information.
In summary, the method for the embodiment of the present application identifies target by the humidity change information in space where destination object
The action of object, because the acquisition of humidity information is not only restricted to blocking for other facilities of object space, to object space
Environmental condition requires less, simple and easy applied widely.
In the step s 120, at least in a seclected time section of an at least position can be obtained by an at least dew cell
Humidity change information, dew cell such as humistor, humicap etc..That is, step S120 may include：
S122. the humidity change information is obtained by an at least dew cell for an at least opening position.
An at least dew cell can be at least one dew cell in an at least opening position, in order to obtain essence
True humidity change information, especially obtain and embody because the motion of destination object produces the humidity change information of disturbance, it is described
An at least dew cell can be deployed in an at least opening position in an array manner, sufficiently to obtain humidity letter as far as possible
Breath.
In a kind of possible implementation, the dew cell is capacitance sensor, and an at least dew cell is
Capacitance sensor array.Capacitance sensor is a kind of sensor that measured change is converted into electric capacitance change.Tested
Mechanical quantity etc. be converted to the sensor of electric capacitance change.Its non-sensitive part is exactly the capacitor for having variable element.This Shen
Each capacitance sensor in capacitance sensor array that please be in the method for embodiment can be made up of two parallel poles, pole
Between plate condenser using air as medium.Ignore edge effect, the electric capacity of plate condenser is ε A/ δ, and ε is that interpolar is situated between in formula
The dielectric constant of matter, A are the effective area that two electrodes cover mutually, and δ is the distance between two electrodes, and the humidity of contrasted between solid dielectric is sent out
Changing will cause the change of dielectric constant, and then cause the change of capacitance sensor electric capacity.In the method for the embodiment of the present application
In, a capacitance sensor can be arranged on to an at least opening position；Or the first pole plate of a capacitance sensor is set
In an at least opening position, the second pole plate is arranged on to the opening position relative with first pole plate；Or multiple electricity will be included
The capacitance sensor array for holding sensor is arranged on an at least opening position；Or the electric capacity including multiple capacitance sensors is passed
Each first pole plate of sensor is arranged on an at least opening position, forms the first pole plate array, and each second pole plate is arranged on and each the
The relative opening position of one pole plate, form the second pole plate array.Each capacitance sensor in the capacitance sensor array can be right
Answer a two-dimensional coordinate.The plate electrode of each capacitance sensor with the air in the presence of object space, destination object and its
Its possible obstacle facility is contrasted between solid dielectric.According to the quantity of used capacitance sensor and deployment form, for each institute
Rheme is put, and the temperature variation curve of time dimension can be constructed by temperature information determined by the number of degrees of capacitance sensor,
And/or the temperature variation curve of time-space dimension, and temperature variation curve is by with the specific mould corresponding to different actions
Formula.
In addition, in order to realize the purpose of the method identification maneuver of the embodiment of the present application, it is also necessary to institute in known target space
The humidity variation characteristic data that an at least position response performs different actions in the destination object are stated, with can be acquired
Corresponding feature is matched in humidity change information, and then identifies the action of destination object.That is, the method for the embodiment of the present application
Also include：
S110. obtain an at least position response in the object space and perform different actions in the destination object
Humidity variation characteristic data.
Correspondingly, step S140 includes：
S142. the humidity change information at least in a seclected time section and the humidity variation characteristic data are done
Match somebody with somebody, identify at least one action that the destination object performs.
An at least position response performs the humidity change of different actions in the destination object in the object space
Characteristic can be stored directly in the executive agent of the method for the embodiment of the present application, or the executive agent is from external equipment
Obtain, or, it can be obtained and stored in by pre-establishing training set in the executive agent, that is, being directed to each type
Destination object, each test position of actual measurement collection object space performs the period of different actions in response to destination object
The humidity information at interior each moment, obtain the humidity changing pattern corresponding to each action.By enough samples, can establish pair
Answer such humidity variation characteristic database of the various motion of diverse location.
Further, since object space integrated environment humidity may also change for some reason, therefore, in order to carry
The accuracy of identification of the method for high the embodiment of the present application, the measured value ripple of dew cell caused by environment overall variation can be removed
It is dynamic.For example, when handling the humidity data acquired in dew cell, by the difference as gradient filter or Gaussian function
Algorithm, appropriate humidity threshold is set, filters out the ambient humidity component that object space is corresponded in the humidity data collected, etc.
