CN118255226A - Computer vision-based elevator non-flat-bed door fault identification method - Google Patents
Computer vision-based elevator non-flat-bed door fault identification method Download PDFInfo
- Publication number
- CN118255226A CN118255226A CN202410477476.8A CN202410477476A CN118255226A CN 118255226 A CN118255226 A CN 118255226A CN 202410477476 A CN202410477476 A CN 202410477476A CN 118255226 A CN118255226 A CN 118255226A
- Authority
- CN
- China
- Prior art keywords
- elevator
- floor
- height difference
- adjusting mechanism
- infrared
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000007246 mechanism Effects 0.000 claims abstract description 74
- 238000005516 engineering process Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 19
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 238000004904 shortening Methods 0.000 claims abstract description 19
- 238000001514 detection method Methods 0.000 claims description 29
- 230000007704 transition Effects 0.000 claims description 24
- 238000012937 correction Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 description 5
- 238000003708 edge detection Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000011418 maintenance treatment Methods 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Maintenance And Inspection Apparatuses For Elevators (AREA)
Abstract
The invention discloses a computer vision-based elevator non-flat landing door fault identification method, and relates to the technical field of elevator fault identification; the technical key points are as follows: the method comprises the steps of utilizing an infrared technology to effectively judge the elevator non-flat landing door faults, combining a binocular camera unit, and identifying fault information by adopting a stereoscopic vision technology to ensure the accuracy of acquiring the height difference between the elevator and the corresponding floor, so that the corresponding instruction can be conveniently made according to the situation; the height difference generated under the non-flat-bed door opening state is calculated and acquired by utilizing a three-dimensional matching algorithm, a data calculation model is built according to the height difference and the original initial length of the gradient adjusting mechanism, and the required elongation pre-estimation or shortening pre-estimation of the gradient adjusting mechanism is generated, so that a longitudinal cylinder in the gradient adjusting mechanism can be conveniently and rapidly and effectively adjusted according to the corresponding pre-estimation, the elevator can be used in an emergency, and the safety of passengers is ensured.
Description
Technical Field
The invention relates to the technical field of elevator fault identification, in particular to a computer vision-based elevator non-flat-bed door opening fault identification method.
Background
Elevator fault identification refers to identifying and judging possible problems of an elevator system by monitoring abnormal conditions or fault signals occurring in the operation of an elevator; these problems may relate to aspects of the mechanical parts, electrical systems, control systems, etc. of the elevator, and the fault identification process generally comprises the following aspects: abnormality monitoring, fault signal identification, fault classification, fault positioning and the like; abnormality monitoring: the elevator system can monitor the running state of the elevator in real time through a sensor or a monitoring device, and the running state comprises information such as the speed, the acceleration, the position, the door opening and closing state and the like of the elevator; when data which is inconsistent with normal conditions is detected, the system marks the data as abnormal; and (3) fault signal identification: the elevator system can recognize through a preset fault signal mode according to the monitored abnormal data; the modes can be rules and models which are defined in advance or algorithms which are trained based on techniques such as machine learning, artificial intelligence and the like; fault classification: after the fault signal is identified, the system classifies the fault signal and determines which type of fault the system belongs to; this facilitates subsequent maintenance and handling; fault location: after determining the type of the fault, the system further locates the specific position of the fault; this may need to be done by further detection, investigation, or by using the self-diagnostic function of the system itself.
The prior application publication number is: CN112801072a, named as a computer vision-based elevator non-flat-bed door fault recognition device and method, includes: the video acquisition module acquires elevator sill images; the acceleration monitoring module is used for monitoring the running acceleration of the elevator so as to pre-judge whether the elevator enters a deceleration and elevator stopping state; the model training module comprises a sill groove target detection model training unit and a non-flat layer door opening classification model training unit and is used for training a model; the detection and identification module comprises an image illumination self-adaptive correction unit, a sill groove detection unit, an image angle self-adaptive correction unit and a non-flat layer door opening identification unit; in the scheme, when the device discovers that the elevator is ready to stop, a detection and identification model is loaded, target images of the elevator car door and the elevator sill groove are analyzed, the elevator door opening state and amplitude are detected, and the elevator door opening faults can be detected and identified in real time, efficiently and accurately by carrying out two categories of flat-layer door opening and non-flat-layer door opening on the target images, but the faults are not regulated or emergency maintenance treatment is not carried out in the scheme.
