CN109484935A - A kind of lift car monitoring method, apparatus and system - Google Patents
A kind of lift car monitoring method, apparatus and system Download PDFInfo
- Publication number
- CN109484935A CN109484935A CN201710822052.0A CN201710822052A CN109484935A CN 109484935 A CN109484935 A CN 109484935A CN 201710822052 A CN201710822052 A CN 201710822052A CN 109484935 A CN109484935 A CN 109484935A
- Authority
- CN
- China
- Prior art keywords
- human body
- depth image
- body target
- information
- lift car
- 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
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000012544 monitoring process Methods 0.000 title claims abstract description 57
- 238000012806 monitoring device Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 230000003287 optical effect Effects 0.000 claims description 8
- 230000001960 triggered effect Effects 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 description 40
- 238000010801 machine learning Methods 0.000 description 12
- 238000012549 training Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000007637 random forest analysis Methods 0.000 description 5
- 238000012706 support-vector machine Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 238000009434 installation Methods 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 241000023320 Luma <angiosperm> Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- OSWPMRLSEDHDFF-UHFFFAOYSA-N methyl salicylate Chemical compound COC(=O)C1=CC=CC=C1O OSWPMRLSEDHDFF-UHFFFAOYSA-N 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000000465 moulding Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0012—Devices monitoring the users of the elevator system
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/02—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Indicating And Signalling Devices For Elevators (AREA)
Abstract
The embodiment of the invention provides a kind of lift car monitoring methods, apparatus and system, this method comprises: obtaining image capture device when detecting that elevator cab door is closed and being directed to lift car scene depth image collected；Whether judge in depth image obtained comprising human body target；When judgement is comprising human body target, it is based on depth image obtained, determines human height's information of included human body target；Whether human height's information determined by judging is lower than preset height threshold value；When judging that identified human height's information is lower than preset height threshold value, alarm.The intelligent monitoring to only having the case where child takes in elevator is realized using the embodiment of the present invention, saves manpower, and is avoided there may be because artificial energy is limited or the factors such as human negligence, and the omission phenomenon occurred.
Description
Technical field
The present invention relates to technical field of computer vision, more particularly to a kind of lift car monitoring method, device and are
System.
Background technique
Elevator is more and more widely used in life as important vertical transport tool.Elevator is convenient
While people go on a journey, security risk has also been brought.In oneainstance, above-mentioned security risk may include elevator car
Only have the case where child takes in compartment, since child is active and security precautions are weak, cannot such as find in time, it is likely that meeting
Lead to tragedy.Although being assembled with image capture device in the lift car much built, current above-mentioned Image Acquisition
Video monitoring system where equipment is only capable of providing image collecting function；Subsequent, monitoring personnel is adopted by manually watching image
Collect equipment acquired image, recognizes whether the case where only child takes in elevator occur.Above-mentioned building may include but not
It is limited to cell, hotel and mansion etc..
In such a way that above-mentioned artificial viewing image is monitored, monitoring personnel is needed to watch image capture device in real time
Acquired image consumes manpower, not smart enough, and there may be because artificial energy is limited or human negligence etc. because
Element, and the omission phenomenon occurred.
Summary of the invention
The embodiment of the present invention is designed to provide a kind of lift car monitoring method, apparatus and system, to realize to electricity
Manpower is saved in the intelligent monitoring for the case where only child takes in terraced, and avoid there may be because artificial energy it is limited or
The factors such as person's human negligence, and the omission phenomenon occurred.Specific technical solution is as follows:
On the one hand, the embodiment of the invention provides a kind of lift car monitoring methods, which comprises
When detecting that elevator cab door is closed, obtains image capture device and be directed to lift car scene depth collected
Image；
It whether determines in depth image obtained comprising human body target；
When determining comprising human body target, it is based on depth image obtained, determines that the human body of included human body target is high
Spend information；
Whether human height's information determined by judging is lower than preset height threshold value；
When judging that identified human height's information is lower than the preset height threshold value, alarm.
Optionally, the step of whether including human body target in determination depth image obtained, comprising:
Characteristics of image is extracted from depth image obtained；
Based on extracted characteristics of image and the first default classifier, determine in depth image obtained whether include
Human body target.
Optionally, the step of whether including human body target in determination depth image obtained, comprising:
Depth image obtained is converted into the point cloud data under world coordinate system；
It whether determines in the point cloud data comprising the corresponding point cloud data of human body target, wherein when judging described cloud
In data when point cloud data corresponding comprising human body target, characterize in depth image obtained comprising human body target.
Optionally, the step of whether including human body target corresponding point cloud data in the determination point cloud data, packet
It includes:
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and the second default classifier, all kinds of cloud subnumbers obtained are determined
It whether include corresponding cloud subdata of human body target in.
Optionally, the step of whether including human body target corresponding point cloud data in the determination point cloud data, packet
It includes:
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and default manikin, all kinds of cloud subdatas obtained are determined
In whether include corresponding cloud subdata of human body target.
Optionally, the depth image that described image acquisition equipment utilization is included acquires sub- equipment sampling depth image；
The step of point cloud data depth image obtained is converted under world coordinate system, comprising:
Obtain the parameter information that the depth image acquires sub- equipment, wherein include in the parameter information: focal length letter
Breath, principal point information, mounting height information and setting angle information；
Using the focus information, the principal point information, depth image obtained is converted under device coordinate system
Point cloud data, wherein the device coordinate system are as follows: the coordinate established based on the optical center that the depth image acquires sub- equipment
System；
Using the mounting height information and the setting angle information, by the point cloud number under the device coordinate system
According to the point cloud data being converted under the world coordinate system.
Optionally, the step of acquisition image capture device is for lift car scene depth image collected, packet
It includes:
It obtains described image acquisition equipment and is directed to lift car scene depth image collected and depth obtained
The corresponding color image of image；
The step of whether including human body target in the determination depth image obtained, comprising:
By whether identifying in the color image comprising human body target, determine in the depth image obtained whether include people
Body target.
Optionally, in described the step of being based on depth image obtained, determining the elevation information of included human body target
Before, the method also includes:
When determining comprising human body target, judge whether included human body target is one；
When judging included human body target for one, execute it is described be based on depth image obtained, determination is included
The step of elevation information of human body target.
Optionally, before described the step of being alarmed, the method also includes:
When judging that identified human height's information is lower than the preset height threshold value, included human body target is determined
Quantity；
When the quantity of identified included human body target is lower than preset quantity, described the step of being alarmed is executed.
Optionally, after described the step of being alarmed, the method also includes:
It controls the lift car and is moved to designated floor；
After the lift car is moved to the designated floor, controls the lift car and stop movement；
After obtaining the door open command for being directed to the lift car, controls the lift car and open compartment door.
On the other hand, the embodiment of the invention provides a kind of lift car monitoring device, described device includes:
First obtains module, is directed to elevator car for when detecting that elevator cab door is closed, obtaining image capture device
Compartment scene depth image collected；
First determining module, for whether determining in depth image obtained comprising human body target；
Second determining module, for being based on depth image obtained when determining comprising human body target, determination is included
Human height's information of human body target；
First judgment module, for judging whether identified human height's information is lower than preset height threshold value；
Alarm module, for being reported when judging that identified human height's information is lower than the preset height threshold value
It is alert.
Optionally, first determining module, is specifically used for
Characteristics of image is extracted from depth image obtained；
Based on extracted characteristics of image and the first default classifier, determine in depth image obtained whether include
Human body target.
Optionally, first determining module includes converting unit and determination unit；
The converting unit, for depth image obtained to be converted to the point cloud data under world coordinate system；
The determination unit, for whether determining in the point cloud data comprising the corresponding point cloud data of human body target,
In, when judging point cloud data corresponding comprising human body target in the point cloud data, characterizes and wrapped in depth image obtained
Containing human body target.
Optionally, the determination unit, is specifically used for
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and the second default classifier, all kinds of cloud subnumbers obtained are determined
It whether include corresponding cloud subdata of human body target in.
Optionally, the determination unit, is specifically used for
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and default manikin, all kinds of cloud subdatas obtained are determined
In whether include corresponding cloud subdata of human body target.
Optionally, the depth image that described image acquisition equipment utilization is included acquires sub- equipment sampling depth image；
The converting unit, is specifically used for
Obtain the parameter information that the depth image acquires sub- equipment, wherein include in the parameter information: focal length letter
Breath, principal point information, mounting height information and setting angle information；
Using the focus information, the principal point information, depth image obtained is converted under device coordinate system
Point cloud data, wherein the device coordinate system are as follows: the coordinate established based on the optical center that the depth image acquires sub- equipment
System；
Using the mounting height information and the setting angle information, by the point cloud number under the device coordinate system
According to the point cloud data being converted under the world coordinate system.
Optionally, described first module is obtained, be specifically used for
It obtains described image acquisition equipment and is directed to lift car scene depth image collected and depth obtained
The corresponding color image of image；
First determining module, is specifically used for
By whether identifying in the color image comprising human body target, determine in the depth image obtained whether include people
Body target.
Optionally, described device further includes the second judgment module；
Second judgment module determines included human body target for being based on depth image obtained described
Before elevation information, when determining comprising human body target, judge whether included human body target is one；As the included people of judgement
When body target is one, second determining module is triggered.
Optionally, described device further includes third determining module；
The third determining module, for before described alarmed, human height's information determined by judge is low
When the preset height threshold value, the quantity of included human body target is determined；
When the quantity of identified included human body target is lower than preset quantity, the alarm module is triggered.
Optionally, described device further includes the first control module, the second control module and third control module；
First control module is moved to specified building for after described alarmed, controlling the lift car
Layer；
Second control module, for controlling the elevator after lift car is moved to the designated floor
Carriage stops movement；
The third control module, for controlling the elevator after obtaining the door open command for being directed to the lift car
Carriage opens compartment door.
On the other hand, a kind of electronic equipment, including processor and memory are provided, wherein the memory, for storing
Computer program；
The processor when for executing the computer program stored on memory, realizes any of the above-described lift car
Monitoring method.
On the other hand, a kind of elevator device is provided, comprising: elevator and any of the above-described lift car monitoring device.
Optionally, the elevator device, including camera for acquiring the depth image in lift car, and are sent to institute
State lift car monitoring device.
In the embodiment of the present invention, when detecting that elevator cab door is closed, obtains image capture device and be directed to lift car
Scene depth image collected；Whether judge in depth image obtained comprising human body target；When judgement includes human body mesh
When mark, it is based on depth image obtained, determines human height's information of included human body target；Human body determined by judging is high
Whether degree information is lower than preset height threshold value；When judging that identified human height's information is lower than preset height threshold value, carry out
Alarm.
When detecting that elevator cab door is closed, obtains image capture device and be directed to lift car scene depth collected
Image, and whether determine in depth image includes that human body target further determines that human body target when determining comprising human body target
Elevation information, and combine preset height threshold value, judge whether included human body target is child, when judging above-mentioned human body mesh
When target elevation information is lower than preset height threshold value, it may be determined that the human body target for being included is child, and can be determined electric at this time
Occurs the case where only having the case where child takes, i.e., only child takes elevator in terraced carriage.Monitoring only child's seating
It in the case where elevator, alarms, subsequent to warn monitoring personnel to be paid close attention to, monitoring personnel can carry out accordingly
Processing avoids the child for taking elevator from occurring dangerous, to realize the intelligent monitoring to only having the case where child takes in elevator, section
Human-saving.Also, when avoiding monitoring personnel and manually being monitored, it is understood that there may be because artificial energy is limited or human negligence
Etc. factors, and occur monitoring omit the case where, in turn result in and the case where tragedy occur.Certainly, it implements any of the products of the present invention
Or method must be not necessarily required to reach all the above advantage simultaneously.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow diagram of lift car monitoring method provided by the embodiment of the present invention；
Fig. 2 is a kind of another flow diagram of lift car monitoring method provided by the embodiment of the present invention；
Fig. 3 is a kind of structural schematic diagram of lift car monitoring device provided by the embodiment of the present invention；
Fig. 4 is a kind of another structural schematic diagram of lift car monitoring device provided by the embodiment of the present invention；
Fig. 5 is the structural schematic diagram of a kind of electronic equipment provided by the embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention provides a kind of lift car monitoring methods, apparatus and system, only have to realize in elevator
Manpower is saved in the intelligent monitoring for the case where child takes, and avoids that there may be because artificial energy is limited or artificial dredge
The factors such as suddenly, and the omission phenomenon occurred.
As shown in Figure 1, may include steps of the embodiment of the invention provides a kind of lift car monitoring method:
S101: it when detecting that elevator cab door is closed, obtains image capture device and is acquired for lift car scene
Depth image；
It is understood that a kind of lift car monitoring method provided by the embodiment of the present invention, can be applied to any
In the electronic equipment that image capture device depth image collected can be obtained, above-mentioned electronic equipment can be computer, intelligence
Mobile phone, monitoring server etc..
Above-mentioned image capture device can be installed in above-mentioned lift car.Above-mentioned Image Acquisition is installed in lift car
When equipment, can also tilt installation with right angle setting, wherein right angle setting may is that prolong lift car lifting direction into
Row installation, above-mentioned image capture device can be installed on the centre of elevator car roof, and face lift car ground, to electricity
Scene in terraced carriage carries out Image Acquisition；Inclination installation can be the direction to go up and down with lift car, and there are certain angles
Direction is installed, and above-mentioned image capture device can be installed on any edge of elevator car roof, to lift car
Interior scene carries out Image Acquisition.It is understood that howsoever installing above-mentioned image capture device, it need to guarantee above-mentioned image
Equipment is acquired, the image in each corner in lift car can be collected to the greatest extent.
Above-mentioned image capture device can acquire the depth information namely three-dimensional information of target in scene to be any
Equipment can include but is not limited to binocular depth video camera, TOF (Time of Flight, flight time) video camera and structure
Light video camera.
It may include the depth information of each point in lift car scene in above-mentioned depth image, i.e., it is each in lift car scene
Point arrives the range information of image capture device.In one implementation, the pixel value of each pixel is in above-mentioned depth image
Are as follows: the depth information of the point in the corresponding lift car scene of each pixel.As it can be seen that depth image have be not illuminated by the light, shade
And the characteristic that the factors such as coloration influence.
S102: it whether determines in depth image obtained comprising human body target；
After electronic equipment obtains above-mentioned depth image, it can identify whether to include human body target from above-mentioned depth image.
In one implementation, electronic equipment can be detected in above-mentioned depth image first with the presence or absence of target, and above-mentioned target can be with
Including human body target and object target.When detecting in above-mentioned depth image there are when target, electronic equipment is from above-mentioned depth map
The each target for being included is determined as in, it is subsequent, identified judge whether each target is human body for each target
Target.
In a kind of situation, electronic equipment can prestore the background depth image of above-mentioned lift car scene, as elevator
When carriage is unloaded, the depth image of lift car scene；When electronic equipment detects that elevator cab door is closed, obtains image and adopt
Collect equipment and be directed to lift car scene depth image collected, by depth image obtained and above-mentioned background depth image into
It is poor that row is made, acquisition differential chart, when there are when foreground pixel point, then can determining in above-mentioned depth image obtained in differential chart
Include target.Certainly, when carrying out depth image obtained and above-mentioned background depth image to make difference, it need to guarantee Image Acquisition
The visual angle of equipment does not change, i.e., the position of image capture device and/or angle do not change.
In one implementation, electronic equipment is identified judge whether each target is human body for each target
The process of target, which may is that, matches each target with pre-set 3 D stereo manikin, when successful match,
It can then determine that target is that human body target can then determine that target is not human body target when it fails to match.
In another implementation, electronic equipment is identified judge whether each target is people for each target
The process of body target is also possible to: using classifier trained in advance, classifies to each target in above-mentioned depth image,
Determine whether each target is human body target.It is understood that above-mentioned classifier trained in advance may is that based on training sample
The resulting machine learning model of this and preset algorithm training, wherein above-mentioned training sample may include human body target and inhuman
The sample depth image of body target.For above-mentioned preset algorithm, above-mentioned preset algorithm can be with are as follows: including convolutional neural networks
Neural network algorithm including related algorithm, at this point, above-mentioned machine learning model can be with are as follows: neural network model.Alternatively, above-mentioned
Preset algorithm can be with are as follows: random forests algorithm, at this point, above-mentioned machine learning model can be with are as follows: random forest disaggregated model.Or
Person, above-mentioned preset algorithm can be with are as follows: algorithm of support vector machine, at this point, above-mentioned machine learning model can be with are as follows: support vector machines point
Class model, etc..
S103: when determining comprising human body target, it is based on depth image obtained, determines the people of included human body target
Body elevation information；
When electronic equipment determines in depth image comprising human body target, electronic equipment can use above-mentioned depth image,
Determine human height's information of included human body target.In one implementation, electronic equipment can be by depth obtained
Image is converted into the point cloud data under world coordinate system, subsequent, utilizes the point cloud under world coordinate system corresponding to human body target
Data determine human height's information of human body target.
In oneainstance, it can be highest point in the point cloud data under world coordinate system corresponding to determining human body target
Cloud data (i.e. the corresponding point cloud data in human body target crown position) and minimum point cloud data (i.e. human body target sole position pair
The point cloud data answered), by the difference in height between above-mentioned highest point cloud data and minimum point cloud data, as human body target
Human height's information.
In another scenario, electronic equipment can be for the point cloud number under world coordinate system corresponding to each human body target
According to the height based on the every bit in the point cloud data under world coordinate system corresponding to each human body target is ranked up, and is calculated
The average value of the height of preceding predetermined quantity point in collating sequence makes a reservation for as the first average value, and after calculating in collating sequence
The average value of the height of quantity point makees the absolute value of above-mentioned first average value and the second average value as the second average value
For human height's information of human body target.
S104: whether human height's information determined by judging is lower than preset height threshold value；
S105: it when judging that identified human height's information is lower than preset height threshold value, alarms.
Electronic equipment, can be by identified human height after determining human height's information of included human body target
Information is compared with preset height threshold value, judges whether identified human height's information is lower than preset height threshold value.One
It plants in implementation, may include one or more human body targets in depth image, when there are multiple human body targets, can incite somebody to action
Human height's information of each human body target is compared with preset height threshold value respectively, judges that the human body of each human body target is high
Whether degree information is lower than preset height threshold value.
When electronic equipment judges that identified human height's information is lower than preset height threshold value, alarm.Wherein, electric
When sub- equipment is alarmed, it can be and alarmed in the form of exporting prompt information, be also possible to issue voice prompt
Form is alarmed, be also possible to alarm in the form of through variation screen intensity, etc., it is all to cause monitor
The mode of the attention of member, can be used as the alarm form in the embodiment of the present invention.
It in one implementation, can when human body target included in electronic equipment determines depth image is multiple
To be that just will do it alarm only when the human height's information for the human body target for being included is below preset height threshold value.
Above-mentioned preset height threshold value can be to be set according to the height of the crowd as child, in a kind of realization side
In formula, the above-mentioned crowd as child may include 10 years old crowd below.
Using the embodiment of the present invention, when detecting that elevator cab door is closed, obtains image capture device and be directed to elevator car
Compartment scene depth image collected, and whether determine in depth image includes human body target, when determining comprising human body target,
It further determines that human height's information of human body target, and combines preset height threshold value, whether judge included human body target
For child, when judging human height's information of above-mentioned human body target lower than preset height threshold value, it may be determined that the human body for being included
Target is child, and can determine and occur only having the case where child takes in lift car at this time, i.e., only child takes elevator
The case where.In the case where monitoring only child's seating elevator, alarm, to warn monitoring personnel to be paid close attention to,
Subsequent, monitoring personnel can carry out respective handling, avoid the child for taking elevator from occurring dangerous, only have to realize in elevator
Manpower is saved in the intelligent monitoring for the case where child takes.Also, when avoiding monitoring personnel and manually being monitored, it is understood that there may be
Because artificial energy is limited or the factors such as human negligence, and the case where monitoring is omitted occurs, the feelings for tragedy occur are in turn resulted in
Condition.
When whether electronic equipment includes human body target in determining depth image, the figure in depth image can be directly utilized
As feature, whether determines in depth image comprising human body target, be also possible to the spy using the corresponding point cloud data of depth image
Sign determines that this is all possible whether comprising human body target in depth image.
In one implementation, in determination depth image obtained whether include human body target (S102) step
Suddenly, may include:
Characteristics of image is extracted from depth image obtained；
Based on extracted characteristics of image and the first default classifier, determine in depth image obtained whether include
Human body target.
It is understood that above-mentioned first default classifier may is that based on training sample and preset algorithm training institute
Machine learning model, wherein above-mentioned training sample can be with are as follows: the sample depth figure comprising human body target and non-human target
Picture, that is, the characteristics of image for training the above-mentioned first default classifier to be utilized can be with are as follows: the image in above-mentioned sample depth image is special
Sign.For above-mentioned preset algorithm, above-mentioned preset algorithm can be with are as follows: the nerve including convolutional neural networks related algorithm
Network algorithm, at this point, above-mentioned machine learning model can be with are as follows: neural network model.Alternatively, above-mentioned preset algorithm can be with are as follows: with
Machine forest algorithm, at this point, above-mentioned machine learning model can be with are as follows: random forest disaggregated model.Alternatively, above-mentioned preset algorithm can be with
Are as follows: algorithm of support vector machine, at this point, above-mentioned machine learning model can be with are as follows: support vector cassification model, etc..
In one implementation, electronic equipment can carry out image characteristics extraction to depth image, such as: it utilizes
Canny operator etc. carries out Edge Gradient Feature to depth image.It is subsequent, by first default point of the input of extracted characteristics of image
Whether in class device, the first default classifier classifies to extracted characteristics of image, to determine in depth image comprising human body
Target.
Alternatively, human body image feature can also be prestored in electronic equipment.Electronic equipment extracts image from depth image
After feature, extracted characteristics of image is matched with the human body image feature prestored, when successful match, then can be determined
It include human body target in depth image.
It in another implementation, whether include human body target (S102) in determination depth image obtained
Step may include:
Depth image obtained is converted into the point cloud data under world coordinate system；
It whether determines in point cloud data comprising the corresponding point cloud data of human body target, wherein when judging to wrap in point cloud data
When point cloud data corresponding containing human body target, characterize in depth image obtained comprising human body target.
In the embodiment of the present invention, above-mentioned depth image can be converted to the point cloud under world coordinate system by electronic equipment first
Whether data, the point cloud data being then based under above-mentioned world coordinate system determine in depth image comprising human body target.Image is adopted
Collection equipment can use included depth image and acquire sub- equipment sampling depth image, and electronic equipment converts depth image to
When point cloud data under world coordinate system, the depth image acquisition in the image capture device for acquiring the depth image can use
Depth image is converted to the point cloud data under world coordinate system by the parameter information of sub- equipment.Wherein, above-mentioned parameter information can be with
The internal reference of sub- equipment is acquired including depth image and outer ginseng, electronic equipment can be based on Zhang Zhengyou calibration methods to image capture device
It is demarcated, so that it is determined that above-mentioned parameter information, alternatively, electronic equipment is demarcated when directly image capture device can dispatch from the factory
Internal reference, as the internal reference in the parameter information mentioned in the embodiment of the present invention, outer ginseng can by measurement and in advance calibration obtain
, this is all possible.In one implementation, point depth image obtained is converted under world coordinate system
The step of cloud data, may include:
Obtain the parameter information that depth image acquires sub- equipment, wherein include in parameter information: focus information, principal point
Information, mounting height information and setting angle information；
Using focus information, principal point information, depth image obtained is converted to the point cloud number under device coordinate system
It can be with according to, wherein device coordinate system are as follows: the coordinate system established based on the optical center that depth image acquires sub- equipment；
The world is converted by the point cloud data under device coordinate system using mounting height information and setting angle information
Point cloud data under coordinate system.
It is understood that above-mentioned principal point can be with are as follows: the optical axis and the intersection point as plane that depth image acquires sub- equipment.
In the embodiment of the present invention, principal point information be may include: two-dimensional coordinate of the principal point in depth image.Wherein, using as main
The two-dimensional coordinate of point, can determine the two-dimensional coordinate of each pixel in depth image.
In one implementation, by the two-dimensional coordinate (u, v) of pixel each in depth image, device coordinate is converted to
Three-dimensional coordinate (X in systemC,YC,ZC), to obtain the point cloud data of depth image.Wherein, above-mentioned preset three-dimensional rectangular coordinate
System can be with are as follows: based on the coordinate system that the optical center that depth image acquires sub- equipment is established, utilized formula is as follows when coordinate is converted:
Wherein, fDxAnd fDyIt is the focal length that depth image acquires sub- equipment；(uD0,vD0) it is that the two-dimentional of above-mentioned principal point is sat
Mark；ZCFor the point in the corresponding scene of pixel (u, v) in above-mentioned depth image, the distance of sub- equipment is acquired to depth image
Information, the i.e. pixel value of pixel (u, v).Wherein, fDxAbove-mentioned depth image determined by indicating acquires the x-axis direction of sub- equipment
On focal length, fDyAbove-mentioned depth image determined by indicating acquires the focal length on the y-axis direction of sub- equipment.Above-mentioned fDxAnd fDy?
It is contained in above-mentioned focus information, determination can directly be demarcated by Zhang Zhengyou calibration method.The two-dimensional coordinate of above-mentioned principal point can also
Directly to demarcate determination by Zhang Zhengyou calibration method.
Above equipment coordinate system can be three-dimensional cartesian coordinate system.It is subsequent, the peace of sub- equipment is acquired using depth image
Elevation information and setting angle information are filled, pitch angle, the deflection angle of sub- equipment are acquired including depth image, determines above equipment
The transformational relation of coordinate system and world coordinate system, and then based on identified transformational relation by the point cloud number under device coordinate system
According to the point cloud data being converted under world coordinate system.
After above-mentioned depth image is converted the point cloud data under world coordinate system by electronic equipment, it can use preset
Clustering algorithm clusters above-mentioned point cloud data, determines each target included in depth image, and further, really
Whether whether fixed each target is human body target, corresponding comprising human body target in the determining point cloud data in a kind of situation
The step of point cloud data may include:
Point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and the second default classifier, all kinds of cloud subnumbers obtained are determined
It whether include corresponding cloud subdata of human body target in.
In another situation, the step of whether including human body target corresponding point cloud data in the determining point cloud data,
May include:
Point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and default manikin, all kinds of cloud subdatas obtained are determined
In whether include corresponding cloud subdata of human body target.
In one implementation, above-mentioned preset clustering algorithm can be density-based algorithms, such as:
(Density-Based Spatial Clustering of Applications with Noise has noisy DBSCAN
Density clustering method) clustering algorithm, wherein DBSCAN clustering algorithm needs two parameters, respectively sweep radius
(eps) and it is minimum comprising counting (minPts).After electronic equipment obtains above-mentioned point cloud data, above-mentioned DBSCAN clustering algorithm is utilized
From above-mentioned point cloud data, the point of optional one not visited (unvisited) (be not labeled as the point accessed or not by
Labeled as the point of noise), as current point, and determine all (including eps) within eps at a distance from the current point
Point, as neighbouring point；Electronic equipment determines whether the quantity put near the current point is more than or equal to minPts, if electronics is set
Standby to determine that the quantity put near the current point is more than or equal to minPts, then electronic equipment is by current point and identified point nearby
It is determined as a cluster, which is labeled as having accessed (visited) by electronic equipment；Electronic equipment is passed for the cluster
Return, handles all points for being not labeled as having accessed (visited) or being not labeled as noise in the cluster in the above described manner, thus
Cluster is extended；If cluster is fully extended, i.e., all the points in cluster are marked as having accessed；Electronic equipment is again optional
One not visited point executes follow-up process as current point.When electronic equipment determines that the quantity put near current point is small
In minPts, then it is noise that electronic equipment, which marks the current point,；Until gathering after not including not visited point in point cloud data
Class is completed.At this point, electronic equipment can determine all kinds of cloud subdatas included in point cloud data, subsequent, electronics is set
It is standby to be handled for every class point cloud subdata, it whether is the corresponding point of human body target with the every a kind of point cloud subdata of determination
Cloud subdata.
In oneainstance, it can be electronic equipment for the default classification of every a kind of point cloud subdata input second obtained
In device so that the second default classifier classifies to every a kind of point cloud subdata, determine every a kind of point cloud subdata whether be
Corresponding cloud subdata of human body target.It is understood that above-mentioned second default classifier can be with are as follows: based on training sample with
And the resulting machine learning model of preset algorithm training, wherein above-mentioned training sample can be with are as follows: comprising human body target and non-human
The corresponding point cloud data of sample depth image of target.For above-mentioned preset algorithm, above-mentioned preset algorithm can be with are as follows: including
Neural network algorithm including convolutional neural networks related algorithm, at this point, above-mentioned machine learning model can be with are as follows: neural network mould
Type.Alternatively, above-mentioned preset algorithm can be with are as follows: random forests algorithm, at this point, above-mentioned machine learning model can be with are as follows: random forest
Disaggregated model.Alternatively, above-mentioned preset algorithm can be with are as follows: algorithm of support vector machine, at this point, above-mentioned machine learning model can be with are as follows:
Support vector cassification model, etc..
In another scenario, it can be electronic equipment for every a kind of point cloud subdata obtained with preset human mould
Type is matched, and when successful match, determines that point cloud subdata is corresponding cloud subdata of human body target, when it fails to match
When, determine that point cloud subdata is not corresponding cloud subdata of human body target.
In one implementation, above-mentioned preset clustering algorithm can also be the clustering algorithm based on grid, such as:
STING clustering algorithm etc. is also possible to the clustering algorithm based on division, such as: K-means clustering algorithm etc., the present invention are real
Example is applied not to be defined the algorithm that can cluster point cloud data, it is all can be to the calculation that point cloud data is clustered
Method can be applied in the embodiment of the present invention.
In one implementation, it can use color image corresponding for lift car depth image collected,
It whether determines in depth image comprising human body target, wherein the acquisition image capture device is adopted for lift car scene
The step of depth image of collection, may include:
It obtains image capture device and is directed to lift car scene depth image collected and depth image obtained
Corresponding color image；
The step of whether including human body target in the determination depth image obtained, may include:
By whether identifying in the color image comprising human body target, determine in the depth image obtained whether include people
Body target.
It can be recognised that from color image by face recognition algorithms comprising human body target, to determine depth image
In whether include human body target, wherein include human body target in depth image in color image when including human body target；When
When not including human body target in color image, human body target is not included in depth image.
In the embodiment of the present invention, above-mentioned color image can be the image of any color mode, such as: RGB (Red
Green Blue, RGB) color mode image or YUV color mode image, etc..Wherein, YUV (also referred to as YCrCb)
It is a kind of colour coding method, " Y " indicates brightness (Luminance or Luma), that is, grayscale value；And " U " and " V " is indicated
Coloration (Chrominance or Chroma), effect are description image color and saturation degree, the color for specified pixel point." color
Degree " defines in terms of two of the color of pixel-and tone and saturation degree can be indicated with Cr and Cb respectively.
In oneainstance, when only child takes elevator, in fact it could happen that dangerous probability may larger,
When in order to preferably avoid above-mentioned only child's seating elevator, there is danger, when only a child takes elevator, prison
Control personnel avoid the above-mentioned child for taking elevator alone from occurring dangerous with greater need for paying close attention to and taking corresponding measure.In a kind of realization
In mode, as shown in Fig. 2, the method may include following steps:
S201: it when detecting that elevator cab door is closed, obtains image capture device and is acquired for lift car scene
Depth image；
S202: it whether determines in depth image obtained comprising human body target；
Wherein, above-mentioned S201 is identical as S101 shown in Fig. 1, and above-mentioned S202 is identical as S102 shown in Fig. 1.
S203: when determining comprising human body target, judge whether included human body target is one；
S204: being based on depth image obtained, determines human height's information of included human body target；
S205: whether human height's information determined by judging is lower than preset height threshold value；
S206: it when judging that identified human height's information is lower than preset height threshold value, alarms.
Wherein, above-mentioned S204 is identical as S103 shown in Fig. 1, and above-mentioned S205 is identical as S104 shown in Fig. 1, above-mentioned
S206 is identical as S105 shown in Fig. 1.
In one implementation, when electronic equipment determines human height's letter of human body target included in lift car
Breath is below preset height threshold value, that is, when determining that human body target included in lift car is child, and works as human body target
Quantity when being lower than preset quantity, need that monitoring personnel is reminded to be paid close attention to, monitoring personnel is taken and accordingly arranged for above situation
It applies, avoids above-mentioned human body target i.e. child from occurring dangerous.It is described alarmed (S105) the step of before, the method may be used also
To include:
When judging that identified human height's information is lower than preset height threshold value, the number of included human body target is determined
Amount；
When the quantity of identified included human body target is lower than preset quantity, described the step of being alarmed is executed.
Wherein, the step of quantity of the included human body target of above-mentioned determination, can the letter of the human height determined by judging
It is carried out between the step of whether breath is lower than preset height threshold value and the step of being alarmed, i.e., the human body determined by judging is high
After degree information is below preset height threshold value, the quantity of included human body target is determined；When identified included human body target
Quantity be lower than preset quantity when, alarm.Wherein, above-mentioned preset quantity can be any positive integer, for example, above-mentioned default
Quantity can be with value for 3, i.e., when electronic equipment determines that the quantity for the human body target for being included in lift car is lower than 3, i.e. child
Quantity be 2 or 1 when, electronic equipment is alarmed.In another example above-mentioned preset quantity can be with value for 2, i.e., when electronics is set
The standby quantity for determining the human body target for being included in lift car is lower than 2, i.e., when the quantity of child is 1, electronic equipment is carried out
Alarm.
In one implementation, it is described alarmed (S105) the step of after, the method can also include:
Control lift car is moved to designated floor；
After lift car is moved to the designated floor, control lift car stops movement；
After obtaining the door open command for being directed to lift car, control lift car opens compartment door.
In the embodiment of the present invention, after electronic equipment is alarmed, the movement that monitoring personnel is directed to lift car can receive
While control instruction or electronic equipment are alarmed, mobile control instruction of automatic trigger etc., this is all possible.Work as electricity
After sub- equipment obtains above-mentioned mobile control instruction, electronic equipment can control above-mentioned lift car and be moved to designated floor, and control
Lift car processed stops at above-mentioned designated layer.
Wherein, above-mentioned designated layer can be pre-set first floor layer, and such as one layer；Alternatively, can be in lift car
On moving direction and be presently in the nearest first floor layer in position with lift car, lift car is presently in position can be with
Are as follows: lift car present position when electronic equipment obtains above-mentioned mobile control instruction.Wherein, the moving direction of lift car can be with
Including rising or falling.
After electronic equipment control lift car stops at above-mentioned designated layer, the car door of above-mentioned lift car can control
It is in close state, until control lift car opens compartment door after obtaining the door open command for lift car.
Corresponding to above method embodiment, the embodiment of the invention also provides a kind of lift car monitoring devices, such as Fig. 3 institute
Show, the apparatus may include:
First obtains module 310, is directed to elevator for when detecting that elevator cab door is closed, obtaining image capture device
Carriage scene depth image collected；
First determining module 320, for whether determining in depth image obtained comprising human body target；
Second determining module 330, for being based on depth image obtained, determining institute when determining comprising human body target
Human height's information comprising human body target；
First judgment module 340, for judging whether identified human height's information is lower than preset height threshold value；
Alarm module 350, for carrying out when judging that identified human height's information is lower than the preset height threshold value
Alarm.
Using the embodiment of the present invention, when detecting that elevator cab door is closed, obtains image capture device and be directed to elevator car
Compartment scene depth image collected, and whether determine in depth image includes human body target, when determining comprising human body target,
It further determines that human height's information of human body target, and combines preset height threshold value, whether judge included human body target
For child, when judging human height's information of above-mentioned human body target lower than preset height threshold value, it may be determined that the human body for being included
Target is child, and can determine and occur only having the case where child takes in lift car at this time, i.e., only child takes elevator
The case where.In the case where monitoring only child's seating elevator, alarm, to warn monitoring personnel to be paid close attention to,
Subsequent, monitoring personnel can carry out respective handling, avoid the child for taking elevator from occurring dangerous, only have to realize in elevator
Manpower is saved in the intelligent monitoring for the case where child takes.Also, when avoiding monitoring personnel and manually being monitored, it is understood that there may be
Because artificial energy is limited or the factors such as human negligence, and the case where monitoring is omitted occurs, the feelings for tragedy occur are in turn resulted in
Condition.
In one implementation, first determining module 320, is specifically used for
Characteristics of image is extracted from depth image obtained；
Based on extracted characteristics of image and the first default classifier, determine in depth image obtained whether include
Human body target.
In one implementation, first determining module 320 includes converting unit and determination unit；
The converting unit, for depth image obtained to be converted to the point cloud data under world coordinate system；
The determination unit, for whether determining in the point cloud data comprising the corresponding point cloud data of human body target,
In, when judging point cloud data corresponding comprising human body target in the point cloud data, characterizes and wrapped in depth image obtained
Containing human body target.
In one implementation, the determination unit, is specifically used for
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and the second default classifier, all kinds of cloud subnumbers obtained are determined
It whether include corresponding cloud subdata of human body target in.
In one implementation, the determination unit, is specifically used for
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and default manikin, all kinds of cloud subdatas obtained are determined
In whether include corresponding cloud subdata of human body target.
In one implementation, it is deep that the depth image that described image acquisition equipment utilization is included acquires sub- equipment acquisition
Spend image；
The converting unit, is specifically used for
Obtain the parameter information that the depth image acquires sub- equipment, wherein include in the parameter information: focal length letter
Breath, principal point information, mounting height information and setting angle information；
Using the focus information, the principal point information, depth image obtained is converted under device coordinate system
Point cloud data, wherein the device coordinate system are as follows: the coordinate established based on the optical center that the depth image acquires sub- equipment
System；
Using the mounting height information and the setting angle information, by the point cloud number under the device coordinate system
According to the point cloud data being converted under the world coordinate system.
In one implementation, described first module 310 is obtained, be specifically used for
It obtains described image acquisition equipment and is directed to lift car scene depth image collected and depth obtained
The corresponding color image of image；
First determining module 320, is specifically used for
By whether identifying in the color image comprising human body target, determine in the depth image obtained whether include people
Body target.
In one implementation, as shown in figure 4, described device can also include the second judgment module 410；
Second judgment module 410 determines included human body target for being based on depth image obtained described
Elevation information before, when determining comprising human body target, judge whether included human body target is one；When judgement is included
When human body target is one, second determining module 330 is triggered.
In one implementation, described device can also include third determining module；
The third determining module, for before described alarmed, human height's information determined by judge is low
When the preset height threshold value, the quantity of included human body target is determined；
When the quantity of identified included human body target is lower than preset quantity, the alarm module 350 is triggered.
In one implementation, described device can also include the first control module, the second control module and third control
Molding block；
First control module is moved to specified building for after described alarmed, controlling the lift car
Layer；
Second control module, for controlling the elevator after lift car is moved to the designated floor
Carriage stops movement；
The third control module, for controlling the elevator after obtaining the door open command for being directed to the lift car
Carriage opens compartment door.
Corresponding to above method embodiment, the embodiment of the present invention also provides a kind of elevator device, comprising: elevator, Yi Jishang
State any lift car monitoring device.
Optionally, the elevator device, including camera for acquiring the depth image in lift car, and are sent to institute
State lift car monitoring device.
Corresponding to above method embodiment, the embodiment of the present invention also provides a kind of electronic equipment, including processor and storage
Device, wherein the memory, for storing computer program；
The processor when for executing the computer program stored on memory, realizes any of the above-described lift car
Monitoring method.
Corresponding to above method embodiment, the embodiment of the invention also provides a kind of electronic equipment, as shown in figure 5, including
Processor 510, communication interface 520, memory 530 and communication bus 540, wherein processor 510, communication interface 520, storage
Device 530 completes mutual communication by communication bus 540,
Memory 530, for storing computer program；
Processor 510 when for executing the computer program stored on memory 530, realizes institute of the embodiment of the present invention
Any lift car monitoring method provided, this method may include steps of:
When detecting that elevator cab door is closed, obtains image capture device and be directed to lift car scene depth collected
Image；
It whether determines in depth image obtained comprising human body target；
When determining comprising human body target, it is based on depth image obtained, determines that the human body of included human body target is high
Spend information；
Whether human height's information determined by judging is lower than preset height threshold value；
When judging that identified human height's information is lower than the preset height threshold value, alarm.
Using the embodiment of the present invention, when detecting that elevator cab door is closed, obtains image capture device and be directed to elevator car
Compartment scene depth image collected, and whether determine in depth image includes human body target, when determining comprising human body target,
It further determines that human height's information of human body target, and combines preset height threshold value, whether judge included human body target
For child, when judging human height's information of above-mentioned human body target lower than preset height threshold value, it may be determined that the human body for being included
Target is child, and can determine and occur only having the case where child takes in lift car at this time, i.e., only child takes elevator
The case where.In the case where monitoring only child's seating elevator, alarm, to warn monitoring personnel to be paid close attention to,
Subsequent, monitoring personnel can carry out respective handling, avoid the child for taking elevator from occurring dangerous, only have to realize in elevator
Manpower is saved in the intelligent monitoring for the case where child takes.Also, when avoiding monitoring personnel and manually being monitored, it is understood that there may be
Because artificial energy is limited or the factors such as human negligence, and the case where monitoring is omitted occurs, the feelings for tragedy occur are in turn resulted in
Condition.
The communication bus that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral Component
Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..The communication bus can be divided into address bus, data/address bus, control bus etc..For just
It is only indicated with a thick line in expression, figure, it is not intended that an only bus or a type of bus.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, RAM), also may include non-easy
The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
CPU), network processing unit (Network Processor, NP) etc.；It can also be digital signal processor (Digital Signal
Processing, DSP), it is specific integrated circuit (Application Specific Integrated Circuit, ASIC), existing
It is field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete
Door or transistor logic, discrete hardware components.
Corresponding to above method embodiment, the embodiment of the invention provides a kind of computer readable storage medium, the meter
It is stored with computer program in calculation machine readable storage medium storing program for executing, realizes that the present invention is implemented when the computer program is executed by processor
Any lift car monitoring method, this method provided by example may include steps of:
When detecting that elevator cab door is closed, obtains image capture device and be directed to lift car scene depth collected
Image；
It whether determines in depth image obtained comprising human body target；
When determining comprising human body target, it is based on depth image obtained, determines that the human body of included human body target is high
Spend information；
Whether human height's information determined by judging is lower than preset height threshold value；
When judging that identified human height's information is lower than the preset height threshold value, alarm.
Using the embodiment of the present invention, when detecting that elevator cab door is closed, obtains image capture device and be directed to elevator car
Compartment scene depth image collected, and whether determine in depth image includes human body target, when determining comprising human body target,
It further determines that human height's information of human body target, and combines preset height threshold value, whether judge included human body target
For child, when judging human height's information of above-mentioned human body target lower than preset height threshold value, it may be determined that the human body for being included
Target is child, and can determine and occur only having the case where child takes in lift car at this time, i.e., only child takes elevator
The case where.In the case where monitoring only child's seating elevator, alarm, to warn monitoring personnel to be paid close attention to,
Subsequent, monitoring personnel can carry out respective handling, avoid the child for taking elevator from occurring dangerous, only have to realize in elevator
Manpower is saved in the intelligent monitoring for the case where child takes.Also, when avoiding monitoring personnel and manually being monitored, it is understood that there may be
Because artificial energy is limited or the factors such as human negligence, and the case where monitoring is omitted occurs, the feelings for tragedy occur are in turn resulted in
Condition.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is all made of relevant mode and describes, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (23)
1. a kind of lift car monitoring method, which is characterized in that the described method includes:
When detecting that elevator cab door is closed, obtains image capture device and be directed to lift car scene depth map collected
Picture；
It whether determines in depth image obtained comprising human body target；
When determining comprising human body target, it is based on depth image obtained, determines human height's letter of included human body target
Breath；
Whether human height's information determined by judging is lower than preset height threshold value；
When judging that identified human height's information is lower than the preset height threshold value, alarm.
2. the method according to claim 1, wherein whether including people in determination depth image obtained
The step of body target, comprising:
Characteristics of image is extracted from depth image obtained；
Based on extracted characteristics of image and the first default classifier, whether determine in depth image obtained comprising human body
Target.
3. the method according to claim 1, wherein whether including people in determination depth image obtained
The step of body target, comprising:
Depth image obtained is converted into the point cloud data under world coordinate system；
It whether determines in the point cloud data comprising the corresponding point cloud data of human body target, wherein when judging the point cloud data
In comprising human body target corresponding point cloud data when, characterize in depth image obtained comprising human body target.
4. according to the method described in claim 3, it is characterized in that, whether including human body mesh in the determination point cloud data
The step of marking corresponding point cloud data, comprising:
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and the second default classifier, determine in all kinds of cloud subdatas obtained
It whether include corresponding cloud subdata of human body target.
5. according to the method described in claim 3, it is characterized in that, whether including human body mesh in the determination point cloud data
The step of marking corresponding point cloud data, comprising:
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and default manikin, determining in all kinds of cloud subdatas obtained is
No includes corresponding cloud subdata of human body target.
6. according to the method described in claim 3, it is characterized in that, the depth image that described image acquisition equipment utilization is included
Acquire sub- equipment sampling depth image；
The step of point cloud data depth image obtained is converted under world coordinate system, comprising:
Obtain the parameter information that the depth image acquires sub- equipment, wherein include in the parameter information: focus information, as
Principal point information, mounting height information and setting angle information；
Using the focus information, the principal point information, depth image obtained is converted to the point under device coordinate system
Cloud data, wherein the device coordinate system are as follows: the coordinate system established based on the optical center that the depth image acquires sub- equipment；
The point cloud data under the device coordinate system is turned using the mounting height information and the setting angle information
Turn to the point cloud data under the world coordinate system.
7. the method according to claim 1, wherein the acquisition image capture device is directed to lift car scene
The step of depth image collected, comprising:
It obtains described image acquisition equipment and is directed to lift car scene depth image collected and depth image obtained
Corresponding color image；
The step of whether including human body target in the determination depth image obtained, comprising:
By whether identifying in the color image comprising human body target, determine in the depth image obtained whether include human body mesh
Mark.
8. method according to claim 1-7, which is characterized in that it is based on depth image obtained described,
Before the step of determining the elevation information of included human body target, the method also includes:
When determining comprising human body target, judge whether included human body target is one；
When judging included human body target for one, execute it is described be based on depth image obtained, determine included human body
The step of elevation information of target.
9. method according to claim 1-7, which is characterized in that before described the step of being alarmed, institute
State method further include:
When judging that identified human height's information is lower than the preset height threshold value, the number of included human body target is determined
Amount；
When the quantity of identified included human body target is lower than preset quantity, described the step of being alarmed is executed.
10. method according to claim 1-7, which is characterized in that after described the step of being alarmed, institute
State method further include:
It controls the lift car and is moved to designated floor；
After the lift car is moved to the designated floor, controls the lift car and stop movement；
After obtaining the door open command for being directed to the lift car, controls the lift car and open compartment door.
11. a kind of lift car monitoring device, which is characterized in that described device includes:
First obtains module, is directed to lift car field for when detecting that elevator cab door is closed, obtaining image capture device
Scape depth image collected；
First determining module, for whether determining in depth image obtained comprising human body target；
Second determining module determines included human body for being based on depth image obtained when determining comprising human body target
Human height's information of target；
First judgment module, for judging whether identified human height's information is lower than preset height threshold value；
Alarm module, for alarming when judging that identified human height's information is lower than the preset height threshold value.
12. device according to claim 11, which is characterized in that first determining module is specifically used for
Characteristics of image is extracted from depth image obtained；
Based on extracted characteristics of image and the first default classifier, whether determine in depth image obtained comprising human body
Target.
13. device according to claim 11, which is characterized in that first determining module includes converting unit and determination
Unit；
The converting unit, for depth image obtained to be converted to the point cloud data under world coordinate system；
The determination unit, for whether determining in the point cloud data comprising the corresponding point cloud data of human body target, wherein when
When judging point cloud data corresponding comprising human body target in the point cloud data, characterize in depth image obtained comprising human body
Target.
14. device according to claim 13, which is characterized in that the determination unit is specifically used for
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and the second default classifier, determine in all kinds of cloud subdatas obtained
It whether include corresponding cloud subdata of human body target.
15. device according to claim 13, which is characterized in that the determination unit is specifically used for
The point cloud data is clustered, all kinds of cloud subdatas are obtained；
Based on all kinds of cloud subdatas obtained and default manikin, determining in all kinds of cloud subdatas obtained is
No includes corresponding cloud subdata of human body target.
16. device according to claim 13, which is characterized in that the depth map that described image acquisition equipment utilization is included
As acquiring sub- equipment sampling depth image；
The converting unit, is specifically used for
Obtain the parameter information that the depth image acquires sub- equipment, wherein include in the parameter information: focus information, as
Principal point information, mounting height information and setting angle information；
Using the focus information, the principal point information, depth image obtained is converted to the point under device coordinate system
Cloud data, wherein the device coordinate system are as follows: the coordinate system established based on the optical center that the depth image acquires sub- equipment；
The point cloud data under the device coordinate system is turned using the mounting height information and the setting angle information
Turn to the point cloud data under the world coordinate system.
17. device according to claim 11, which is characterized in that described first obtains module, is specifically used for
It obtains described image acquisition equipment and is directed to lift car scene depth image collected and depth image obtained
Corresponding color image；
First determining module, is specifically used for
By whether identifying in the color image comprising human body target, determine in the depth image obtained whether include human body mesh
Mark.
18. the described in any item devices of 1-17 according to claim 1, which is characterized in that described device further includes the second judgement mould
Block；
Second judgment module determines the height of included human body target for being based on depth image obtained described
Before information, when determining comprising human body target, judge whether included human body target is one；When the included human body mesh of judgement
When being designated as one, second determining module is triggered.
19. the described in any item devices of 1-17 according to claim 1, which is characterized in that described device further includes that third determines mould
Block；
The third determining module, for before described alarmed, human height's information determined by judge is lower than institute
When stating preset height threshold value, the quantity of included human body target is determined；
When the quantity of identified included human body target is lower than preset quantity, the alarm module is triggered.
20. the described in any item devices of 1-17 according to claim 1, which is characterized in that described device further includes the first control mould
Block, the second control module and third control module；
First control module is moved to designated floor for after described alarmed, controlling the lift car；
Second control module, for controlling the lift car after lift car is moved to the designated floor
Stop movement；
The third control module, for controlling the lift car after obtaining the door open command for being directed to the lift car
Open compartment door.
21. a kind of electronic equipment, which is characterized in that including processor and memory, wherein the memory, based on storing
Calculation machine program；
The processor when for executing the computer program stored on memory, realizes that claim 1-10 is any described
Lift car monitoring method.
22. a kind of elevator device characterized by comprising elevator, and the elevator as described in any in claim 11-20
Carriage monitoring device.
23. elevator device according to claim 22 characterized by comprising
Camera for acquiring the depth image in lift car, and is sent to the lift car monitoring device.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710822052.0A CN109484935B (en) | 2017-09-13 | 2017-09-13 | Elevator car monitoring method, device and system |
| PCT/CN2018/101545 WO2019052318A1 (en) | 2017-09-13 | 2018-08-21 | Method, apparatus and system for monitoring elevator car |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710822052.0A CN109484935B (en) | 2017-09-13 | 2017-09-13 | Elevator car monitoring method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109484935A true CN109484935A (en) | 2019-03-19 |
| CN109484935B CN109484935B (en) | 2020-11-20 |
Family
ID=65688677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710822052.0A Active CN109484935B (en) | 2017-09-13 | 2017-09-13 | Elevator car monitoring method, device and system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109484935B (en) |
| WO (1) | WO2019052318A1 (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111031278A (en) * | 2019-11-25 | 2020-04-17 | 广州恒龙信息技术有限公司 | Monitoring method and system based on structured light and TOF |
| CN111689327A (en) * | 2020-06-19 | 2020-09-22 | 迅达（中国）电梯有限公司 | Elevator car control system |
| CN112184737A (en) * | 2020-10-10 | 2021-01-05 | 上海阅目科技有限公司 | Children take alone elevator detection and alarm system based on image analysis |
| CN112257570A (en) * | 2020-10-20 | 2021-01-22 | 江苏濠汉信息技术有限公司 | Method and device for detecting whether safety helmet of constructor is not worn based on visual analysis |
| CN112529953A (en) * | 2020-12-17 | 2021-03-19 | 深圳市普渡科技有限公司 | Elevator space state judgment method and device and storage medium |
| CN112693981A (en) * | 2020-12-17 | 2021-04-23 | 日立楼宇技术（广州）有限公司 | Adjusting method, device and equipment of touch control box and storage medium |
| CN113361410A (en) * | 2021-06-07 | 2021-09-07 | 上海数川数据科技有限公司 | Child detection box filtering algorithm based on grid clustering |
| CN113492405A (en) * | 2021-06-17 | 2021-10-12 | 达闼机器人有限公司 | Robot door opening method and device, storage medium and electronic equipment |
| CN114063572A (en) * | 2020-07-31 | 2022-02-18 | 北京熵行科技有限公司 | Non-sensing intelligent device control method, electronic device and control system |
| CN114180431A (en) * | 2021-12-22 | 2022-03-15 | 佛山市和众电梯技术有限公司 | Car elevator safe operation monitored control system |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113104687A (en) * | 2021-04-14 | 2021-07-13 | 合肥工业大学 | Method and device for judging elevator space state based on computer vision recognition |
| CN113269111B (en) * | 2021-06-03 | 2024-04-05 | 昆山杜克大学 | Video monitoring-based elevator abnormal behavior detection method and system |
| CN113838097B (en) * | 2021-09-29 | 2024-01-09 | 成都新潮传媒集团有限公司 | Camera lens angle deviation detection method, device and storage medium |
| CN114155557B (en) * | 2021-12-07 | 2022-12-23 | 美的集团(上海)有限公司 | Positioning method, positioning device, robot and computer-readable storage medium |
| WO2023245402A1 (en) * | 2022-06-21 | 2023-12-28 | Kone Corporation | A method and a triggering system for triggering an acquisition of image data of an empty elevator car |
| CN116137074A (en) * | 2023-02-22 | 2023-05-19 | 常熟理工学院 | Automatic detection method and system for passengers in elevator car |
| CN115909321B (en) * | 2023-03-08 | 2023-05-12 | 成都睿瞳科技有限责任公司 | Identification method, system and storage medium for elevator car |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5465308A (en) * | 1990-06-04 | 1995-11-07 | Datron/Transoc, Inc. | Pattern recognition system |
| CN101200252A (en) * | 2007-12-10 | 2008-06-18 | 中国科学院计算技术研究所 | Target place optimized dispatching method and system |
| CN103413145A (en) * | 2013-08-23 | 2013-11-27 | 南京理工大学 | Articulation point positioning method based on depth image |
| CN203728342U (en) * | 2014-03-05 | 2014-07-23 | 中国石油大学(华东) | Mistaken stopping preventive system for elevator |
| CN103971135A (en) * | 2014-05-05 | 2014-08-06 | 中国民航大学 | Human body target detection method based on head and shoulder depth information features |
| CN104176570A (en) * | 2013-05-22 | 2014-12-03 | 日立电梯（中国）有限公司 | Control method and control system for lift |
| CN104598890A (en) * | 2015-01-30 | 2015-05-06 | 南京邮电大学 | Human body behavior recognizing method based on RGB-D video |
| CN104751146A (en) * | 2015-04-13 | 2015-07-01 | 中国科学技术大学 | Indoor human body detection method based on 3D (three-dimensional) point cloud image |
| CN105923481A (en) * | 2016-06-30 | 2016-09-07 | 爱默生电梯有限公司 | Elevator capable of preventing risk of children accidental entering |
| CN106315324A (en) * | 2015-06-16 | 2017-01-11 | 奥的斯电梯公司 | Elevator system capable of monitoring child's use and control method for same |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102657532B (en) * | 2012-05-04 | 2014-09-17 | 深圳泰山在线科技有限公司 | Height measuring method and device based on body posture identification |
| WO2016068869A1 (en) * | 2014-10-28 | 2016-05-06 | Hewlett-Packard Development Company, L.P. | Three dimensional object recognition |
| CN105129550B (en) * | 2015-08-03 | 2016-09-14 | 罗普特（厦门）科技集团有限公司 | The elevator safety preventing minor from individually taking controls platform |
| CN206486130U (en) * | 2017-01-19 | 2017-09-12 | 北京北迅电梯公司 | Elevator early warning system |
-
2017
- 2017-09-13 CN CN201710822052.0A patent/CN109484935B/en active Active
-
2018
- 2018-08-21 WO PCT/CN2018/101545 patent/WO2019052318A1/en active Application Filing
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5465308A (en) * | 1990-06-04 | 1995-11-07 | Datron/Transoc, Inc. | Pattern recognition system |
| CN101200252A (en) * | 2007-12-10 | 2008-06-18 | 中国科学院计算技术研究所 | Target place optimized dispatching method and system |
| CN104176570A (en) * | 2013-05-22 | 2014-12-03 | 日立电梯（中国）有限公司 | Control method and control system for lift |
| CN103413145A (en) * | 2013-08-23 | 2013-11-27 | 南京理工大学 | Articulation point positioning method based on depth image |
| CN203728342U (en) * | 2014-03-05 | 2014-07-23 | 中国石油大学(华东) | Mistaken stopping preventive system for elevator |
| CN103971135A (en) * | 2014-05-05 | 2014-08-06 | 中国民航大学 | Human body target detection method based on head and shoulder depth information features |
| CN104598890A (en) * | 2015-01-30 | 2015-05-06 | 南京邮电大学 | Human body behavior recognizing method based on RGB-D video |
| CN104751146A (en) * | 2015-04-13 | 2015-07-01 | 中国科学技术大学 | Indoor human body detection method based on 3D (three-dimensional) point cloud image |
| CN106315324A (en) * | 2015-06-16 | 2017-01-11 | 奥的斯电梯公司 | Elevator system capable of monitoring child's use and control method for same |
| CN105923481A (en) * | 2016-06-30 | 2016-09-07 | 爱默生电梯有限公司 | Elevator capable of preventing risk of children accidental entering |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111031278A (en) * | 2019-11-25 | 2020-04-17 | 广州恒龙信息技术有限公司 | Monitoring method and system based on structured light and TOF |
| CN111689327A (en) * | 2020-06-19 | 2020-09-22 | 迅达（中国）电梯有限公司 | Elevator car control system |
| CN114063572B (en) * | 2020-07-31 | 2024-05-31 | 北京熵行科技有限公司 | Non-perception intelligent device control method, electronic device and control system |
| CN114063572A (en) * | 2020-07-31 | 2022-02-18 | 北京熵行科技有限公司 | Non-sensing intelligent device control method, electronic device and control system |
| CN112184737A (en) * | 2020-10-10 | 2021-01-05 | 上海阅目科技有限公司 | Children take alone elevator detection and alarm system based on image analysis |
| CN112257570A (en) * | 2020-10-20 | 2021-01-22 | 江苏濠汉信息技术有限公司 | Method and device for detecting whether safety helmet of constructor is not worn based on visual analysis |
| CN112529953B (en) * | 2020-12-17 | 2022-05-03 | 深圳市普渡科技有限公司 | Elevator space state judgment method and device and storage medium |
| CN112529953A (en) * | 2020-12-17 | 2021-03-19 | 深圳市普渡科技有限公司 | Elevator space state judgment method and device and storage medium |
| CN112693981A (en) * | 2020-12-17 | 2021-04-23 | 日立楼宇技术（广州）有限公司 | Adjusting method, device and equipment of touch control box and storage medium |
| CN113361410A (en) * | 2021-06-07 | 2021-09-07 | 上海数川数据科技有限公司 | Child detection box filtering algorithm based on grid clustering |
| CN113361410B (en) * | 2021-06-07 | 2022-08-26 | 上海数川数据科技有限公司 | Child detection box filtering algorithm based on grid clustering |
| CN113492405B (en) * | 2021-06-17 | 2022-08-02 | 达闼机器人股份有限公司 | Robot door opening method and device, storage medium and electronic equipment |
| CN113492405A (en) * | 2021-06-17 | 2021-10-12 | 达闼机器人有限公司 | Robot door opening method and device, storage medium and electronic equipment |
| CN114180431A (en) * | 2021-12-22 | 2022-03-15 | 佛山市和众电梯技术有限公司 | Car elevator safe operation monitored control system |
| CN114180431B (en) * | 2021-12-22 | 2023-12-15 | 佛山市和众电梯技术有限公司 | Safety operation monitoring system for elevator car |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019052318A1 (en) | 2019-03-21 |
| CN109484935B (en) | 2020-11-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109484935A (en) | A kind of lift car monitoring method, apparatus and system | |
| CN108446630B (en) | Intelligent monitoring method for airport runway, application server and computer storage medium | |
| WO2020078229A1 (en) | Target object identification method and apparatus, storage medium and electronic apparatus | |
| CN110118973B (en) | Intelligent sensing and identifying method and device for warehouse and electronic equipment | |
| KR102122859B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN109670441A (en) | A kind of realization safety cap wearing knows method for distinguishing, system, terminal and computer readable storage medium | |
| US20160260306A1 (en) | Method and device for automated early detection of forest fires by means of optical detection of smoke clouds | |
| CN104599427A (en) | Intelligent image type fire alarming system for highway tunnel | |
| CN110619277A (en) | Multi-community intelligent deployment and control method and system | |
| CN109766779A (en) | It hovers personal identification method and Related product | |
| US11935378B2 (en) | Intrusion detection methods and devices | |
| CN111488803A (en) | Airport target behavior understanding system integrating target detection and target tracking | |
| CN112183472A (en) | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet | |
| CN108805184B (en) | Image recognition method and system for fixed space and vehicle | |
| CN112270253A (en) | High-altitude parabolic detection method and device | |
| CN110674753A (en) | Theft early warning method, terminal device and storage medium | |
| CN109977840A (en) | A kind of airport scene monitoring method based on deep learning | |
| CN112668478B (en) | Well lid monitoring method and device, electronic equipment and storage medium | |
| CN113965733A (en) | Binocular video monitoring method, system, computer equipment and storage medium | |
| CN111862129A (en) | Virtual fence system for sample sealing storage | |
| CN116597371A (en) | Dangerous object early warning method, system and computer equipment based on image monitoring | |
| CN115861905A (en) | Hotel management system based on Internet of things | |
| CN115953815A (en) | Monitoring method and device for infrastructure site | |
| US11830126B2 (en) | Accurate representation of camera field of view in two-dimensional mapping applications | |
| CN115601712A (en) | Image data processing method and system suitable for field safety measures |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |