CN116137074A - Automatic detection method and system for passengers in elevator car - Google Patents
Automatic detection method and system for passengers in elevator car Download PDFInfo
- Publication number
- CN116137074A CN116137074A CN202310152622.5A CN202310152622A CN116137074A CN 116137074 A CN116137074 A CN 116137074A CN 202310152622 A CN202310152622 A CN 202310152622A CN 116137074 A CN116137074 A CN 116137074A
- Authority
- CN
- China
- Prior art keywords
- passenger
- passengers
- elevator car
- fight
- preset time
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 63
- 238000012544 monitoring process Methods 0.000 claims abstract description 92
- 210000000707 wrist Anatomy 0.000 claims abstract description 61
- 230000006399 behavior Effects 0.000 claims abstract description 46
- 230000009471 action Effects 0.000 claims abstract description 24
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000000034 method Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 5
- 210000003414 extremity Anatomy 0.000 description 4
- 239000004973 liquid crystal related substance Substances 0.000 description 4
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B50/00—Energy efficient technologies in elevators, escalators and moving walkways, e.g. energy saving or recuperation technologies
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Indicating And Signalling Devices For Elevators (AREA)
Abstract
The invention discloses an automatic detection method and system for the fight behavior of passengers in an elevator car, wherein the detection method comprises the following steps: acquiring monitoring videos in an elevator car, and acquiring N monitoring images corresponding to N frames of monitoring videos in preset time; acquiring the number of passengers in the elevator car according to the N monitoring images, and judging whether the number of passengers is more than 1; if the number of the passengers is greater than 1, acquiring the skeleton information of each passenger in the N monitoring images according to the N monitoring images through a Lightweight OpenPose algorithm, acquiring the coordinate information of each joint point of each passenger according to the skeleton information of each passenger in the N monitoring images, and extracting the wrist coordinate information of each passenger; calculating the movement speed of the wrist of the corresponding passenger in the preset time according to the wrist coordinate information; judging whether the corresponding passengers take the fight action according to the movement speed of the wrists of the passengers within the preset time. Therefore, the real-time accurate detection of the fight behaviors of passengers in the elevator car can be automatically carried out.
Description
Technical Field
The invention belongs to the technical field of behavior detection, and relates to an automatic detection method and an automatic detection system for a fight behavior of passengers in an elevator car.
Background
The elevator is widely applied to daily life of people, and a fight event in the elevator car frequently happens, and once the fight behavior occurs in the elevator car, the elevator is difficult to rescue in time due to the special environment in the elevator car. Therefore, it is important to find out the fight behavior inside the elevator car in time, otherwise irrecoverable consequences will be caused.
In the related art, whether the passenger fights the car or not is judged through the monitoring video shot by the monitoring camera installed in the elevator car by people, however, the manual monitoring mode consumes manpower resources, and the situation that real-time accurate monitoring cannot be realized due to fatigue or distraction and other factors easily occurs, so that the reliability is low.
Disclosure of Invention
The invention aims to provide an automatic detection method for the fight behavior of passengers in an elevator car.
Another object of the invention is to provide an automatic detection system for the behavior of passengers fighting in an elevator car.
The technical solution for realizing the purpose of the invention is as follows:
an automatic detection method for the fight behavior of passengers in an elevator car comprises the following steps: acquiring a monitoring video in the elevator car, and acquiring N monitoring images corresponding to N frames of the monitoring video in preset time according to the monitoring video; acquiring the number of passengers in the elevator car according to N monitoring images through a target detection model, and judging whether the number of passengers is larger than 1; if the number of the passengers is greater than 1, acquiring skeleton information of each passenger in N monitoring images according to N monitoring images through a Lightweight OpenPose algorithm, acquiring coordinate information of each joint point of each passenger according to the skeleton information of each passenger in N monitoring images, and extracting wrist coordinate information of each passenger; calculating the movement speed of the wrist of the corresponding passenger in the preset time according to the wrist coordinate information; judging whether the corresponding passengers take the fight action according to the movement speed of the wrists of the passengers in the preset time.
In one embodiment of the present invention, determining whether the corresponding passenger performs a fight action according to the movement speed of the wrist of each passenger in the preset time includes: detecting whether the moving speed drifts within the preset time by adopting a drift detection algorithm; and if the movement speed is detected to drift within the preset time, judging that the corresponding passenger takes a fight action.
In one embodiment of the present invention, after detecting that the movement speed is shifted within the preset time, the method further includes: extracting the coordinate information of the local articulation point of each passenger from the coordinate information of each articulation point of each passenger; calculating the total kinetic energy of the local articulation point according to the coordinate information of the local articulation point and the number of passengers; judging whether the total kinetic energy is larger than a preset kinetic energy threshold value or not; and if the total kinetic energy is larger than the preset kinetic energy threshold value, judging that the corresponding passenger takes a fight action.
An automatic detection system for a passenger's fight behavior in an elevator car, comprising: the first acquisition module is used for acquiring the monitoring video in the elevator car and acquiring N monitoring images corresponding to N frames of the monitoring video in preset time according to the monitoring video; the first judging module is used for acquiring the number of passengers in the elevator car according to N monitoring images through a target detection model and judging whether the number of passengers is larger than 1; the second acquisition module is used for acquiring the skeleton information of each passenger in the N monitoring images according to the N monitoring images through a Lightweight OpenPose algorithm when the number of passengers is greater than 1, acquiring the coordinate information of each joint point of each passenger according to the skeleton information of each passenger in the N monitoring images, and extracting the wrist coordinate information of each passenger; the calculation module is used for calculating the movement speed of the wrist of the corresponding passenger in the preset time according to the wrist coordinate information; the second judging module is used for judging whether the corresponding passengers take the fight action or not according to the movement speed of the wrists of the passengers within the preset time.
In one embodiment of the present invention, the second determining module is specifically configured to: detecting whether the moving speed drifts within the preset time by adopting a drift detection algorithm; and if the movement speed is detected to drift within the preset time, judging that the corresponding passenger takes a fight action.
In one embodiment of the present invention, the second determining module is further configured to, after detecting that the movement speed has shifted within the preset time,: extracting the coordinate information of the local articulation point of each passenger from the coordinate information of each articulation point of each passenger; calculating the total kinetic energy of the local articulation point according to the coordinate information of the local articulation point and the number of passengers; judging whether the total kinetic energy is larger than a preset kinetic energy threshold value or not; and if the total kinetic energy is larger than the preset kinetic energy threshold value, judging that the corresponding passenger takes a fight action.
A computer device comprising a memory, a processor and a computer program stored on the memory and operable on the processor, said processor implementing the above-described method for automatically detecting the behavior of a passenger fight in an elevator car when executing said computer program.
A non-transitory computer readable storage medium having stored thereon a computer program which when executed by a processor implements the above-described method of automatically detecting a passenger fight behavior in an elevator car.
Compared with the prior art, the invention has the remarkable advantages that:
the invention can realize the automatic detection of the fight behavior of passengers in the elevator car without manual participation, thereby saving manpower resources, and being capable of accurately detecting in real time and having higher reliability.
Drawings
Fig. 1 is a flow chart of a method for automatically detecting a passenger fight behavior in an elevator car according to an embodiment of the invention.
Fig. 2 is a block schematic diagram of an automatic detection system for passenger fight behavior in an elevator car according to an embodiment of the 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.
Fig. 1 is a flow chart of a method for automatically detecting a passenger fight behavior in an elevator car according to an embodiment of the invention.
As shown in fig. 1, the automatic detection method for the fight behavior of passengers in an elevator car according to the embodiment of the invention can comprise the following steps:
s1, acquiring monitoring videos in an elevator car, and acquiring N monitoring images corresponding to N frames of monitoring videos in preset time according to the monitoring videos.
Specifically, a surveillance video in an elevator car may be captured by a surveillance camera mounted in the elevator car. It will be appreciated that, in general, the motion of passengers in an elevator car is not large, so that one monitoring image can be extracted from the monitoring video every N frames (for example, every 10 frames) within a preset time, and N monitoring images are extracted altogether, so as to prevent the occurrence of repeated data.
S2, acquiring the number of passengers in the elevator car according to the N monitoring images through the target detection model, and judging whether the number of passengers is larger than 1.
Specifically, before the number of passengers in the elevator car is obtained according to N monitoring images through the target detection model, K historical monitoring images can be extracted from the historical detection video, the passenger image data sets in the K historical monitoring images are marked sequentially by LabelImg image marking software so as to mark the heads of all passengers in each historical monitoring image, and the K marked historical monitoring images are divided into a training set and a testing set according to a preset proportion so as to train a target detection algorithm and obtain the target detection model. Specifically, the training set may be trained by using the csp-dark 53 network model of YOLOv4, and network parameters during training may be modified according to the performance of the GPU (Graphics Processing Unit, image processing unit), for example, the batch size=64, the division=128, the learning rate=0.001, and the training number 10000 times may be set. And setting a weight file storage interval, and reserving when the fitting condition occurs after training is finished. And after training, the training curve converges to obtain a weight file model. And testing the video of the passengers in the elevator car by using the weight file model obtained by training 10000 times so as to finally obtain a target detection model.
N monitoring images are input into a trained target detection model to output the number of passengers in the elevator car. Then, it is judged whether the number of passengers is greater than 1.
It will be appreciated that if the number of passengers in the elevator car is 0 or 1, no fight action will occur in the elevator car, so that no further subsequent detection is necessary; if the number of passengers in the elevator car is greater than 1, a fight behavior may occur in the elevator car, and therefore, detection of a subsequent fight behavior of passengers in the elevator car is required.
And S3, if the number of the passengers is greater than 1, acquiring the skeleton information of each passenger in the N monitoring images according to the N monitoring images through a Lightweight OpenPose algorithm, acquiring the coordinate information of each joint point of each passenger according to the skeleton information of each passenger in the N monitoring images, and extracting the wrist coordinate information of each passenger.
Specifically, first, each joint point of each passenger in N monitoring images is detected through Lightweight OpenPose algorithm, coordinate information of each joint point and PAF of each limb are obtained, limb connection is obtained according to the coordinate information of each joint point and the PAF of each limb, and the limbs are spliced to form a human skeleton, so that skeleton information of each passenger in N monitoring images is obtained.
And then, acquiring coordinate information of each joint point of each passenger according to the skeleton information of each passenger in the N monitoring images, and extracting wrist coordinate information of each passenger.
And S4, calculating the movement speed of the wrist of the corresponding passenger in the preset time according to the wrist coordinate information.
Specifically, the wrist of the passenger can be selected as a key point for matching feature points between different frames, and the wrist coordinate information of the jth passenger detected by the ith frame of monitoring video (i.e. the ith monitoring image) is recorded as
After the preset frame is passed, the coordinates of the corresponding wrist are recorded again, and the Euclidean distance d between the coordinates and the corresponding wrist is calculated ij Wherein the preset frame may be 3 frames, and continuously calculating Euclidean distance d of the corresponding wrist at intervals of 3 frames from the first frame to the last frame ij ：
Wherein, the liquid crystal display device comprises a liquid crystal display device,
d ij for the Euclidean distance of wrist movement every 3 frames starting from frame 4, +.>
For the ith frame of surveillance video, the coordinates of the wrist of the jth passenger in the y-direction,/->
Representing the ith-3 frames, the jth passenger's wrist is in the y-directionCoordinates of->
Representing the coordinates of the wrist of the jth passenger in the x-direction,/-in the ith frame of surveillance video>
And (3) representing the coordinates of the wrist of the jth passenger in the x direction when the ith-3 frames of monitoring video are displayed. Δx ij Representing the distance by which the wrist of the jth passenger moves in the x-direction, Δy, during the ith frame of surveillance video ij And when the ith frame of monitoring video is represented, the wrist of the jth passenger moves in the y direction.
Continuously calculating the movement speed of the corresponding wrist from the 4 th frame of the video
I.e.
Wherein, the liquid crystal display device comprises a liquid crystal display device,
indicating the movement speed of the wrist of the jth passenger in the ith frame from the 4 th frame monitoring video.
S5, judging whether the corresponding passengers take the fight action according to the movement speed of the wrists of the passengers in the preset time.
In one embodiment of the present invention, determining whether a fight behavior occurs to each passenger according to a movement speed of the wrist of the passenger in a preset time includes: detecting whether the moving speed drifts in a preset time by adopting a drift detection algorithm; if the detected moving speed drifts within the preset time, judging that the corresponding passenger takes a fight action.
Specifically, after the movement speed of the wrist of each passenger in the preset time is obtained, the movement speed can be used as the input of a signal, and a drift detection algorithm, for example, a Page-henkey test algorithm is adopted to detect the drift condition of the wrist speed of the corresponding passenger, that is, whether the abrupt change condition occurs.
Specifically, the Page-Hinkey test algorithm can detect whether the movement speed drifts within a preset time by the following formula:
v ω,t ＝min(m v,t-1 ,M v,t )，
wherein M is v,0 ＝0，v t The moving speed at the time t is the moving speed at the time t,
v is t The average value is updated every frame. j (j) v The constant for controlling the detection sensitivity is set by an experiment manually. M is M v,t For a cumulative value, representing the degree of deviation of the signal value accumulation from the expected value, m v,t Is the minimum of this cumulative value. />
Is M v,t And m v,t Is a difference in (c). As can be seen from the above formula, when v t When the change is slow, the person is added with->
Also change slowly, at the same time M v,t Will also change slowly. When v t When there is a sudden and severe change, there is also a corresponding result +.>
And M v,t Abrupt changes.
Thus, the present invention first uses a target detection model to detect the heads of passengers in an elevator car, thereby determining the number of passengers in the elevator car. Meanwhile, the skeleton and joint point information of the passengers are extracted by using a Lightweight OpenPose human skeleton extraction algorithm. And then, calculating the motion condition of the local joint point through the extracted coordinate information of the skeleton feature points, selecting the wrist of the passenger as the feature point, and detecting whether the wrist speed in each frame has mutation or not by using a Page-Hinkley test algorithm. And determining a threshold value through an experimental method, comparing the detected motion condition with a set threshold value, and judging whether the fight behavior occurs or not. Therefore, automatic detection of the fight behavior of passengers in the elevator car can be realized without manual participation, not only is manpower resources saved, but also the passengers can be accurately detected in real time, and the reliability is higher.
It should be noted that, in practical application, when passengers make phone calls, handshakes and other actions in the elevator car, the wrist speed will also generate mutation, thus causing false detection, so the total kinetic energy of the local articulation point can be further selected as a judging condition to assist in judging the fight behavior.
In one embodiment of the present invention, after detecting that the movement speed is shifted within the preset time, the method further includes: extracting the coordinate information of the local joint point of each passenger from the coordinate information of each joint point of each passenger; calculating the total kinetic energy of the local articulation point according to the coordinate information of the local articulation point and the number of passengers; judging whether the total kinetic energy is larger than a preset kinetic energy threshold value or not; if the total kinetic energy is larger than the preset kinetic energy threshold value, judging that the corresponding passenger takes a fight action.
Specifically, according to the motion change of the human body during the fight process, local articulation points of each passenger, such as left shoulder, right shoulder, left hip, right hip, neck, left wrist and right wrist, can be selected. The coordinate information of the local node of each passenger is extracted from the coordinate information of each node of each passenger, and the moving speed of each local node is calculated according to the coordinate information of the local node. And then calculating the total kinetic energy of the local articulation points according to the moving speed of each local articulation point by the following formula:
After the total kinetic energy is calculated, the total kinetic energy can be compared with a preset kinetic energy threshold value, and if the total kinetic energy is larger than the preset kinetic energy threshold value, the corresponding passenger is judged to take the fight action. Otherwise, judging that the corresponding passengers do not take the fight action.
In summary, according to the automatic detection method for the fight behavior of the passengers in the elevator car according to the embodiment of the invention, the monitoring video in the elevator car is obtained, the N monitoring images corresponding to the N monitoring videos in the preset time are obtained according to the monitoring video, the number of the passengers in the elevator car is obtained according to the N monitoring images through the target detection model, whether the number of the passengers is greater than 1 is judged, when the number of the passengers is greater than 1, the skeleton information of each passenger in the N monitoring images is obtained according to the N monitoring images through the Lightweight OpenPose algorithm, the coordinate information of each joint point of each passenger is obtained according to the skeleton information of each passenger in the N monitoring images, the wrist coordinate information of each passenger is extracted, the movement speed of the wrist of the corresponding passenger in the preset time is calculated according to the wrist coordinate information, and whether the fight behavior of the corresponding passenger occurs is judged according to the movement speed of the wrist of the passenger in the preset time. Therefore, automatic detection of the fight behavior of passengers in the elevator car can be realized without manual participation, not only is manpower resources saved, but also the passengers can be accurately detected in real time, and the reliability is higher.
Corresponding to the automatic detection method of the passengers 'fight behavior in the elevator car in the embodiment, the invention provides an automatic detection system of the passengers' fight behavior in the elevator car.
As shown in fig. 2, the automatic detection system for the behavior of passengers fighting in an elevator car according to the embodiment of the invention may comprise: the first acquisition module 100, the first judgment module 200, the second acquisition module 300, the calculation module 400, and the second judgment module 500.
The first obtaining module 100 is configured to obtain a monitoring video in an elevator car, and obtain N monitoring images corresponding to N monitoring videos in a preset time according to the monitoring video; the first judging module 200 is used for acquiring the number of passengers in the elevator car according to the N monitoring images through the target detection model and judging whether the number of passengers is larger than 1; the second obtaining module 300 is configured to obtain, when the number of passengers is greater than 1, skeleton information of each passenger in the N monitoring images according to the N monitoring images by using a Lightweight OpenPose algorithm, and obtain coordinate information of each node of each passenger according to the skeleton information of each passenger in the N monitoring images, and extract wrist coordinate information of each passenger; the calculation module 400 is configured to calculate a movement speed of the wrist of the corresponding passenger in a preset time according to the wrist coordinate information; the second judging module 500 is configured to judge whether the corresponding passenger takes a fight action according to the movement speed of the wrist of each passenger in the preset time.
In one embodiment of the present invention, the second determining module 500 is specifically configured to: detecting whether the moving speed drifts in a preset time by adopting a drift detection algorithm; if the detected moving speed drifts within the preset time, judging that the corresponding passenger takes a fight action.
In one embodiment of the present invention, the second determining module 500 is further configured to, after detecting that the movement speed has shifted within the preset time: extracting the coordinate information of the local joint point of each passenger from the coordinate information of each joint point of each passenger; calculating the total kinetic energy of the local articulation point according to the coordinate information of the local articulation point and the number of passengers; judging whether the total kinetic energy is larger than a preset kinetic energy threshold value or not; if the total kinetic energy is larger than the preset kinetic energy threshold value, judging that the corresponding passenger takes a fight action.
It should be noted that, for more specific embodiments of the automatic detection system for a passenger fight behavior in an elevator car according to the present invention, reference may be made to the above-described embodiments of the automatic detection method for a passenger fight behavior in an elevator car, and details thereof are not given here for avoiding redundancy.
According to the automatic detection system for the fight behavior of the passengers in the elevator car, the first acquisition module acquires the monitoring video in the elevator car, N monitoring images corresponding to N frames of monitoring video in preset time are acquired according to the monitoring video, the first judgment module acquires the number of the passengers in the elevator car according to the N monitoring images through the target detection model, judges whether the number of the passengers is larger than 1, the second acquisition module acquires the skeleton information of each passenger in the N monitoring images according to the N monitoring images through a Lightweight OpenPose algorithm when the number of the passengers is larger than 1, the coordinate information of each joint point of each passenger is acquired according to the skeleton information of each passenger in the N monitoring images, wrist coordinate information of each passenger is extracted, the calculation module calculates the movement speed of the wrist of the corresponding passenger in the preset time according to the wrist coordinate information, and the second judgment module judges whether the fight behavior of the corresponding passenger occurs according to the movement speed of the wrist of the passengers in the preset time. Therefore, automatic detection of the fight behavior of passengers in the elevator car can be realized without manual participation, not only is manpower resources saved, but also the passengers can be accurately detected in real time, and the reliability is higher.
The invention also provides computer equipment corresponding to the embodiment.
The computer device of the embodiment of the invention can comprise a memory, a processor and a computer program stored on the memory and running on the processor, wherein the automatic detection method of the passengers' fight behavior in the elevator car is realized when the processor executes the computer program.
According to the computer equipment provided by the embodiment of the invention, the automatic detection of the fight behavior of passengers in the elevator car can be realized without manual participation, so that not only is the manpower resource saved, but also the detection can be accurately performed in real time, and the reliability is higher.
The present invention also proposes a non-transitory computer-readable storage medium corresponding to the above-described embodiments.
The non-transitory computer readable storage medium of the embodiment of the invention stores a computer program which, when executed by a processor, implements the above-described automatic detection method of the behavior of passengers fighting in an elevator car.
According to the non-transitory computer readable storage medium, the automatic detection of the passenger fight behavior in the elevator car can be realized without manual participation, so that not only is the manpower resource saved, but also the detection can be accurately performed in real time, and the reliability is higher.
The foregoing examples are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the foregoing examples, and any other changes, modifications, substitutions, combinations, and simplifications that do not depart from the spirit and principles of the present invention should be made therein and are intended to be equivalent substitutes within the scope of the present invention.
Claims (8)
1. An automatic detection method for the fight behavior of passengers in an elevator car is characterized by comprising the following steps:
acquiring a monitoring video in the elevator car, and acquiring N monitoring images corresponding to N frames of the monitoring video in preset time according to the monitoring video;
acquiring the number of passengers in the elevator car according to N monitoring images through a target detection model, and judging whether the number of passengers is larger than 1;
if the number of the passengers is greater than 1, acquiring skeleton information of each passenger in N monitoring images according to N monitoring images through a Lightweight OpenPose algorithm, acquiring coordinate information of each joint point of each passenger according to the skeleton information of each passenger in N monitoring images, and extracting wrist coordinate information of each passenger;
calculating the movement speed of the wrist of the corresponding passenger in the preset time according to the wrist coordinate information;
judging whether the corresponding passengers take the fight action according to the movement speed of the wrists of the passengers in the preset time.
2. The method for detecting the fight behavior of passengers in an elevator car according to claim 1, wherein determining whether the corresponding passenger has a fight behavior based on the movement speed of the wrist of each passenger within the preset time comprises:
detecting whether the moving speed drifts within the preset time by adopting a drift detection algorithm;
and if the movement speed is detected to drift within the preset time, judging that the corresponding passenger takes a fight action.
3. The method for detecting a fight behavior of a passenger in an elevator car according to claim 2, further comprising, after detecting that the moving speed has shifted within the preset time:
extracting the coordinate information of the local articulation point of each passenger from the coordinate information of each articulation point of each passenger;
calculating the total kinetic energy of the local articulation point according to the coordinate information of the local articulation point and the number of passengers;
judging whether the total kinetic energy is larger than a preset kinetic energy threshold value or not; and if the total kinetic energy is larger than the preset kinetic energy threshold value, judging that the corresponding passenger takes a fight action.
4. An automatic detection system for a passenger fight behavior in an elevator car, comprising:
the first acquisition module is used for acquiring the monitoring video in the elevator car and acquiring N monitoring images corresponding to N frames of the monitoring video in preset time according to the monitoring video;
the first judging module is used for acquiring the number of passengers in the elevator car according to N monitoring images through a target detection model and judging whether the number of passengers is larger than 1;
the second acquisition module is used for acquiring the skeleton information of each passenger in the N monitoring images according to the N monitoring images through a Lightweight OpenPose algorithm when the number of passengers is greater than 1, acquiring the coordinate information of each joint point of each passenger according to the skeleton information of each passenger in the N monitoring images, and extracting the wrist coordinate information of each passenger;
the calculation module is used for calculating the movement speed of the wrist of the corresponding passenger in the preset time according to the wrist coordinate information;
the second judging module is used for judging whether the corresponding passengers take the fight action according to the movement speed of the wrists of the passengers in the preset time.
5. The automatic detection system of claim 4, wherein the second determination module is specifically configured to:
detecting whether the moving speed drifts within the preset time by adopting a drift detection algorithm;
and if the movement speed is detected to drift within the preset time, judging that the corresponding passenger takes a fight action.
6. The automatic detection system for a passenger fight behavior in an elevator car according to claim 5, wherein the second determination module, after detecting that the movement speed has shifted within the preset time, is further configured to:
extracting the coordinate information of the local articulation point of each passenger from the coordinate information of each articulation point of each passenger;
calculating the total kinetic energy of the local articulation point according to the coordinate information of the local articulation point and the number of passengers;
judging whether the total kinetic energy is larger than a preset kinetic energy threshold value or not; and if the total kinetic energy is larger than the preset kinetic energy threshold value, judging that the corresponding passenger takes a fight action.
7. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements a method for automatic detection of a passenger fight behavior in an elevator car according to any of claims 1-3.
8. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a method for automatic detection of a passenger fight behavior in an elevator car according to any of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310152622.5A CN116137074A (en) | 2023-02-22 | 2023-02-22 | Automatic detection method and system for passengers in elevator car |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310152622.5A CN116137074A (en) | 2023-02-22 | 2023-02-22 | Automatic detection method and system for passengers in elevator car |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116137074A true CN116137074A (en) | 2023-05-19 |
Family
ID=86333464
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310152622.5A Pending CN116137074A (en) | 2023-02-22 | 2023-02-22 | Automatic detection method and system for passengers in elevator car |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116137074A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107000981A (en) * | 2014-11-26 | 2017-08-01 | 奥的斯电梯公司 | The elevator safety and control system moved based on passenger |
| WO2019052318A1 (en) * | 2017-09-13 | 2019-03-21 | 杭州海康威视数字技术股份有限公司 | Method, apparatus and system for monitoring elevator car |
| CN110765964A (en) * | 2019-10-30 | 2020-02-07 | 常熟理工学院 | Method for detecting abnormal behaviors in elevator car based on computer vision |
| US20210174074A1 (en) * | 2019-09-27 | 2021-06-10 | Beijing Sensetime Technology Development Co., Ltd. | Human detection method and apparatus, computer device and storage medium |
| CN114241379A (en) * | 2021-12-16 | 2022-03-25 | 成都新潮传媒集团有限公司 | Passenger abnormal behavior identification method, device and equipment and passenger monitoring system |
| KR20220113631A (en) * | 2021-02-05 | 2022-08-16 | 호서대학교 산학협력단 | Dangerous situation detection device and dangerous situation detection method |
-
2023
- 2023-02-22 CN CN202310152622.5A patent/CN116137074A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107000981A (en) * | 2014-11-26 | 2017-08-01 | 奥的斯电梯公司 | The elevator safety and control system moved based on passenger |
| WO2019052318A1 (en) * | 2017-09-13 | 2019-03-21 | 杭州海康威视数字技术股份有限公司 | Method, apparatus and system for monitoring elevator car |
| US20210174074A1 (en) * | 2019-09-27 | 2021-06-10 | Beijing Sensetime Technology Development Co., Ltd. | Human detection method and apparatus, computer device and storage medium |
| CN110765964A (en) * | 2019-10-30 | 2020-02-07 | 常熟理工学院 | Method for detecting abnormal behaviors in elevator car based on computer vision |
| KR20220113631A (en) * | 2021-02-05 | 2022-08-16 | 호서대학교 산학협력단 | Dangerous situation detection device and dangerous situation detection method |
| CN114241379A (en) * | 2021-12-16 | 2022-03-25 | 成都新潮传媒集团有限公司 | Passenger abnormal behavior identification method, device and equipment and passenger monitoring system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210302B (en) | Multi-target tracking method, device, computer equipment and storage medium | |
| US11048942B2 (en) | Method and apparatus for detecting a garbage dumping action in real time on video surveillance system | |
| CN110688929B (en) | Human skeleton joint point positioning method and device | |
| US20220180534A1 (en) | Pedestrian tracking method, computing device, pedestrian tracking system and storage medium | |
| US20130070105A1 (en) | Tracking device, tracking method, and computer program product | |
| CN110889376A (en) | Safety helmet wearing detection system and method based on deep learning | |
| CN112016435A (en) | Abnormal behavior detection method and device, electronic equipment and storage medium | |
| CN113221771B (en) | Living body face recognition method, device, apparatus, storage medium and program product | |
| CN112597877A (en) | Factory personnel abnormal behavior detection method based on deep learning | |
| US20220366570A1 (en) | Object tracking device and object tracking method | |
| CN114005167A (en) | Remote sight estimation method and device based on human skeleton key points | |
| CN116434325A (en) | Method, device, equipment and storage medium for detecting specific action | |
| CN114155492A (en) | High-altitude operation safety belt hanging rope high-hanging low-hanging use identification method and device and electronic equipment | |
| CN110580708B (en) | Rapid movement detection method and device and electronic equipment | |
| CN116002480A (en) | Automatic detection method and system for accidental falling of passengers in elevator car | |
| CN116137074A (en) | Automatic detection method and system for passengers in elevator car | |
| CN116580454A (en) | Motion evaluation method and device based on target detection and human body posture estimation | |
| CN114518106B (en) | Method, system, medium and equipment for detecting update of vertical elements of high-precision map | |
| CN116229502A (en) | Image-based tumbling behavior identification method and equipment | |
| CN115546825A (en) | Automatic monitoring method for safety inspection normalization | |
| CN111695404B (en) | Pedestrian falling detection method and device, electronic equipment and storage medium | |
| CN114758414A (en) | Pedestrian behavior detection method, device, equipment and computer storage medium | |
| CN114663972A (en) | Target marking method and device based on motion difference | |
| JP2018124786A (en) | Image processing device, image processing method, and image processing program | |
| CN112580633A (en) | Public transport passenger flow statistical device and method |
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 |