Deng.
In order to further improve the precision of the method identification maneuver of the embodiment of the present application, other means of identification are may also be combined with.
In a kind of possible implementation, the method for the embodiment of the present application may also include：
S160. the optical imagery of the object space is obtained.The optical imagery can be for example based on being arranged on object space
In the image of object space that obtains of vision sensor (shooting first-class), each of object space can be identified based on image procossing
Object distribution can also identify the action of destination object to a certain extent, be combined on the basis of humidity change information by the light
Learn image, it is possible to increase the action recognition precision of destination object, set for example, can be positioned according to the optical imagery and then reject obstacle
Apply or other objects etc. caused by interference, etc..Correspondingly, step S140 can further comprise：
S144. according at least to the humidity change information and the optical imagery, identify what the destination object performed
At least one action.
In a kind of possible implementation, the method for the embodiment of the present application may also include：
S170. the depth image of the object space is obtained.
The depth image can be such as being obtained based on the depth transducer being arranged in object space (TOF depth transducers)
The depth image of the object space obtained, the depth information that object space can be obtained based on the depth information image also can be one
Determine to identify the action that target exclusively enjoys in degree, combined the depth image on the basis of humidity change information, it is possible to increase mesh
The action recognition precision of object is marked, is made for example, can be positioned according to the depth image and then reject obstacle facility or other objects etc.
Into interference, etc..Correspondingly, step S140 can further comprise：
S146. according at least to the humidity change information and the depth image, identify what the destination object performed
At least one action.
Also can be in combination with optical imagery and depth image in the method for the embodiment of the present application, and then realize higher knowledge
Other precision.That is, the method for the embodiment of the present application may also include：
S180. the optical imagery and depth image of the object space are obtained.
Correspondingly, step S140 can further comprise：
S148. according at least to the humidity change information, the optical imagery and the depth image, the mesh is identified
Mark at least one action that object performs.
To sum up, the method for the embodiment of the present application identifies the action of destination object, letter by destination object humidity change information
It is single easy to be applied widely, and disclosure satisfy that certain required precision.
It will be understood by those skilled in the art that in the above method of the application embodiment, the sequence number of each step
Size is not meant to the priority of execution sequence, and the execution sequence of each step should be determined with its function and internal logic, without answering
Any restriction is formed to the implementation process of the application embodiment.
In addition, the embodiment of the present application additionally provides a kind of computer-readable medium, including following grasp is carried out when executed
The computer-readable instruction of work：Perform the operation of each step of the method in above-mentioned Fig. 1 illustrated embodiments.
The embodiment of the present application additionally provides a kind of action recognition device for performing above-mentioned action identification method, in the above method
What is be previously mentioned in embodiment is used to obtain the sensor of various information and view data, can belong to the device of the embodiment of the present application,
Or the device of the embodiment of the present application can obtain the data of each sensor, correspondingly, in addition to each part described below, institute
Stating device also includes that the communication module to communicate with arbitrary equipment outside device can be realized as needed.As shown in Fig. 2 (a), this Shen
Please the action recognition 200 of embodiment include：
First acquisition module 220, for obtaining at least one choosing of an at least position in the object space where destination object
The humidity change information fixed time in section.
In the ideal case, because destination object athletic meeting produces disturbance to the moisture distribution of object space, different is dynamic
Work will produce different disturbances to object space moisture distribution, and in response to different actions, in object space at diverse location
The humidity that can be obtained is on time dimension also by the change with respective AD HOC.In order to identify the action of destination object,
Humidity change information of at least position at least in a seclected time section in object space is obtained in the step s 120.Wherein, institute
It can be any number of point or the region in object space to state an at least position, and can preferably be collected in optional object space
To the position of humidity change, for example, close to the point of destination object or the region of face destination object, such position can be obvious
Ground detects that the motion of destination object disturbs to caused by the moisture distribution of object space.Correspondingly, the humidity change information
Can be in the humidity change information to a little less time dimension being collected into；Or at least it can be collected into a region
The humidity change information of time dimension, or the humidity change information of time-space dimension.Seclected time section is also with any
A period of time of duration, it is corresponding specific with specifically action that the duration should meet that corresponding temperature change granularity identifies enough
Changing pattern.
Identification module 240, for according at least to the humidity change information at least in a seclected time section, described in identification
Destination object is at least one action at least performed in a seclected time section.
As described above, the humidity for the different actions that an at least position response described in identification module 240 performs in destination object
Variation characteristic, destination object can be matched in the humidity change information acquired in the first acquisition module 220 and performs specific move
The temperature change of caused AD HOC when making, and then realize and the dynamic of destination object is identified according to obtained humidity change information
Make.
In summary, the device of the embodiment of the present application identifies target by the humidity change information in space where destination object
The action of object, because the acquisition of humidity information is not only restricted to blocking for other facilities of object space, to object space
Environmental condition requires less, simple and easy applied widely.
First acquisition module 220 can obtain the space humidity distributed intelligence by an at least dew cell, for example, wet sensitive
Resistance, humicap etc..As shown in Fig. 2 (b), the first acquisition module 220 may include：
Acquiring unit 222, change for obtaining the humidity by an at least dew cell for an at least opening position
Information.
In the device of the embodiment of the present application, the first acquisition module 220 can pass through the acquisition that communicated with an at least dew cell
The humidity change information, an at least dew cell can belong to the device 200 of the present embodiment, be independently of the present embodiment
Device 200, or the first acquisition module 220 itself includes an at least dew cell.
An at least dew cell can be at least one dew cell in an at least opening position, in order to obtain essence
True humidity change information, especially obtain and embody because the motion of destination object produces the humidity change information of disturbance, it is described
An at least dew cell can be deployed in an at least opening position in an array manner, sufficiently to obtain humidity letter as far as possible
Breath.
In a kind of possible implementation, the dew cell is capacitance sensor, and an at least dew cell is
Capacitance sensor array.Capacitance sensor is a kind of sensor that measured change is converted into electric capacitance change.Tested
Mechanical quantity etc. be converted to the sensor of electric capacitance change.Its non-sensitive part is exactly the capacitor for having variable element.This Shen
Each capacitance sensor in capacitance sensor array that please be in the device of embodiment can be made up of two parallel poles, pole
Between plate condenser using air as medium.Ignore edge effect, the electric capacity of plate condenser is ε A/ δ, and ε is that interpolar is situated between in formula
The dielectric constant of matter, A are the effective area that two electrodes cover mutually, and δ is the distance between two electrodes, and the humidity of contrasted between solid dielectric is sent out
Changing will cause the change of dielectric constant, and then cause the change of capacitance sensor electric capacity.In the device of the embodiment of the present application
In, a capacitance sensor can be arranged on to an at least opening position；Or the first pole plate of a capacitance sensor is set
In an at least opening position, the second pole plate is arranged on to the opening position relative with first pole plate；Or multiple electricity will be included
The capacitance sensor array for holding sensor is arranged on an at least opening position；Or the electric capacity including multiple capacitance sensors is passed
Each first pole plate of sensor is arranged on an at least opening position, forms the first pole plate array, and each second pole plate is arranged on and each the
The relative opening position of one pole plate, form the second pole plate array.Each capacitance sensor in the capacitance sensor array can be right
Answer a two-dimensional coordinate.The plate electrode of each capacitance sensor with the air in the presence of object space, destination object and its
Its possible obstacle facility is contrasted between solid dielectric.According to the quantity of used capacitance sensor and deployment form, for each institute
Rheme is put, and the temperature variation curve of time dimension can be constructed by temperature information determined by the number of degrees of capacitance sensor,
And/or the temperature variation curve of time-space dimension, and temperature variation curve is by with the specific mould corresponding to different actions
Formula.
In addition, in order to realize the purpose of the device identification maneuver of the embodiment of the present application, it is also necessary to institute in known target space
The humidity variation characteristic data that an at least position response performs different actions in the destination object are stated, with can be acquired
Corresponding feature is matched in humidity change information, and then identifies the action of destination object.That is, as shown in Fig. 2 (c), this Shen
Please the device 200 of embodiment also include：
Second acquisition module 210, for obtaining in the object space an at least position response in the target pair
Humidity variation characteristic data as performing different actions.
Correspondingly, identification module 240 can further comprise：
First recognition unit 242, for by the humidity change information at least in a seclected time section and the humidity
Variation characteristic data match, and identify at least one action that the destination object performs.
An at least position response performs the humidity change of different actions in the destination object in the object space
Characteristic can be stored directly in the device 200 of the embodiment of the present application, or the device 300 obtains from external equipment,
Or can be obtained and stored in by pre-establishing training set in the executive agent, that is, being directed to each type of target pair
As each test position of actual measurement collection object space performs each moment in the period of different actions in response to destination object
Humidity information, obtain the humidity changing pattern corresponding to each action.By enough samples, corresponding different positions can be established
The such humidity variation characteristic database for the various motion put.
Further, since object space integrated environment humidity may also change for some reason, therefore, in order to carry
The accuracy of identification of the device of high the embodiment of the present application, the measured value ripple of dew cell caused by environment overall variation can be removed
It is dynamic.For example, when handling the humidity data acquired in dew cell, by the difference as gradient filter or Gaussian function
Algorithm, appropriate humidity threshold is set, filters out the ambient humidity component that object space is corresponded in the humidity data collected, etc.
Deng.
In order to further improve the precision of the device identification maneuver of the embodiment of the present application, other means of identification are may also be combined with.
As shown in Fig. 2 (d), in a kind of possible implementation, the device 200 of the embodiment of the present application may also include：
3rd acquisition module 260, for obtaining the optical imagery of the object space.The optical imagery can be for example based on
The image for the object space that the vision sensor (shooting is first-class) being arranged in object space obtains, can be known based on image procossing
Each object distribution of other object space can also identify the action of destination object to a certain extent, in the base of humidity change information
Combined the optical imagery on plinth, it is possible to increase the action recognition precision of destination object, for example, can be positioned according to the optical imagery
And then interference, etc. caused by rejecting obstacle facility or other objects etc..Correspondingly, identification module 240 can also further comprise：
Second recognition unit 244, for according at least to the humidity change information and the optical imagery, described in identification
At least one action that destination object performs.
In the device of the embodiment of the present application, the 3rd acquisition module 260 can be by communicating described in acquisition with vision sensor
Optical imagery, the vision sensor can belong to the device 200 of the present embodiment, be independently of the device 200 of the present embodiment, or
The acquisition module 260 of person the 3rd itself includes the vision sensor.
As shown in Fig. 2 (e), in a kind of possible implementation, the device 200 of the embodiment of the present application may also include：
4th acquisition module 270, for obtaining the depth image of the object space.
The depth image can be such as being obtained based on the depth transducer being arranged in object space (TOF depth transducers)
The depth image of the object space obtained, the depth information that object space can be obtained based on the depth information image also can be one
Determine to identify the action that target exclusively enjoys in degree, combined the depth image on the basis of humidity change information, it is possible to increase mesh
The action recognition precision of object is marked, is made for example, can be positioned according to the depth image and then reject obstacle facility or other objects etc.
Into interference, etc..Correspondingly, identification module 240 can further comprise：
3rd recognition unit 246, for according at least to the humidity change information and the depth image, described in identification
At least one action that destination object performs.
In the device of the embodiment of the present application, the 4th acquisition module 270 can pass through the acquisition that communicated with the depth transducer
The temperature information, the depth transducer can belong to the device 200 of the present embodiment, be independently of the dress of the present embodiment
Putting the 200, or the 4th acquisition module 270 itself includes the depth transducer.
Also can be in combination with optical imagery and depth image in the device of the embodiment of the present application, and then realize higher knowledge
Other precision.That is, as shown in Fig. 2 (f), the device 200 of the embodiment of the present application may also include：
5th acquisition module 280, for obtaining the optical imagery and depth image of the object space.
Correspondingly, identification module 240 can also further comprise：
4th recognition unit 248, for according at least to the humidity change information, the optical imagery and the depth
Image, identify at least one action that the destination object performs.
In the device of the embodiment of the present application, the 5th acquisition module 280 can by with the vision sensor and depth
Sensor communication obtains the humidity change information, and the vision sensor and depth transducer can belong to the device of the present embodiment
200, be independently of the device 200 of the present embodiment, or the 5th acquisition module 280 itself include the vision sensor with
Depth transducer.
To sum up, the device of the embodiment of the present application identifies the dynamic of destination object by destination object space humidity change information
Make, it is simple and easy applied widely, and disclosure satisfy that certain required precision.
Fig. 3 is the structural representation for another action recognition device 300 that the embodiment of the present application provides, and the application is specifically real
Specific implementation of the example not to action recognition device 300 is applied to limit.As shown in figure 3, the action recognition device 300 can wrap
Include：
Processor (processor) 310, communication interface (Communications Interface) 320, memory
(memory) 330 and communication bus 340.Wherein：
Processor 310, communication interface 320 and memory 330 complete mutual communication by communication bus 340.
Communication interface 320, for being communicated with the network element of such as client etc..
Processor 310, for configuration processor 332, it can specifically realize to act in the device embodiment of above-mentioned Fig. 2 (a) and know
The correlation function of other device.
Specifically, program 332 can include program code, and described program code includes computer-managed instruction.
Processor 310 is probably a central processor CPU, or specific integrated circuit ASIC (Application
Specific Integrated Circuit), or it is arranged to implement the integrated electricity of one or more of the embodiment of the present application
Road.Program 332 specifically can be used for so that the action recognition device 300 performs following steps：
The change letter of the humidity at least in a seclected time section of an object space at least position where acquisition destination object
Breath；
According at least to the humidity change information at least in a seclected time section, identify the destination object it is described extremely
At least one action performed in a few seclected time section.
Corresponding to the specific implementation of each step may refer in corresponding steps and the unit in above-described embodiment in program 332
Description, will not be described here.It is apparent to those skilled in the art that for convenience and simplicity of description, it is above-mentioned to retouch
The equipment and the specific work process of module stated, the corresponding process description in preceding method embodiment is may be referred to, herein no longer
Repeat.
It is apparent to those skilled in the art that for convenience and simplicity of description, the equipment of foregoing description
With the specific work process of module, the corresponding description in aforementioned means embodiment is may be referred to, will not be repeated here.
Although subject matter described herein is held in the execution on the computer systems of binding operation system and application program
There is provided in capable general context, but it will be appreciated by the appropriately skilled person that may also be combined with other kinds of program module
To perform other realizations.In general, program module include perform particular task or realize particular abstract data type routine,
Program, component, data structure and other kinds of structure.It will be understood by those skilled in the art that subject matter described herein can
To be put into practice using other computer system configurations, including portable equipment, multicomputer system, based on microprocessor or can compile
Journey consumption electronic product, minicom, mainframe computer etc., it is possible to use task by communication network by being connected wherein
In the DCE that remote processing devices perform.In a distributed computing environment, program module can be located locally and far
Journey memory storage device both in.
Those of ordinary skill in the art are it is to be appreciated that the list of each example described with reference to the embodiments described herein
Member and method and step, it can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
Performed with hardware or software mode, application-specific and design constraint depending on technical scheme.Professional and technical personnel
Described function can be realized using distinct methods to each specific application, but this realization is it is not considered that exceed
Scope of the present application.
If the function is realized in the form of SFU software functional unit and is used as independent production marketing or in use, can be with
It is stored in a computer read/write memory medium.Based on such understanding, the technical scheme of the application is substantially in other words
The part to be contributed to original technology or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are causing a computer equipment (can be
People's computer, server, or network equipment etc.) perform each embodiment methods described of the application all or part of step.
And foregoing computer read/write memory medium include with storage as computer-readable instruction, data structure, program module or its
Any mode or technology of the information such as his data are come the physics volatibility realized and non-volatile, removable and can not be situated between because of east
Matter.Computer read/write memory medium specifically includes, but is not limited to, USB flash disk, mobile hard disk, read-only storage (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), Erasable Programmable Read Only Memory EPROM
(EPROM), EEPROM (EEPROM), flash memory or other solid-state memory technologies, CD-ROM, numeral are more
Functional disc (DVD), HD-DVD, blue light (Blue-Ray) or other light storage devices, tape, disk storage or other magnetic storages
Equipment can be used for any other medium that stores information needed and can be accessed by computer.
Embodiment of above is merely to illustrate the present invention, and not limitation of the present invention, about the common of technical field
Technical staff, without departing from the spirit and scope of the present invention, it can also make a variety of changes and modification, thus it is all
Equivalent technical scheme falls within scope of the invention, and scope of patent protection of the invention should be defined by the claims.
Claims (18)
1. a kind of action identification method, it is characterised in that methods described includes：
The humidity change information at least in a seclected time section of an object space at least position where acquisition destination object；
An at least position response performs the humidity variation characteristic data of different actions in the destination object described in obtaining；
According at least to the humidity change information at least in a seclected time section, identify the destination object described at least one
At least one action performed in seclected time section；
The identification destination object includes at least one action at least performed in a seclected time section：
The humidity change information at least in a seclected time section is matched with the humidity variation characteristic data, identifies institute
State at least one action of destination object execution.
2. according to the method for claim 1, it is characterised in that the object space at least one obtained where destination object
The humidity change information at least in a seclected time section of position includes：
The humidity change information is obtained by least dew cell for being arranged on an at least opening position.
3. method according to any one of claim 1 to 2, it is characterised in that methods described also includes：
Obtain the optical imagery of the object space.
4. according to the method for claim 3, it is characterised in that the identification destination object is selected described at least one
At least one action performed in period includes：
According at least to the humidity change information and the optical imagery, the destination object performs described at least one is identified
Action.
5. method according to any one of claim 1 to 2, it is characterised in that methods described also includes：
Obtain the depth image of the object space.
6. according to the method for claim 5, it is characterised in that the identification destination object is selected described at least one
At least one action performed in period includes：
According at least to the humidity change information and the depth image, the destination object performs described at least one is identified
Action.
7. method according to any one of claim 1 to 2, it is characterised in that methods described also includes：
Obtain the optical imagery and depth image of the object space.
8. according to the method for claim 7, it is characterised in that the identification destination object is selected described at least one
At least one action performed in period includes：
According at least to the humidity change information, the optical imagery and the depth image, identify that the destination object is held
Capable at least one action.
9. according to the method for claim 2, it is characterised in that the dew cell is capacitance sensor.
10. a kind of action recognition device, it is characterised in that described device includes：
One first acquisition module, for obtaining an at least seclected time section for the object space at least position where destination object
Interior humidity change information；
One second acquisition module, for obtaining an at least position response in the humidity of the different actions of destination object execution
Variation characteristic data；
One identification module, for according at least to the humidity change information at least in a seclected time section, identifying the target
Object is at least one action at least performed in a seclected time section；
The identification module, including：
One first recognition unit, it is special for the humidity change information at least in a seclected time section and the humidity to be changed
Sign data match, and identify at least one action that the destination object performs.
11. device according to claim 10, it is characterised in that first acquisition module includes：
One acquiring unit, for obtaining the humidity change by being arranged on an at least dew cell for an at least opening position
Information.
12. the device according to any one of claim 10 to 11, it is characterised in that described device also includes：
One the 3rd acquisition module, for obtaining the optical imagery of the object space.
13. device according to claim 12, it is characterised in that the identification module includes：
One second recognition unit, for according at least to the humidity change information and the optical imagery, identifying the target
At least one action that object performs.
14. the device according to any one of claim 10 to 11, it is characterised in that described device also includes：
One the 4th acquisition module, for obtaining the depth image of the object space.
15. device according to claim 14, it is characterised in that the identification module includes：
One the 3rd recognition unit, for according at least to the humidity change information and the depth image, identifying the target
At least one action that object performs.
16. the device according to any one of claim 10 to 11, it is characterised in that described device also includes：
One the 5th acquisition module, for obtaining the optical imagery and depth image of the object space.
17. device according to claim 16, it is characterised in that the identification module includes：
One the 4th recognition unit, for according at least to the humidity change information, the optical imagery and the depth image,
Identify at least one action that the destination object performs.
18. device according to claim 11, it is characterised in that the dew cell is capacitance sensor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410563395.6A CN104318248B (en) | 2014-10-21 | 2014-10-21 | Action identification method and action recognition device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410563395.6A CN104318248B (en) | 2014-10-21 | 2014-10-21 | Action identification method and action recognition device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104318248A CN104318248A (en) | 2015-01-28 |
| CN104318248B true CN104318248B (en) | 2018-04-06 |
Family
ID=52373477
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410563395.6A Active CN104318248B (en) | 2014-10-21 | 2014-10-21 | Action identification method and action recognition device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104318248B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108960059A (en) * | 2018-06-01 | 2018-12-07 | 众安信息技术服务有限公司 | A kind of video actions recognition methods and device |
| CN109800659B (en) * | 2018-12-26 | 2021-05-25 | 中国科学院自动化研究所南京人工智能芯片创新研究院 | Action recognition method and device |
| CN109886102B (en) * | 2019-01-14 | 2020-11-17 | 华中科技大学 | Fall-down behavior time-space domain detection method based on depth image |
| CN110084211B (en) * | 2019-04-30 | 2020-12-18 | 苏州大学 | Action recognition method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103141080A (en) * | 2010-09-23 | 2013-06-05 | 索尼电脑娱乐公司 | User interface system and method using thermal imaging |
| CN103370674A (en) * | 2011-02-21 | 2013-10-23 | 皇家飞利浦有限公司 | Gesture recognition system |
| CN103390174A (en) * | 2012-05-07 | 2013-11-13 | 深圳泰山在线科技有限公司 | Physical education assisting system and method based on human body posture recognition |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5759161B2 (en) * | 2010-12-16 | 2015-08-05 | キヤノン株式会社 | Object recognition device, object recognition method, learning device, learning method, program, and information processing system |
| JP5386539B2 (en) * | 2011-05-12 | 2014-01-15 | 富士重工業株式会社 | Environment recognition device |
-
2014
- 2014-10-21 CN CN201410563395.6A patent/CN104318248B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103141080A (en) * | 2010-09-23 | 2013-06-05 | 索尼电脑娱乐公司 | User interface system and method using thermal imaging |
| CN103370674A (en) * | 2011-02-21 | 2013-10-23 | 皇家飞利浦有限公司 | Gesture recognition system |
| CN103390174A (en) * | 2012-05-07 | 2013-11-13 | 深圳泰山在线科技有限公司 | Physical education assisting system and method based on human body posture recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104318248A (en) | 2015-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gu et al. | Paws: Passive human activity recognition based on wifi ambient signals | |
| Kim et al. | Target localization using ensemble support vector regression in wireless sensor networks | |
| CN104318248B (en) | Action identification method and action recognition device | |
| Mekruksavanich et al. | A multichannel cnn-lstm network for daily activity recognition using smartwatch sensor data | |
| Kolosnjaji et al. | Neural network-based user-independent physical activity recognition for mobile devices | |
| CN106456000A (en) | Motion based estimation of biometric signals | |
| Koc et al. | Comparison of linear correlation and a statistical dependency measure for inferring spatial relation of temperature sensors in buildings | |
| Banjarey et al. | Human activity recognition using 1D convolutional neural network | |
| CN104392204B (en) | Gesture recognition method and gesture recognition device | |
| Ferreira et al. | User-centric coordinates for applications leveraging 3-axis accelerometer data | |
| Schmid et al. | SVM versus MAP on accelerometer data to distinguish among locomotor activities executed at different speeds | |
| CN107169334B (en) | The user authen method based on straight punch motion detection for hand wearable device | |
| Menter et al. | Application of machine learning-based pattern recognition in iot devices | |
| Zeng et al. | Accelerometer-based gait recognition via deterministic learning | |
| Galván-Tejada et al. | Magnetic-field feature extraction for indoor location estimation | |
| Martinelli et al. | Daily movement recognition for dead reckoning applications | |
| AU2020102094A4 (en) | GROUP ACTIVITY RECOGNITION BY INTEGRATION AND FUSION OF INDIVIDUAL MULTISENSORY IoT DATA | |
| Choudhary et al. | A seismic sensor based human activity recognition framework using deep learning | |
| Yu et al. | Deep learning-based vibration signal personnel positioning system | |
| CN108696377A (en) | A method of realizing mobile subscriber's Activity recognition | |
| Hashim et al. | Machine learning-based human activity recognition using neighbourhood component analysis | |
| Dehzangi et al. | Motion-based gait identification using spectro-temporal transform and convolutional neural networks | |
| Hossen et al. | Smartphone-Based Drivers Context Recognition | |
| Nana et al. | The elderly’s falling motion recognition based on kinect and wearable sensors | |
| Duffy et al. | An investigation into smartphone based weakly supervised activity recognition systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |