CN107187980B - The detection method and detection device of elevator door folding in a kind of lift appliance - Google Patents
The detection method and detection device of elevator door folding in a kind of lift appliance Download PDFInfo
- Publication number
- CN107187980B CN107187980B CN201710384717.4A CN201710384717A CN107187980B CN 107187980 B CN107187980 B CN 107187980B CN 201710384717 A CN201710384717 A CN 201710384717A CN 107187980 B CN107187980 B CN 107187980B
- Authority
- CN
- China
- Prior art keywords
- label
- elevator door
- area
- candidate
- label area
- 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.)
- Expired - Fee Related
Links
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
- B66B5/0018—Devices monitoring the operating condition of the elevator system
Landscapes
- Image Analysis (AREA)
Abstract
本发明实施例公开了一种电梯装置中电梯门开合的检测方法及检测装置，该检测方法通过在电梯装置第一电梯门和第二电梯门上设置第一标签和第二标签，并在轿厢内设置深度相机，来利用深度相机采集第一标签和第二标签图像，获得所述第一标签和第二标签的世界坐标信息，并基于第一标签和第二标签的世界坐标信息获得第一电梯门和第二电梯门之间的开合程度信息，安装方便，方法简单，可以适用于所有的新旧电梯平台，适用范围较广，且在使用过程中，不存在机械或零件的损耗，后续基本不需要人工的维护，且检测精度不会随着使用时间的积累而降低。
The embodiment of the present invention discloses a detection method and a detection device for the opening and closing of an elevator door in an elevator device. The detection method comprises setting a first label and a second label on the first elevator door and the second elevator door of the elevator device, and A depth camera is set in the car to use the depth camera to collect images of the first label and the second label, obtain the world coordinate information of the first label and the second label, and obtain the world coordinate information based on the first label and the second label. The opening and closing degree information between the first elevator door and the second elevator door is easy to install and the method is simple. It can be applied to all new and old elevator platforms, with a wide range of applications, and there is no loss of machinery or parts during use. , the subsequent manual maintenance is basically not required, and the detection accuracy will not decrease with the accumulation of use time.
Description
Technical field
The present invention relates to the detection sides that elevator door in elevator door detection technique field more particularly to a kind of lift appliance opens and closes
Method and detection device.
Background technique
The carriage of usual elevator is required to whether confirmation car door is in completely closed state before up and down motion, with
Ensure the safety of personnel near lift car the inside and elevator.Moreover, when elevator leads to electricity due to mechanical breakdown etc.
There can be very big security risk when terraced door acceleration switch.Therefore, the folding condition of real-time monitoring elevator door, for improving electricity
Terraced safety in utilization has very important significance.
Traditional detection device generally includes gate switch and encoder, wherein the gate switch includes being arranged one
Pin on a door and the hole being arranged on another, pin and hole completely to it is upper when determine that elevator door is in completely enclosed shape
State, the encoder fan the folding condition between door for the mobile motor control two of selectivity, and provide between two fan doors away from
From information.
But above-mentioned detection device, dependent on mechanical configuration, with the extension of use time, machine components can damage
Consumption, subsequent maintenance cost is higher, moreover, above-mentioned detection device is needed in elevator factory, just design is entered, for it is this
The elevator of kind function can not be reequiped easily.
Summary of the invention
In order to solve the above technical problems, the embodiment of the invention provides the detection sides that elevator door in a kind of lift appliance opens and closes
Method and detection device, it is higher to solve the subsequent maintenance cost of existing detection device, in addition, it is desirable to just be designed in elevator factory
The problem of entering, can not easily being reequiped for the elevator of no this function.
To solve the above problems, the embodiment of the invention provides following technical solutions:
The detection method of elevator door folding in a kind of lift appliance, the lift appliance include lift car and elevator door,
The elevator door includes: the first elevator door and the second elevator door being oppositely arranged, and is provided with the first mark on first elevator door
It signs, is provided with the second label on second elevator door, is provided with depth camera in the lift car, the detection method packet
It includes:
The motor area of first label and second label is selected in the display interface of the depth camera image
Domain is as target area；
In the gray level image of depth camera output, edge extracting is carried out to the corresponding region in the target area,
And the edge is formed into the region of closure as the first candidate label area；
In the depth image of depth camera output, processing is split to the corresponding region in the target area,
Obtain at least one second candidate label area；
It, will be with described first by the described first candidate label area and at least one described second candidate label area comparison
The size and location of candidate label area meets the second candidate label area of the first preset condition as third label area；
Calculate world coordinates of each pixel of third label area under camera coordinates system；
World coordinates based on each pixel of third label area under the camera coordinates system calculates described first
The opening and closing degree information of elevator door and the second elevator door.
Optionally, in the gray level image of depth camera output, side is carried out to the corresponding region in the target area
Edge extracts, and includes: as the first candidate label area using the region that the edge forms closure
The edge binary map of the target area is calculated using Canny operator；
In the edge binary map, remove the point that closure is not formed in the gray level image, the side that obtains that treated
Edge binary map；
Connected domain is extracted in edge binary map after treatment, using the boundary rectangle of the connected domain as the first candidate mark
Sign region.
Optionally, in the depth image of depth camera output, the corresponding region in the target area is divided
Processing is cut, obtaining at least one second candidate label area includes:
The segmentation result of target area described in the depth image is obtained using the method for cluster or region growing；
The segmentation result is screened, the second preset condition will be met with first label and second label
Region screen, as the second candidate label area.
Optionally, calculating world coordinates of each pixel of third label area under camera coordinates system includes:
Obtain the seat of the depth information and each pixel of each pixel in the third label area under image coordinate system
It marks (x, y)；
World coordinates of each pixel under camera coordinates system is calculated using following formula:
Wherein, Xworld、Yworld、ZworldIt is the world coordinates under camera coordinates system respectively；X, y is under image coordinate system
Abscissa and ordinate；Deep is the depth value at (x, y) pixel；cx, cyIt is depth camera center under image coordinate system respectively
The coordinate value of point, Tx, TyIt is the size (in millimeters) in the single pixel direction point x and the direction y, f respectivelyx, fyIt is institute respectively
State the focus information on the direction depth camera x and the direction y.
Optionally, the world coordinates based on each pixel of third label area under the camera coordinates system calculates
The opening and closing degree information of first elevator door and the second elevator door includes:
It is averaged to world coordinates of each pixel of third label area under the camera coordinates system, obtains institute
State the world coordinates information of the first tag hub point and the world coordinates information of the second tag hub point；
According to the world coordinates information of the world coordinates information of the first tag hub point and the second tag hub point, meter
Calculate the Euclidean distance of the first tag hub point and the second tag hub point；
First label and institute are subtracted using the Euclidean distance of the first tag hub point and the second tag hub point
The size for stating the second label obtains the opening and closing degree information between first elevator door and second elevator door.
Optionally, the opening and closing degree information between first elevator door and second elevator door includes first electricity
The distance between terraced door and second elevator door and described state open and close speed between the first elevator door and second elevator door
Degree.
The detection device of elevator door folding in a kind of ladder device, the lift appliance includes lift car and elevator door, institute
Stating elevator door includes the first elevator door and the second elevator door being oppositely arranged, and the detection device includes:
The detection device includes:
The first label being set on first elevator door and the second label being set on second elevator door；
The depth camera being set in the lift car；
The processor being electrically connected with the depth camera, the processor are used for: in the display of the depth camera image
Select the moving region of first label and second label as target area in interface；
In the gray level image of depth camera output, edge extracting is carried out to the corresponding region in the target area,
And the edge is formed into the region of closure as the first candidate label area；
In depth camera output depth image, processing is split to the target area, obtains at least one
Second candidate label area；
It, will be with the described first candidate mark by the described first candidate label area and each described second candidate label area comparison
The size and location in label region meets the second candidate label area of the first preset condition as third label area；
Calculate world coordinates of each pixel of third label area under camera coordinates system；
World coordinates based on each pixel of third label area under the camera coordinates system calculates described first
The opening and closing degree information of elevator door and the second elevator door.
Optionally, first label and second label be in the same horizontal line.
Optionally, first preset condition includes:
The difference in areas of the described first candidate label area and the second candidate label area is exhausted under image coordinate system
The 0.2 of predeterminable area area is less than to value, is formulated are as follows: | Sa-Sb| < 0.2min (Sa, Sb), wherein SaFor the first candidate
The area of label area, SbFor the area of the second candidate label area, the predeterminable area is the described first candidate label area
With the lesser region of area in the described second candidate label area；
The y coordinate system difference of the described first candidate label area and the second candidate label area is absolute under image coordinate system
Value is less than twice of the predeterminable area height；It is formulated as | ya-yb| < 2*min (ha,hb)；yaFor the first candidate label
The y-coordinate information in region, ybFor the y-coordinate information of the second candidate label area, haFor on the first candidate direction label area y
Height value, hbFor the height value on the second candidate direction label area y.
Optionally, the opening and closing degree information between first elevator door and second elevator door includes first electricity
The distance between terraced door and second elevator door and described state open and close speed between the first elevator door and second elevator door
Degree.
Compared with prior art, above-mentioned technical proposal has the advantage that
The detection method and detection device that elevator door opens and closes in lift appliance provided by the embodiment of the present invention, can pass through
The first label and the second label are set on the first elevator door and the second elevator door, and depth camera is set in carriage, is come sharp
The first label and the second label image are acquired with depth camera, obtains the world coordinates letter of first label and the second label
Breath, and the folding between the first elevator door of world coordinates information acquisition and the second elevator door based on the first label and the second label
Degree information, easy for installation, method is simple, can be adapted for all new and old elevator platforms, and the scope of application is wider.
Moreover, elevator door opens and closes in lift appliance provided by the embodiment of the present invention detection method and detection device, it can
To provide the distance between the first elevator door and the second elevator door information and folding velocity information simultaneously, facilitate security system to electricity
The operating condition of ladder device is accurately and efficiently monitored and is carried out big data analysis.
In addition, elevator door opens and closes in lift appliance provided by the embodiment of the present invention detection method and detection device, are adopted
With being based on computer vision technique, what the depth information that the result detected is all based on depth camera present frame obtained, and phase
The imaging of machine deviation will not occur with the accumulation of time, therefore, detection method and detection provided by the embodiment of the present invention
Device, it is in use, subsequent not need artificial maintenance, and detection accuracy substantially there is no mechanical or part loss
It will not reduce with the accumulation of usage time.
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 electricity applied by the detection method of elevator door folding in lift appliance provided by one embodiment of the invention
The structural schematic diagram of ladder device；
Fig. 2 is the detection method flow chart of elevator door folding in lift appliance provided by one embodiment of the 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.
In the following description, numerous specific details are set forth in order to facilitate a full understanding of the present invention, but the present invention can be with
Implemented using other than the one described here other way, those skilled in the art can be without prejudice to intension of the present invention
In the case of do similar popularization, therefore the present invention is not limited by the specific embodiments disclosed below.
The embodiment of the invention provides the detection methods that elevator door in a kind of lift appliance opens and closes, as shown in Figure 1, the electricity
Ladder device includes lift car 10 and elevator door 20, and the elevator door 20 includes the first elevator door 21 and the second electricity being oppositely arranged
Terraced door 22, wherein be provided with the first label 31 on first elevator door 21, be provided with the second mark on second elevator door 22
32 are signed, is provided with depth camera (not shown) in the lift car 10, as shown in Fig. 2, this method comprises:
S1: the movement of first label and second label is selected in the display interface of the depth camera image
Region is as target area (i.e. ROI, region of interest).
In embodiments of the present invention, the moving region of the first label is selected in the display interface of the depth camera image
When with the moving region of the second label as target area, the target area can be rectangle, as shown in figure 1 dashed region 40,
Or ellipse or other 2D shapes, the present invention to this and without limitation, specifically depend on the circumstances.
It should be noted that in embodiments of the present invention, in order to reduce the area of the target area, reducing subsequent calculating
The calculation amount of distance between first label and second label, in one embodiment of the invention, first mark
Label and second label are located in same horizontal line, but the present invention is to this and without limitation, specifically depends on the circumstances.
Optionally, in one embodiment of the invention, the movement of first label is completely covered in the target area
The moving region in region and second label, and it is greater than the moving region of first label and the movement of second label
The sum of area in region, it is preferred that in order to reduce the area of the target area as far as possible, the target area is completely covered
The moving region of the moving region of first label and second label, and it is slightly larger than the moving region of first label
With the sum of the area of moving region of second label, but the present invention is to this and without limitation, as long as guaranteeing the target area
The moving region of first label and the moving region of second label is completely covered in domain, and is not less than first label
Moving region and second label the sum of the area of moving region.
S2: in the gray level image of depth camera output, edge is carried out to the corresponding region in the target area and is mentioned
It takes, and the edge is formed into the region of closure as the first candidate label area.
Specifically, in one embodiment of the invention, in the gray level image of depth camera output, to the mesh
It marks the corresponding region in region and carries out edge extracting, and the edge is formed into the region of closure as the first candidate label area packet
It includes:
S21: the edge binary map of the target area is calculated using Canny operator；
S22: in the edge binary map, removing the point that closure is not formed in the gray level image, such as isolated point
Or line segment etc., the edge binary map that obtains that treated.
S23: extracting connected domain in edge binary map after treatment, waits the boundary rectangle of the connected domain as first
Select label area.Wherein, the connected domain refers in the edge binary map do not have the region in hole for intuitive, stringent next
Say Topological Space X become be connection, and if only if one of the following conditions set up: (1) X cannot be expressed as two it is disjoint non-
It is spaced apart the union of collection；(2)A ≠ X or
It should be noted that in embodiments of the present invention, Canny operator be Australian Scientists John F.Canny in
The multistage edge detection algorithm of one proposed in 1986, for extracting the marginal point in image.
It should also be noted that, in embodiments of the present invention, the described first candidate label area is preferably two, right respectively
Answer the region where first label and the region where second label.
S3: in the depth image of depth camera output, place is split to the corresponding region in the target area
Reason obtains at least one second candidate label area.
Specifically, in one embodiment of the invention, in depth map, processing is split to the target area,
Obtaining at least one second candidate label area includes:
S31: the segmentation result of target area described in the depth image is obtained using the method for cluster or region growing；
S32: screening the segmentation result, will meet second with first label or second label and preset
The region of condition screens, as the second candidate label area.
Specifically, in one embodiment of the invention, it is pre- to meet second with first label or second label
If the region of condition at least meets the following conditions: regular shape, the area ratio with first label or second label
Not less than 0.8 and it is not more than 1.2.
It should be noted that in embodiments of the present invention, cluster is that the set of pixel is based on gray value or depth value
The process for multiple classes that divide into several classes forms like object；Region growing is a kind of by the pixel collection with similar quality
Constitute region, detailed process includes: first to look for a pixel (to be denoted as seed picture at random in the region that each needs is divided
Element) as growth starting point；Then with sub-pixel will there is the pixel of same or similar property to close in field around sub-pixel
And into the region where the sub-pixel.
Depth it should also be noted that, in embodiments of the present invention, around the sub-pixel in field with sub-pixel
The absolute value of the difference of value less than 10mm pixel be denoted as with sub-pixel have same or similar property pixel, i.e., with
There is sub-pixel the depth value of the pixel of same or similar property to meet following relationship: | deptha-depthb| < 10mm；
Wherein, depthaIt is the depth value at sub-pixel, depthbBe around sub-pixel in field with sub-pixel have it is identical or
The depth value of the pixel of similar quality.
S4: the described first candidate label area and each described second candidate label area comparison will be waited with described first
The size and location of label area is selected to meet the second candidate label area of the first preset condition as third label area.
It should be noted that the preset condition includes:
The difference in areas of the described first candidate label area and the second candidate label area is exhausted under image coordinate system
The 0.2 of predeterminable area area is less than to value, is formulated are as follows: | Sa-Sb| < 0.2*min (Sa,Sb), wherein SaIt is waited for first
Select the area of label area, SbFor the area of the second candidate label area, the predeterminable area is the described first candidate label area
The lesser region of area in domain and the second candidate label area；
The y coordinate system difference of the described first candidate label area and the second candidate label area is absolute under image coordinate system
Value is less than twice of the predeterminable area height；It is formulated as | ya-yb| < 2*min (ha,hb)；yaFor the first candidate label
The y-coordinate information in region, ybFor the y-coordinate information of the second candidate label area, haFor on the first candidate direction label area y
Height value, hbFor the height value on the second candidate direction label area y.
It should also be noted that, in embodiments of the present invention, the third label area is two, respectively corresponds described the
Region where one label and second label.
S5: world coordinates of each pixel of third label area under camera coordinates system is calculated.
It is image coordinate system respectively it should be noted that generally comprising three kinds of coordinate systems in computer vision field, camera shooting
Machine coordinate system (also referred to as camera coordinates system), world coordinate system.Generally select a reference frame in the environment to describe to image
The position of machine and object, the coordinate system are known as world coordinate system.It is used as the world to sit simultaneously camera coordinate system in the present invention
Mark system, so it is expressed as the world coordinates under camera coordinate system camera (camera coordinates system).
Specifically, in one embodiment of the invention, calculating each pixel of third label area in the camera
World coordinates under coordinate system includes:
S51: the depth information of each pixel and each pixel are obtained in the third label area under image coordinate system
Coordinate (x, y)；
S52: world coordinates of each pixel under camera coordinates system is calculated using following formula:
Wherein, Xworld、Yworld、ZworldIt is the world coordinates under camera coordinates system respectively；X, y is under image coordinate system
Abscissa and ordinate；Deep is the depth value at (x, y) pixel；cx, cyIt is depth camera center under image coordinate system respectively
The coordinate value of point, Tx, TyIt is the size (in millimeters) in the single pixel direction point x and the direction y, f respectivelyx, fyIt is institute respectively
State the focus information on the direction depth camera x and the direction y.
S6: based on world coordinates of each pixel of third label area under the camera coordinates system, described in calculating
The opening and closing degree information of first elevator door and the second elevator door.
Specifically, in one embodiment of the invention, based on each pixel of third label area in the camera
World coordinates under coordinate system, the opening and closing degree information for calculating first elevator door and the second elevator door include:
S61: it is averaged, obtains to world coordinates of each pixel of third label area in the camera coordinates system
To the world coordinates information of the first tag hub point and the world coordinates information of the second tag hub point；
S62: believed according to the world coordinates of the world coordinates information of the first tag hub point and the second tag hub point
Breath calculates the Euclidean distance of the first tag hub point and the second tag hub point；Wherein, Euclidean distance (euclidean
Metric) (also referred to as euclidean metric) is the distance definition generallyd use, is referred to true between two points in m-dimensional space
Actual distance from or vector natural length (i.e. the distance of the point to origin).Euclidean distance in two and three dimensions space is just
It is the actual range between two o'clock.It is in the present invention exactly the actual distance between two points.
S63: first label is subtracted using the Euclidean distance of the first tag hub point and the second tag hub point
With the size of second label, the opening and closing degree information between first elevator door and second elevator door is obtained.
On the basis of the above embodiments, in one embodiment of the invention, first elevator door and described second
Opening and closing degree information between elevator door includes the distance between first elevator door and second elevator door and described
It states and opens and closes speed, specific formula for calculation between the first elevator door and second elevator door are as follows:
Wherein, D is the opening and closing degree (unit is millimeter) of elevator door, i.e., between described first elevator door and the second elevator door
Distance；(Xleft,Yleft,Zleft) and (Xright,Yright,Zright) be respectively the first label and second two labels of label generation
Boundary's coordinate (unit is millimeter)；L is the sum of actual size of first label and the second label (unit is millimeter)；V is institute
Show the speed of related movement of the first label and the second label (specifically, when the movement velocity of first label is V1, the second mark
The movement velocity of label is 0, then V=V1；When the movement velocity of first label is 0, the movement velocity of the second label is V1, then
V=V1；When the movement velocity of first label is V1, the movement velocity of the second label is V1, then V=2V1)；Dt is the time
Variable quantity；DD is the changing value of elevator door opening degree in period dt, i.e. the first elevator door and the second elevator door in period dt
The distance between changing value.
Correspondingly, the embodiment of the invention also provides the detection device that elevator door in a kind of ladder device opens and closes, the elevator
Device includes lift car and elevator door, and the elevator door includes the first elevator door and the second elevator door being oppositely arranged, described
Detection device includes:
The first label being set on first elevator door and the second label being set on second elevator door；
The depth camera being set in the lift car；
The processor being electrically connected with the depth camera, the processor are used for: in the display of the depth camera image
Select the moving region of first label and second label as target area in interface；
In the gray level image of depth camera output, edge extracting is carried out to the corresponding region in the target area,
And the edge is formed into the region of closure as the first candidate label area；
In depth camera output depth image, processing is split to the corresponding region in the target area, is obtained
To at least one the second candidate label area；
It, will be with the described first candidate mark by the described first candidate label area and each described second candidate label area comparison
The size and location in label region meets the second candidate label area of the first preset condition as third label area；
Calculate world coordinates of each pixel of third label area under camera coordinates system；
World coordinates based on each pixel of third label area under the camera coordinates system calculates described first
The opening and closing degree information of elevator door and the second elevator door.
On the basis of the above embodiments, in one embodiment of the invention, the target area can be rectangle,
It can be ellipse or other 2D shapes, the present invention to this and without limitation, specifically depends on the circumstances.
It should be noted that in embodiments of the present invention, in order to reduce the area of the target area, reducing subsequent calculating
The calculation amount of distance between first label and second label, in one embodiment of the invention, first mark
Label and second label are located in same horizontal line, but the present invention is to this and without limitation, specifically depends on the circumstances.
Optionally, in one embodiment of the invention, the movement of first label is completely covered in the target area
The moving region in region and second label, and it is greater than the moving region of first label and the movement of second label
The sum of area in region, it is preferred that in order to reduce the area of the target area as far as possible, the target area is completely covered
The moving region of the moving region of first label and second label, and it is slightly larger than the moving region of first label
With the sum of the area of moving region of second label, but the present invention is to this and without limitation, as long as guaranteeing the target area
The moving region of first label and the moving region of second label is completely covered in domain, and is not less than first label
Moving region and second label the sum of the area of moving region.
Based on any of the above embodiments, in one embodiment of the invention, the processor is being executed to institute
It states the corresponding region picture in target area and carries out edge extracting, and the edge is formed into the region of closure as the first candidate label area
It is specifically used for when domain:
The edge binary map of the target area is calculated using Canny operator；
In the edge binary map, remove the point that closure is not formed in the gray level image, such as isolated point or line
Section etc., the edge binary map that obtains that treated.
Connected domain is extracted in edge binary map after treatment, using the boundary rectangle of the connected domain as the first candidate mark
Sign region.Wherein, the connected domain refers in the edge binary map do not have the region in hole for intuitive, strictly opens up
Flutterring space X to become is connection, is set up and if only if one of the following conditions: (1) X cannot be expressed as two and disjoint non-be spaced apart
The union of collection；(2)A ≠ X or
It should be noted that in embodiments of the present invention, the described first candidate label area is preferably two, respectively correspond
The region where region and second label where first label.
On the basis of the above embodiments, in one embodiment of the invention, the processor is being executed in the depth
It spends in the depth image of camera output, processing is split to the target area, obtain at least one second candidate label area
It is specifically used for when domain:
The segmentation result of target area described in depth map is obtained using the method for cluster or region growing；
The segmentation result is screened, the second preset condition will be met with first label or second label
Region screen, as the second candidate label area.
Specifically, in one embodiment of the invention, it is pre- to meet second with first label or second label
If the region of condition at least meets the following conditions: regular shape, the area ratio with first label or second label
Not less than 0.8 and it is not more than 1.2.
It should be noted that in embodiments of the present invention, cluster is that the set of pixel is based on gray value or depth value
The process for multiple classes that divide into several classes forms like object；Region growing is a kind of by the pixel collection with similar quality
Constitute region, detailed process includes: first to look for a pixel (to be denoted as seed picture at random in the region that each needs is divided
Element) as growth starting point；Then with sub-pixel will there is the pixel of same or similar property to close in field around sub-pixel
And into the region where the sub-pixel.
Depth it should also be noted that, in embodiments of the present invention, around the sub-pixel in field with sub-pixel
The absolute value of the difference of value less than 10mm pixel be denoted as with sub-pixel have same or similar property pixel, i.e., with
There is sub-pixel the depth value of the pixel of same or similar property to meet following relationship: | deptha-depthb| < 10mm；
Wherein, depthaIt is the depth value at sub-pixel, depthbBe around sub-pixel in field with sub-pixel have it is identical or
The depth value of the pixel of similar quality.
Based on any of the above embodiments, in one embodiment of the invention, first preset condition includes:
The difference in areas of the described first candidate label area and the second candidate label area is exhausted under image coordinate system
The 0.2 of predeterminable area area is less than to value, is formulated are as follows: | Sa-Sb| < 0.2*min (Sa,Sb), wherein SaIt is waited for first
Select the area of label area, SbFor the area of the second candidate label area, the predeterminable area is the described first candidate label area
The lesser region of area in domain and the second candidate label area；
The y coordinate system difference of the described first candidate label area and the second candidate label area is absolute under image coordinate system
Value is less than twice of the predeterminable area height；It is formulated as | ya-yb| < 2*min (ha,hb)；yaFor the first candidate label
The y-coordinate information in region, ybFor the y-coordinate information of the second candidate label area, haFor on the first candidate direction label area y
Height value, hbFor the height value on the second candidate direction label area y.
It should be noted that in embodiments of the present invention, the third label area is two, described first is respectively corresponded
Region where label and second label.
Based on any of the above embodiments, in one embodiment of the invention, the processor is executing calculating
It is specifically used for when world coordinates of each pixel of third label area under camera coordinates system:
Obtain the seat of the depth information and each pixel of each pixel in the third label area under image coordinate system
It marks (x, y)；
World coordinates of each pixel under camera coordinates system is calculated using following formula:
Wherein, Xworld、Yworld、ZworldIt is the world coordinates under camera coordinates system respectively；X, y is under image coordinate system
Abscissa and ordinate；Deep is the depth value at (x, y) pixel；cx, cyIt is depth camera center under image coordinate system respectively
The coordinate value of point, Tx, TyIt is the size (in millimeters) in the single pixel direction point x and the direction y, f respectivelyx, fyIt is institute respectively
State the focus information on the direction depth camera x and the direction y.
It is image coordinate system respectively it should be noted that generally comprising three kinds of coordinate systems in computer vision field, camera shooting
Machine coordinate system (also referred to as camera coordinates system), world coordinate system.Generally select a reference frame in the environment to describe to image
The position of machine and object, the coordinate system are known as world coordinate system.It is used as the world to sit simultaneously camera coordinate system in the present invention
Mark system, so it is expressed as the world coordinates under camera coordinate system camera (camera coordinates system).
Based on any of the above embodiments, in one embodiment of the invention, the processor is for being based on
World coordinates of each pixel of third label area under the camera coordinates system calculates first elevator door and second
It is specifically used for when the opening and closing degree information of elevator door:
It is averaged to world coordinates of each pixel of third label area in the camera coordinates system, obtains institute
State the world coordinates information of the first tag hub point and the world coordinates information of the second tag hub point；
According to the world coordinates information of the world coordinates information of the first tag hub point and the second tag hub point, meter
Calculate the Euclidean distance of the first tag hub point and the second tag hub point；Wherein, Euclidean distance (euclidean
Metric) (also referred to as euclidean metric) is the distance definition generallyd use, is referred to true between two points in m-dimensional space
Actual distance from or vector natural length (i.e. the distance of the point to origin).Euclidean distance in two and three dimensions space is just
It is the actual range between two o'clock.It is in the present invention exactly the actual distance between two points.
Based on any of the above embodiments, in one embodiment of the invention, first elevator door and described
Opening and closing degree information between second elevator door include the distance between first elevator door and second elevator door and
Described state opens and closes speed, specific formula for calculation between the first elevator door and second elevator door are as follows:
Wherein, D is the opening and closing degree (unit is millimeter) of elevator door, i.e., between described first elevator door and the second elevator door
Distance；(Xleft, Tleft, Zleft) and (Xright, Tright, Zright) be respectively the first label and second two labels of label generation
Boundary's coordinate (unit is millimeter)；L is the sum of actual size of first label and the second label (unit is millimeter)；V is institute
Show the speed of related movement of the first label and the second label (specifically, when the movement velocity of first label is V1, the second mark
The movement velocity of label is 0, then V=V1；When the movement velocity of first label is 0, the movement velocity of the second label is V1, then
V=V1；When the movement velocity of first label is V1, the movement velocity of the second label is V1, then V=2V1)；Dt is the time
Variable quantity；DD is the changing value of elevator door opening degree in period dt, i.e. the first elevator door and the second elevator door in period dt
The distance between changing value.
From the foregoing, it will be observed that elevator door opens and closes in lift appliance provided in an embodiment of the present invention detection method and detection device,
Can be by the way that the first label and the second label be arranged on the first elevator door and the second elevator door, and depth phase is set in carriage
Machine, Lai Liyong depth camera acquire the first label and the second label image, obtain the world of first label and the second label
Coordinate information, and between the first elevator door of world coordinates information acquisition and the second elevator door based on the first label and the second label
Opening and closing degree information, easy for installation, method is simple, can be adapted for all new and old elevator platforms, and the scope of application is wider.
Moreover, elevator door opens and closes in lift appliance provided by the embodiment of the present invention detection method and detection device, it can
To provide the distance between the first elevator door and the second elevator door information and folding velocity information simultaneously, facilitate security system to electricity
The operating condition of ladder device is accurately and efficiently monitored and is carried out big data analysis.
In addition, elevator door opens and closes in lift appliance provided by the embodiment of the present invention detection method and detection device, are adopted
With being based on computer vision technique, what the depth information that the result detected is all based on depth camera present frame obtained, and phase
The imaging of machine deviation will not occur with the accumulation of time, therefore, detection method and detection provided by the embodiment of the present invention
Device, it is in use, subsequent not need artificial maintenance, and detection accuracy substantially there is no mechanical or part loss
It will not reduce with the accumulation of usage time.
Various pieces are described in a progressive manner in this specification, and what each some importance illustrated is and other parts
Difference, same and similar part may refer to each other between various pieces.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to embodiment illustrated herein, and is to fit to consistent with the principles and novel features disclosed in this article
Widest scope.
Claims (10)
1. the detection method that elevator door opens and closes in a kind of lift appliance, the lift appliance includes lift car and elevator door, institute
Stating elevator door includes: the first elevator door and the second elevator door being oppositely arranged, which is characterized in that is arranged on first elevator door
There is the first label, the second label is provided on second elevator door, is provided with depth camera in the lift car, the detection
Method includes:
The moving region of first label and second label is selected to make in the display interface of the depth camera image
For target area；
In the gray level image of depth camera output, edge extracting is carried out to the corresponding region in the target area, and will
The edge forms the region of closure as the first candidate label area；
In the depth image of depth camera output, processing is split to the corresponding region in the target area, is obtained
At least one second candidate label area；
It, will be with first candidate by the described first candidate label area and at least one described second candidate label area comparison
The size and location of label area meets the second candidate label area of the first preset condition as third label area；
Calculate world coordinates of each pixel of third label area under camera coordinates system；
World coordinates based on each pixel of third label area under the camera coordinates system calculates first elevator
The opening and closing degree information of door and the second elevator door.
2. detection method according to claim 1, which is characterized in that in the gray level image of depth camera output,
Edge extracting is carried out to the corresponding region in the target area, and the edge is formed into the region of closure as the first candidate label
Region includes:
The edge binary map of the target area is calculated using Canny operator；
In the edge binary map, remove the point that closure is not formed in the gray level image, the edge two that obtains that treated
Value figure；
Connected domain is extracted in edge binary map after treatment, using the boundary rectangle of the connected domain as the first candidate label area
Domain.
3. detection method according to claim 1, which is characterized in that in the depth image of depth camera output,
Processing is split to the corresponding region in the target area, obtaining at least one second candidate label area includes:
The segmentation result of target area described in the depth image is obtained using the method for cluster or region growing；
The segmentation result is screened, the area of the second preset condition will be met with first label and second label
Domain screens, as the second candidate label area.
4. detection method according to claim 1, which is characterized in that calculate each pixel of third label area in phase
World coordinates under machine coordinate system includes:
Obtain the depth information and each pixel of each pixel in the third label area under image coordinate system coordinate (x,
y)；
World coordinates of each pixel under camera coordinates system is calculated using following formula:
Wherein, Xworld、Yworld、ZworldIt is the world coordinates under camera coordinates system respectively；X, y is the horizontal seat under image coordinate system
Mark and ordinate；Deep is the depth value at (x, y) pixel；cx, cyIt is depth camera central point under image coordinate system respectively
Coordinate value, Tx, Ty are the size (in millimeters) in the single pixel direction point x and the direction y, f respectivelyx, fyIt is the depth respectively
Spend the focus information on the direction camera x and the direction y.
5. detection method according to claim 1, which is characterized in that based on each pixel of third label area in institute
The world coordinates under camera coordinates system is stated, the opening and closing degree information for calculating first elevator door and the second elevator door includes:
It is averaged to world coordinates of each pixel of third label area under the camera coordinates system, obtains described
The world coordinates information of 1 tag hub point and the world coordinates information of the second tag hub point；
According to the world coordinates information of the world coordinates information of the first tag hub point and the second tag hub point, institute is calculated
State the Euclidean distance of the first tag hub point and the second tag hub point；
First label and described are subtracted using the Euclidean distance of the first tag hub point and the second tag hub point
The sum of the size of two labels obtains the opening and closing degree information between first elevator door and second elevator door.
6. detection method according to claim 5, which is characterized in that first elevator door and second elevator door it
Between opening and closing degree information include the distance between first elevator door and second elevator door and it is described state first electricity
Speed is opened and closed between terraced door and second elevator door.
7. the detection device that elevator door opens and closes in a kind of lift appliance, the lift appliance includes lift car and elevator door, institute
Stating elevator door includes the first elevator door and the second elevator door being oppositely arranged, which is characterized in that
The detection device includes:
The first label being set on first elevator door and the second label being set on second elevator door；
The depth camera being set in the lift car；
The processor being electrically connected with the depth camera, the processor are used for: in the display interface of the depth camera image
It is interior to select the moving region of first label and second label as target area；
In the gray level image of depth camera output, edge extracting is carried out to the corresponding region in the target area, and will
The edge forms the region of closure as the first candidate label area；
The depth camera output depth image in, processing is split to the target area, obtain at least one second
Candidate label area；
It, will be with the described first candidate label area by the described first candidate label area and each described second candidate label area comparison
The size and location in domain meets the second candidate label area of the first preset condition as third label area；
Calculate world coordinates of each pixel of third label area under camera coordinates system；
World coordinates based on each pixel of third label area under the camera coordinates system calculates first elevator
The opening and closing degree information of door and the second elevator door.
8. detection device according to claim 7, which is characterized in that first label and second label are same
On horizontal line.
9. detection device according to claim 7, which is characterized in that first preset condition includes:
The area absolute value of the difference of the described first candidate label area and the second candidate label area under image coordinate system
Less than 0.2 times of predeterminable area area, it is formulated are as follows: | Sa-Sb| < 0.2min (Sa, Sb), wherein SaFor the first candidate mark
Sign the area in region, SbFor the area of the second candidate label area, the predeterminable area be the described first candidate label area and
The lesser region of area in described second candidate label area；
The y coordinate system absolute value of the difference of the described first candidate label area and the second candidate label area is small under image coordinate system
In twice of the predeterminable area height；It is formulated as | ya-yb| < 2*min (ha,hb)；yaFor the first candidate label area
Y-coordinate information, ybFor the y-coordinate information of the second candidate label area, haFor the height on the first candidate direction label area y
Value, hbFor the height value on the second candidate direction label area y.
10. detection device according to claim 7, which is characterized in that first elevator door and second elevator door
Between opening and closing degree information include the distance between first elevator door and second elevator door and described state first
Speed is opened and closed between elevator door and second elevator door.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710384717.4A CN107187980B (en) | 2017-05-26 | 2017-05-26 | The detection method and detection device of elevator door folding in a kind of lift appliance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710384717.4A CN107187980B (en) | 2017-05-26 | 2017-05-26 | The detection method and detection device of elevator door folding in a kind of lift appliance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107187980A CN107187980A (en) | 2017-09-22 |
| CN107187980B true CN107187980B (en) | 2019-03-15 |
Family
ID=59874450
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710384717.4A Expired - Fee Related CN107187980B (en) | 2017-05-26 | 2017-05-26 | The detection method and detection device of elevator door folding in a kind of lift appliance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107187980B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107986126A (en) * | 2017-11-17 | 2018-05-04 | 浙江新再灵科技股份有限公司 | The detecting system and method for a kind of elevator malfunction |
| US11577932B2 (en) | 2018-07-26 | 2023-02-14 | Otis Elevator Company | Elevator component inspection systems |
| EP3599204B1 (en) | 2018-07-26 | 2022-01-12 | Otis Elevator Company | Elevator component inspection system and method |
| CN109607347A (en) * | 2019-01-08 | 2019-04-12 | 浙江新再灵科技股份有限公司 | A kind of elevator door condition detection method and system based on image histogram |
| CN110127479B (en) * | 2019-04-17 | 2020-08-21 | 浙江工业大学 | Elevator door switch abnormity detection method based on video analysis |
| CN110589648B (en) * | 2019-09-03 | 2021-04-16 | 浙江新再灵科技股份有限公司 | Vertical ladder door label detection method based on deep learning |
| CN110642113B (en) * | 2019-10-21 | 2021-04-16 | 北京猎户星空科技有限公司 | Elevator door state detection method and device |
| CN115830364A (en) * | 2022-11-07 | 2023-03-21 | 浙江大华技术股份有限公司 | State detection method of opening and closing device, computer equipment and storage device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101585478A (en) * | 2008-05-23 | 2009-11-25 | 株式会社日立制作所 | Safety device and security control method for elevator door |
| CN102036899A (en) * | 2008-05-22 | 2011-04-27 | 奥蒂斯电梯公司 | Video-based system and method of elevator door detection |
| CN202390065U (en) * | 2011-12-02 | 2012-08-22 | 广州都盛机电有限公司 | Elevator dynamic image recognition and control system |
| CN103010883A (en) * | 2011-09-28 | 2013-04-03 | 鸿富锦精密工业（深圳）有限公司 | Elevator safety control device and working method thereof |
| CN106204659A (en) * | 2016-07-26 | 2016-12-07 | 浙江捷尚视觉科技股份有限公司 | Elevator switch door detection method based on light stream |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5772156B2 (en) * | 2011-04-01 | 2015-09-02 | 三菱電機株式会社 | Door system and elevator equipment |
| JP5969148B1 (en) * | 2016-01-13 | 2016-08-17 | 東芝エレベータ株式会社 | Elevator system |
-
2017
- 2017-05-26 CN CN201710384717.4A patent/CN107187980B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102036899A (en) * | 2008-05-22 | 2011-04-27 | 奥蒂斯电梯公司 | Video-based system and method of elevator door detection |
| CN101585478A (en) * | 2008-05-23 | 2009-11-25 | 株式会社日立制作所 | Safety device and security control method for elevator door |
| CN103010883A (en) * | 2011-09-28 | 2013-04-03 | 鸿富锦精密工业（深圳）有限公司 | Elevator safety control device and working method thereof |
| CN202390065U (en) * | 2011-12-02 | 2012-08-22 | 广州都盛机电有限公司 | Elevator dynamic image recognition and control system |
| CN106204659A (en) * | 2016-07-26 | 2016-12-07 | 浙江捷尚视觉科技股份有限公司 | Elevator switch door detection method based on light stream |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107187980A (en) | 2017-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107187980B (en) | The detection method and detection device of elevator door folding in a kind of lift appliance | |
| CN103268480B (en) | A kind of Visual Tracking System and method | |
| CN117593592B (en) | Intelligent scanning and identifying system and method for foreign matters at bottom of vehicle | |
| CN104680555B (en) | Cross the border detection method and out-of-range monitoring system based on video monitoring | |
| KR101593187B1 (en) | Device and method surveiling innormal behavior using 3d image information | |
| WO2016055031A1 (en) | Straight line detection and image processing method and relevant device | |
| CN102194443B (en) | Display method and system for window of video picture in picture and video processing equipment | |
| CN105913441B (en) | A shadow removal method for improving object detection performance in video | |
| CN101315701A (en) | Moving Target Image Segmentation Method | |
| CN103530874A (en) | People flow counting method on basis of Kinect | |
| AU2020300067B2 (en) | Layered motion representation and extraction in monocular still camera videos | |
| CN105160649A (en) | Multi-target tracking method and system based on kernel function unsupervised clustering | |
| CN108320300A (en) | A kind of space-time context visual tracking method of fusion particle filter | |
| US11747278B2 (en) | Deposit detection device for detecting a partial covering location, a not-adhere location, and a diffuse reflection location | |
| KR20170015299A (en) | Method and apparatus for object tracking and segmentation via background tracking | |
| CN108446586B (en) | Method for detecting specific action of train driver | |
| CN104866842B (en) | A kind of human body target intrusion detection method | |
| CN108876820A (en) | A kind of obstruction conditions based on average drifting move down object tracking method | |
| JP4156084B2 (en) | Moving object tracking device | |
| CN103077534B (en) | Spatiotemporal object moving target detecting method | |
| CN103577833A (en) | Abnormal intrusion detection method based on motion template | |
| CN109118516A (en) | A kind of target is from moving to static tracking and device | |
| Huang et al. | Target motion compensation with optical flow clustering during visual tracking | |
| CN104240268B (en) | A kind of pedestrian tracting method based on manifold learning and rarefaction representation | |
| Li et al. | A shadow detection method based on improved Gaussian Mixture Model |
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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 323000 room 303-5, block B, building 1, No. 268, Shiniu Road, nanmingshan street, Liandu District, Lishui City, Zhejiang Province Patentee after: Zhejiang Guangpo Intelligent Technology Co.,Ltd. Address before: Hangzhou City, Zhejiang province 310030 Xihu District three Town Shi Xiang Road No. 859 Zijin and building 3 building 1301-1 room Patentee before: HANGZHOU GENIUS PROS TECHNOLOGY Co.,Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190315 |