According to the file and the prior art, if the elevator has a non-flat-bed door opening fault, a computer vision recognition technology is generally adopted to analyze the image, then other algorithms are combined to judge the non-flat-bed door opening fault, the operation of stopping and overhauling is generally carried out when the fault is judged to occur, passengers in the elevator need to pass through the step surface if the passengers want to go out from the elevator, the height difference generated between the elevator and the floor corresponding to the floor surface is the so-called step surface due to the fault, the passengers can fall or fall down when passing through the step surface, if the elevator needs to be used in an emergency, the formed step surface is unfavorable for the passengers, and the use danger of the elevator is increased.
Disclosure of Invention
(One) solving the technical problems
Aiming at the defects of the prior art, the invention provides a computer vision-based elevator non-flat-bed door opening fault identification method, which utilizes an infrared technology to effectively judge elevator non-flat-bed door opening faults, combines a binocular camera unit, and adopts a stereoscopic vision technology to identify fault information so as to ensure the accuracy of acquiring the height difference between an elevator and a corresponding floor and facilitate the follow-up corresponding instruction according to the situation; the height difference generated under the non-flat-bed door opening state is calculated and acquired by utilizing a three-dimensional matching algorithm, a data calculation model is built according to the height difference and the original initial length of the gradient adjusting mechanism, and the required elongation pre-estimation or shortening pre-estimation of the gradient adjusting mechanism is generated, so that a longitudinal cylinder in the gradient adjusting mechanism can be conveniently and rapidly and effectively adjusted according to the corresponding pre-estimation, the elevator can be used in an emergency, the safety of passengers is guaranteed, and the problem in the background technology is solved.
(II) technical scheme
In order to achieve the above purpose, the invention is realized by the following technical scheme:
A computer vision-based elevator non-flat-bed door fault identification method comprises the following steps:
S1, assembling a gradient adjusting mechanism at the bottom of an elevator, and arranging an infrared detection unit at a position where the gradient adjusting mechanism is level with a corresponding floor, judging whether an infrared signal in the infrared detection unit is received after the elevator reaches the corresponding floor, and if the infrared signal can be received, not responding; otherwise, sending out an early warning signal;
s2, triggering the binocular shooting unit under the condition of receiving the early warning signal, and identifying fault information by adopting a stereoscopic vision technology to obtain the height difference generated between the elevator and the corresponding floor;
s3, comparing the obtained height difference between the elevator and the corresponding floor with a preset standard threshold value, and if the absolute value of the height difference exceeds the standard threshold value, sending out a warning signal; otherwise, triggering and adjusting an emergency instruction;
S4, executing an emergency adjustment instruction, when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be positive according to the built rule engine, sending an elongation signal to the gradient adjustment mechanism, building a data calculation model according to the height difference generated between the elevator and the corresponding floor, and generating an elongation pre-estimation L Long length of the gradient adjustment mechanism;
when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be negative, a shortening signal is sent to the gradient adjusting mechanism, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor and the collected initial length of the gradient adjusting mechanism, and a shortening pre-estimation L Short length of the gradient adjusting mechanism is generated.
Further, the gradient adjusting mechanism comprises a longitudinal cylinder and a transition plate; the vertical cylinder is assembled in the elevator floor, and the vertical cylinder is located the transition board below all the time, the output of vertical cylinder is articulated mutually with the transition board lower surface, and the one end of transition board is connected through setting up the pivot with elevator floor inner wall both sides, the other end of transition board is towards the floor.
Further, the infrared detection unit comprises an infrared emitter and an infrared receiver, the infrared generator is configured on the surface of the corresponding floor, the infrared receiver is embedded in the end face of the transition plate facing the floor, each time the elevator reaches the corresponding floor, the infrared generator emits an infrared signal, if the infrared receiver receives the infrared signal, the elevator is flush with the floor, and if the infrared receiver does not receive the infrared signal, the elevator does not open the door in a flat floor.
Further, the binocular camera unit that triggers is installed in the binocular camera on elevator upper portion, and the shooting face of binocular camera covers the horizontal ground of elevator gate and floor, acquires the image of elevator gate and floor horizontal ground, and the elevator gate represents left image, and stair horizontal ground represents right image.
Further, the process of acquiring the height difference generated between the elevator and the corresponding floor is as follows:
Stereoscopic vision matching: the method comprises the steps of performing stereo matching on images acquired by a binocular camera unit, namely matching corresponding pixel points in left and right images to obtain parallax information between the left and right images, wherein the parallax information represents position differences of objects in the left and right images, and depth information of the objects corresponding to the left and right images from a camera is obtained through parallax calculation;
and (3) calculating a height difference: calculating a height difference by using depth information obtained by stereoscopic vision; the height difference is obtained by calculating the difference between the depth values of the elevator doorway and the floor level ground according to the depth information in the left image and the right image, and the formula is as follows:
hr＝S Ladder -S Layer(s)
wherein hr represents the height difference, S Ladder represents the distance between the binocular camera and the floor of the elevator, S Layer(s) represents the distance between the binocular camera and the floor of the corresponding elevator, if hr is positive, the elevator door is lower than the depth of the floor of the corresponding elevator, and if hr is negative, the elevator door is higher than the depth of the floor of the corresponding elevator.
Further, the stereo matching algorithm includes any one of disparity mapping and block matching.
Further, the preset standard threshold value represents the maximum value of the height difference that the gradient adjusting mechanism can handle, and the warning signal sent in S3 is different from the warning signal sent in S1 in the expression mode.
Further, the equation according to which the elongation predicted amount L Long length of the gradient adjusting mechanism is generated is as follows:
Wherein G is a constant correction coefficient, the value range of G is 0-1, and L 0 represents the initial length of the gradient adjusting mechanism.
Further, the equation according to which the elongation predicted amount L Short length of the gradient adjusting mechanism is generated is as follows:
Wherein G is a constant correction coefficient, the value range of G is 0-1, and L 0 represents the initial length of the gradient adjusting mechanism.
A computer vision-based elevator non-flat-bed door fault identification system comprises a signal judgment module, a fault identification module, a comparison judgment module and an early warning adjustment module;
The signal judging module is used for assembling the gradient adjusting mechanism at the bottom of the elevator, arranging an infrared detection unit at the position where the gradient adjusting mechanism is level with the corresponding floor, judging whether an infrared signal in the infrared detection unit is received after the elevator reaches the corresponding floor, and if the infrared signal can be received, not responding; otherwise, sending out an early warning signal;
the fault identification module triggers the binocular shooting unit under the condition of receiving the early warning signal, and identifies fault information by adopting a stereoscopic vision technology to obtain the height difference generated between the elevator and the corresponding floor;
the comparison judging module is used for comparing the obtained height difference generated between the elevator and the corresponding floor with a preset standard threshold value, and sending out a warning signal if the absolute value of the height difference exceeds the standard threshold value; otherwise, triggering and adjusting an emergency instruction;
The early warning adjustment module executes an emergency adjustment instruction, when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be positive according to the built rule engine, an elongation signal is sent to the gradient adjustment mechanism, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor, and an elongation pre-estimation L Long length of the gradient adjustment mechanism is generated;
when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be negative, a shortening signal is sent to the gradient adjusting mechanism, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor and the collected initial length of the gradient adjusting mechanism, and a shortening pre-estimation L Short length of the gradient adjusting mechanism is generated.
(III) beneficial effects
The invention provides a computer vision-based elevator non-flat landing door fault identification method, which has the following beneficial effects:
1. The infrared detection unit is adopted, the infrared technology is utilized to effectively judge the elevator non-flat-bed door opening fault, the binocular camera unit is combined to identify fault information, the stereoscopic vision technology is adopted to ensure that the accuracy of the height difference generated between the elevator and the corresponding floor is ensured to be obtained, the corresponding instruction is convenient to be made according to the condition subsequently, when the absolute value of the height difference exceeds the standard threshold value, the elevator is stopped for maintenance, otherwise, the elevator can be ensured to be used in an emergency, and the practicability and the flexibility of the overall design scheme are reflected;
2. The binocular camera unit is used for calculating and acquiring the height difference generated under the non-flat landing door opening state by utilizing the stereo matching algorithm, a data calculation model is built according to the height difference and the original initial length of the gradient adjusting mechanism, the required elongation pre-estimation or shortening pre-estimation of the gradient adjusting mechanism is generated, the longitudinal cylinder in the gradient adjusting mechanism can be conveniently and rapidly and effectively adjusted according to the corresponding pre-estimation, the elevator can be used in an emergency mode, a step surface is prevented from being formed due to the difference value generated between the elevator and the floor corresponding to the floor, and the safety of passengers is guaranteed.
Drawings
Fig. 1 is a flowchart showing the overall steps of a method for identifying a failure of an elevator door in a non-flat landing door in the present invention;
fig. 2 is a schematic diagram of a modular structure of an elevator non-flat-bed door failure recognition system according to the present invention;
Fig. 3 is a schematic view of the whole structure of the gradient adjusting mechanism in the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1: referring to fig. 1-3, the present embodiment provides a method for identifying a failure of an elevator door in a non-flat floor based on computer vision, comprising the following steps:
S1, assembling a gradient adjusting mechanism at the bottom of an elevator, arranging an infrared detection unit at the position where the gradient adjusting mechanism is level with a corresponding floor, judging whether an infrared signal in the infrared detection unit is received after the elevator reaches the corresponding floor, and if the infrared signal can be received, indicating that the elevator is not abnormal in operation, and enabling the system to not respond; if the infrared signal can not be received, the abnormal operation of the elevator is indicated, the fault of the non-flat landing door of the elevator is generated, and an early warning signal is sent;
the gradient adjusting mechanism comprises a longitudinal cylinder 101 and a transition plate 102;
The longitudinal air cylinder 101 is assembled in the elevator floor, the longitudinal air cylinder 101 is always positioned below the transition plate 102, the output end of the longitudinal air cylinder 101 is hinged with the surface of the transition plate 102, one end of the transition plate 102 is connected with two sides of the inner wall of the elevator floor through a rotating shaft, and the other end of the transition plate faces to floors;
The longitudinal cylinder 101 functions as: the transition plate 102 is driven to rotate, so that the transition plate 102 can cope with the state of the elevator non-flat landing door, and the step surface generated between the elevator and the floor due to the height difference can be solved by the transition plate 102 in an inclined state; a plurality of brushes are further arranged on the end face, facing the floors, of the transition plate 102, and gaps generated between the transition plate 102 and the floors can be blocked to a certain extent; the surface of the elevator floor is provided with a groove for placing the transition plate 102 in the unused state, referring to the state shown in the middle drawing in fig. 3;
the infrared detection unit comprises an infrared emitter and an infrared receiver, the infrared emitter is configured on the surface of the corresponding floor, the infrared receiver is embedded in the end face of the transition plate 102 facing the floor, each time the elevator abuts against the corresponding floor, the infrared generator emits an infrared signal, if the infrared receiver receives the infrared signal, the state of the elevator and the floor is shown in the middle drawing in fig. 3, if the infrared receiver fails to receive the infrared signal, the state of the elevator and the floor is not mutually flush, and the state shown in the upper drawing or the lower drawing in fig. 3 is possibly shown, namely, one or two expression forms in the non-flat-layer door opening state are shown;
s2, triggering the binocular shooting unit under the condition of receiving the early warning signal, and identifying fault information by adopting a stereoscopic vision technology to obtain the height difference generated between the elevator and the corresponding floor;
The triggered binocular camera unit is a binocular camera arranged at the upper part of the elevator, the shooting surface of the binocular camera covers the elevator gate and the horizontal ground of the floor, images of the elevator gate and the horizontal ground of the floor are obtained, the elevator gate represents a left image, and the horizontal ground of the stair represents a right image;
The process of obtaining the height difference generated between the elevator and the corresponding floor is as follows:
Stereoscopic vision matching: the method comprises the steps of performing stereo matching on images acquired by a binocular camera unit, namely matching corresponding pixel points in left and right images to obtain parallax information between the left and right images, wherein the parallax information represents position differences of objects in the left and right images, and depth information of the objects corresponding to the left and right images from a camera is obtained through parallax calculation;
common stereo matching algorithms include disparity mapping and block matching;
and (3) calculating a height difference: calculating a height difference by using depth information obtained by stereoscopic vision;
The height difference is obtained by calculating the difference between the depth values of the elevator doorway and the floor level according to the depth information in the left and right images, and the formula is as follows:
hr＝S Ladder -S Layer(s)
Wherein hr represents the height difference, S Ladder represents the distance between the binocular camera and the floor of the elevator, S Layer(s) represents the distance between the binocular camera and the floor of the corresponding floor, if hr is positive, the door of the elevator is lower than the depth of the floor of the corresponding floor, the state is shown in the lower graph in fig. 3, if hr is negative, the door of the elevator is higher than the depth of the floor of the corresponding floor, the state is shown in the upper graph in fig. 3;
It should be noted that, the installation position of the binocular camera needs to consider the coverage range of the visual angle and the effect of stereo matching; generally, the camera should be mounted in place inside the elevator to ensure that images of the elevator doorway and floor level can be captured and that a proper view angle is maintained;
For example: assuming that the binocular camera is arranged at the upper part of the elevator and faces the elevator gate and the floor level ground, depth information of the elevator gate and the floor level ground can be obtained through a stereo matching technology, and the height difference is obtained through calculating the difference of depth values of the binocular camera and the floor level ground; for example, the depth of the elevator doorway is 1.5 meters and the depth of the floor level is 0.8 meters, then the difference in height is 0.7 meters.
If the infrared detection unit in the step S1 is not used, in the step S2, a horizontal line detection technology can be used for judging whether the elevator gate and the floor horizontal ground are on the same horizontal line; in the matched stereo images, whether the elevator doorway and the floor level ground are on the same horizontal line or not can be judged by detecting the horizontal line; the horizontal line detection can be realized through edge detection and straight line fitting, and the common method comprises Hough transformation and RANSAC algorithm, the horizontal line detection technology is not used in the embodiment, so that the operation capacity of binocular shooting is reduced, the binocular shooting is ensured to be only used for acquiring the height difference generated between an elevator and a corresponding floor, and the accuracy of the height difference data can be ensured to a certain extent;
It should be noted that, the stereo matching technology refers to determining the position of an object in a three-dimensional space by comparing parallax information of corresponding pixel points in the binocular image; the following is a specific description of stereo matching technology:
Stereo matching algorithm: the stereo matching algorithm aims at finding similar pixel point corresponding relations in left and right images, and the common algorithm comprises the following steps: dividing an image into small blocks based on block matching of windows, and finding a corresponding relation by comparing the similarity between the blocks, wherein common algorithms comprise mean value difference, normalized cross correlation and the like; matching based on local features: matching by using local feature descriptors in the image, such as SIFT, SURF and the like; matching based on global optimization: obtaining an optimal matching result through a global optimization method such as dynamic planning, graph cutting and the like;
disparity map: after the stereo matching is carried out, a parallax image can be obtained, wherein each pixel point represents the position difference of the corresponding object in the left image and the right image, and the parallax image can intuitively display the depth information of the object;
Parameter adjustment and optimization: many parameters in the stereo matching algorithm need to be adjusted, such as window size, matching cost function and the like; adjusting these parameters can affect the accuracy and stability of the matching results;
The horizontal line detection technology is a technology for detecting a horizontal line in an image and is used for judging whether an object is positioned on the same horizontal line or not; the following is a specific description of the horizontal line detection technique:
Edge detection: firstly, carrying out edge detection on an image so as to find possible straight line candidates;
and (3) straight line detection: selecting possible straight lines from the edge detection result by using Hough transformation or other straight line detection algorithms;
Straight line fitting: fitting the detected straight line to obtain more accurate straight line parameters;
horizontal line screening: screening out a straight line in the horizontal direction, namely a horizontal line, according to the slope of the straight line or an included angle between the straight line and the horizontal direction;
parameter adjustment and optimization: some parameters in the horizontal line detection algorithm need to be adjusted, such as a threshold value of the edge detection algorithm, the accuracy of straight line fitting and the like.
Specifically, through adopting infrared detection unit, utilize infrared technique to accomplish the effective judgement to elevator non-flat layer open door trouble to combine binocular camera unit, adopt stereoscopic vision technique to discern trouble information, in order to ensure to obtain the accuracy that produces the difference in height between elevator and the corresponding floor, be convenient for follow-up according to the circumstances and make corresponding instruction, when the absolute value of difference in height exceeds standard threshold, then shut down and wait to repair, otherwise, then can also ensure that the elevator can carry out emergent use, embodied whole design's practicality and flexibility.
S3, comparing the obtained height difference between the elevator and the corresponding floor with a preset standard threshold, if the absolute value of the height difference exceeds the standard threshold, sending out a warning signal, and executing a shutdown to-be-repaired instruction; if the absolute value of the height difference does not exceed the standard threshold value, triggering and adjusting an emergency instruction to ensure that the current elevator can be normally used;
The preset standard threshold value represents the maximum value of the height difference which can be handled by the gradient adjusting mechanism, and the standard threshold value is set according to historical data or actual conditions;
The mode adopted by the alarm signal sent out by the position is different from that adopted by the early warning signal, and the alarm can be carried out by flashing of different indicator lamps, for example: the warning signal at the position can warn in a manner of flashing a red warning lamp, and the warning signal can warn in a manner of flashing a yellow warning lamp;
S4, executing an emergency adjustment instruction, sending an extension signal to the gradient adjustment mechanism when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be positive according to the built rule engine, and sending a shortening signal to the gradient adjustment mechanism when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be negative;
If an elongation signal is obtained, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor, and an elongation pre-estimation L Long length of the gradient adjusting mechanism is generated; if a shortening signal is obtained, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor and the initial length of the collected gradient adjusting mechanism, and a shortening pre-estimation L Short length of the gradient adjusting mechanism is generated; the gradient adjustment mechanism performs an adjustment operation according to the obtained elongation preset amount L Long length or shortening preset amount L Short length .
Wherein, the equation according to which the elongation pre-estimation L Long length of the gradient adjusting mechanism is generated is as follows:
Wherein G is a constant correction coefficient, the specific value of which can be adjusted and set by a user or generated by fitting an analytical function, and the value range of G is 0 to 1, l 0 represents the initial length of the gradient adjusting mechanism, that is, the initial elongation of the longitudinal cylinder 101 in the gradient adjusting mechanism, for example: when the initial length L 0 of the gradient adjusting mechanism is 5, the height difference hr is 10, and the value of g is 0.7, the estimated elongation L Long length of the gradient adjusting mechanism is 4.326, and the units of the estimated elongation L Long length are unified to cm, so that the estimated elongation of the gradient adjusting mechanism is 4.326cm.
The formula for generating the shortening pre-estimate L Short length of the gradient adjustment mechanism is the same as the formula for generating the lengthening pre-estimate L Long length of the gradient adjustment mechanism, and the hr 2 is set so that both the shortening pre-estimate L Short length and the lengthening pre-estimate L Long length are positive numbers;
After the gradient adjustment mechanism performs adjustment operation according to the obtained elongation preset value L Long length or the shortening preset value L Short length , the binocular camera unit may be used to perform verification, if an infrared signal is received, the gradient adjustment mechanism may adjust accurately, if an infrared signal is not received, the gradient adjustment mechanism may adjust accurately, the data calculation model needs to be built again, and different constant correction coefficients G need to be selected for adjustment until the infrared signal is received.
Specifically, the binocular camera unit is used for calculating and acquiring the height difference generated under the non-flat-bed door opening state by utilizing the stereo matching algorithm, a data calculation model is built according to the height difference and the original initial length of the gradient adjusting mechanism, the required elongation pre-estimated value or the shortening pre-estimated value of the gradient adjusting mechanism is generated, the longitudinal cylinder 101 in the gradient adjusting mechanism is convenient to quickly and effectively adjust according to the corresponding pre-estimated value, the elevator can be used in an emergency mode, the step surface formed by the difference value generated by the elevator and the corresponding floor surface is avoided, and the safety of passengers is guaranteed.
Example 2: referring to fig. 2, based on embodiment 1, the present embodiment further provides a system for identifying a failure of an elevator non-flat-bed door based on computer vision, where the system includes a signal judging module, a failure identifying module, a comparing and judging module, and an early warning adjusting module;
The signal judging module is used for assembling the gradient adjusting mechanism at the bottom of the elevator, arranging an infrared detection unit at the position where the gradient adjusting mechanism is level with the corresponding floor, judging whether an infrared signal in the infrared detection unit is received after the elevator reaches the corresponding floor, and if the infrared signal can be received, not responding; otherwise, sending out an early warning signal;
the fault identification module triggers the binocular shooting unit under the condition of receiving the early warning signal, and identifies fault information by adopting a stereoscopic vision technology to obtain the height difference generated between the elevator and the corresponding floor;
the comparison judging module is used for comparing the obtained height difference generated between the elevator and the corresponding floor with a preset standard threshold value, and sending out a warning signal if the absolute value of the height difference exceeds the standard threshold value; otherwise, triggering and adjusting an emergency instruction;
The early warning adjustment module executes an emergency adjustment instruction, when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be positive according to the built rule engine, an elongation signal is sent to the gradient adjustment mechanism, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor, and an elongation pre-estimation L Long length of the gradient adjustment mechanism is generated;
when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be negative, a shortening signal is sent to the gradient adjusting mechanism, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor and the collected initial length of the gradient adjusting mechanism, and a shortening pre-estimation L Short length of the gradient adjusting mechanism is generated.
In the application, the related formulas are all the numerical calculation after dimensionality removal, and the formulas are one formulas for obtaining the latest real situation by software simulation through collecting a large amount of data, and the formulas are set by a person skilled in the art according to the actual situation.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application.
Claims (9)
1. The elevator non-flat-bed door fault identification method based on computer vision is characterized by comprising the following steps of: s1, assembling a gradient adjusting mechanism at the bottom of an elevator, and arranging an infrared detection unit at a position where the gradient adjusting mechanism is level with a corresponding floor, judging whether an infrared signal in the infrared detection unit is received after the elevator reaches the corresponding floor, and if the infrared signal can be received, not responding; otherwise, sending out an early warning signal;
s2, triggering the binocular shooting unit under the condition of receiving the early warning signal, and identifying fault information by adopting a stereoscopic vision technology to obtain the height difference generated between the elevator and the corresponding floor;
s3, comparing the obtained height difference between the elevator and the corresponding floor with a preset standard threshold value, and if the absolute value of the height difference exceeds the standard threshold value, sending out a warning signal; otherwise, triggering and adjusting an emergency instruction;
S4, executing an emergency adjustment instruction, when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be positive according to the built rule engine, sending an elongation signal to the gradient adjustment mechanism, building a data calculation model according to the height difference generated between the elevator and the corresponding floor, and generating an elongation pre-estimation L Long length of the gradient adjustment mechanism;
when the height difference generated between the elevator and the corresponding floor obtained in real time is judged to be negative, a shortening signal is sent to the gradient adjusting mechanism, a data calculation model is built according to the height difference generated between the elevator and the corresponding floor and the collected initial length of the gradient adjusting mechanism, and a shortening pre-estimation L Short length of the gradient adjusting mechanism is generated.
2. The elevator non-flat bed door failure recognition method based on computer vision according to claim 1, wherein: the gradient adjusting mechanism comprises a longitudinal cylinder (101) and a transition plate (102); the vertical cylinder (101) is assembled in the elevator floor, the vertical cylinder (101) is always located below the transition plate (102), the output end of the vertical cylinder (101) is hinged to the lower surface of the transition plate (102), one end of the transition plate (102) is connected with two sides of the inner wall of the elevator floor through a rotating shaft, and the other end of the transition plate (102) faces the floor.
3. The elevator non-flat bed door failure recognition method based on computer vision according to claim 2, wherein: the infrared detection unit comprises an infrared emitter and an infrared receiver, the infrared emitter is configured on the surface of the corresponding floor, the infrared receiver is embedded in the end face of the transition plate (102) facing the floor, each time the elevator arrives at the corresponding floor, the infrared generator emits an infrared signal, if the infrared receiver receives the infrared signal, the infrared receiver indicates that the elevator is level with the floor, and if the infrared receiver does not receive the infrared signal, the infrared receiver indicates that the elevator does not have the fault of opening the door in a flat floor.
4. A computer vision based elevator non-flat bed door failure recognition method according to claim 3, characterized in that: the binocular camera unit that triggers is installed in the binocular camera on elevator upper portion, and the shooting face of binocular camera covers the level ground of elevator gate and floor, acquires the image of elevator gate and floor level ground, and the elevator gate represents left image, and the stair level ground represents right image.
5. The computer vision-based elevator non-flat bed door failure recognition method according to claim 4, wherein: the process of obtaining the height difference generated between the elevator and the corresponding floor is as follows:
Stereoscopic vision matching: the method comprises the steps of performing stereo matching on images acquired by a binocular camera unit, namely matching corresponding pixel points in left and right images to obtain parallax information between the left and right images, wherein the parallax information represents position differences of objects in the left and right images, and depth information of the objects corresponding to the left and right images from a camera is obtained through parallax calculation;
and (3) calculating a height difference: calculating a height difference by using depth information obtained by stereoscopic vision; the height difference is obtained by calculating the difference between the depth values of the elevator doorway and the floor level ground according to the depth information in the left image and the right image, and the formula is as follows:
hr＝S Ladder -S Layer(s)
wherein hr represents the height difference, S Ladder represents the distance between the binocular camera and the floor of the elevator, S Layer(s) represents the distance between the binocular camera and the floor of the corresponding elevator, if hr is positive, the elevator door is lower than the depth of the floor of the corresponding elevator, and if hr is negative, the elevator door is higher than the depth of the floor of the corresponding elevator.
6. The computer vision-based elevator non-flat bed door failure recognition method of claim 5, wherein: the stereo matching algorithm includes any one of disparity mapping and block matching.
7. The computer vision-based elevator non-flat bed door failure recognition method of claim 6, wherein: the preset standard threshold value represents the maximum value of the height difference which can be handled by the gradient adjusting mechanism, and the warning signal sent in the S3 is different from the early warning signal sent in the S1 in expression mode.
8. The computer vision-based elevator non-flat bed door failure recognition method of claim 7, wherein: the equation from which the elongation pre-estimation L Long length of the gradient adjustment mechanism is generated is as follows:
Wherein G is a constant correction coefficient, the value range of G is 0-1, and L 0 represents the initial length of the gradient adjusting mechanism.
9. The computer vision-based elevator non-flat bed door failure recognition method of claim 8, wherein: the equation from which the elongation pre-estimation L Short length of the gradient adjustment mechanism is generated is as follows:
Wherein G is a constant correction coefficient, the value range of G is 0-1, and L 0 represents the initial length of the gradient adjusting mechanism.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410477476.8A CN118255226A (en) | 2024-04-19 | 2024-04-19 | Computer vision-based elevator non-flat-bed door fault identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410477476.8A CN118255226A (en) | 2024-04-19 | 2024-04-19 | Computer vision-based elevator non-flat-bed door fault identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118255226A true CN118255226A (en) | 2024-06-28 |
Family
ID=91603707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410477476.8A Pending CN118255226A (en) | 2024-04-19 | 2024-04-19 | Computer vision-based elevator non-flat-bed door fault identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118255226A (en) |
-
2024
- 2024-04-19 CN CN202410477476.8A patent/CN118255226A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102065975B1 (en) | Safety Management System Using a Lidar for Heavy Machine | |
| EP2300949B1 (en) | Video-based system and method of elevator door detection | |
| CN1315715C (en) | Camera for monitoring escalator and mobile footway | |
| CN112623919B (en) | Escalator intelligent monitoring management system based on computer vision | |
| US20200031627A1 (en) | Elevator component inspection systems | |
| KR101852057B1 (en) | unexpected accident detecting system using images and thermo-graphic image | |
| KR20190046351A (en) | Method and Apparatus for Detecting Intruder | |
| Rabiee et al. | IVOA: Introspective vision for obstacle avoidance | |
| CN108341313A (en) | A kind of method for early warning and its multimedia Forewarning Terminal of elevator safety operation environmental monitoring | |
| US20230050216A1 (en) | Aircraft door camera system for engine inlet monitoring | |
| CN109711257A (en) | A kind of banister condition detection method and system based on image recognition | |
| KR20220023726A (en) | Deep learning based realtime process monitoring system and method | |
| US20220083782A1 (en) | Item monitoring for doorbell cameras | |
| CN111523386B (en) | High-speed railway platform door monitoring and protecting method and system based on machine vision | |
| CN109345787A (en) | A kind of anti-outer damage monitoring and alarming system of the transmission line of electricity based on intelligent image identification technology | |
| CN112643719A (en) | Tunnel security detection method and system based on inspection robot | |
| CN115600124A (en) | Subway tunnel inspection system and inspection method | |
| KR101989376B1 (en) | Integrated track circuit total monitoring system | |
| CN110597165B (en) | Steel piling monitoring system and steel piling monitoring method | |
| KR20180081645A (en) | Embedded vision system for real-time detection of safety anomaly in subway screen door and method thereof | |
| KR101542134B1 (en) | The apparatus and method of surveillance a rock fall based on smart video analytic | |
| CN118255226A (en) | Computer vision-based elevator non-flat-bed door fault identification method | |
| CN116630259A (en) | Method and device for detecting looseness of nut in signal box | |
| CN112801072B (en) | Elevator non-flat-layer door opening fault recognition device and method based on computer vision | |
| CN111646349B (en) | Elevator protection method and device based on TOF image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |