CN115620228A - Subway shield door passenger door-rushing early warning method based on video analysis - Google Patents

Subway shield door passenger door-rushing early warning method based on video analysis Download PDF

Info

Publication number
CN115620228A
CN115620228A CN202211251732.9A CN202211251732A CN115620228A CN 115620228 A CN115620228 A CN 115620228A CN 202211251732 A CN202211251732 A CN 202211251732A CN 115620228 A CN115620228 A CN 115620228A
Authority
CN
China
Prior art keywords
door
target
passenger
rushing
early warning
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202211251732.9A
Other languages
Chinese (zh)
Other versions
CN115620228B (en
Inventor
王曦明
刘光杰
孙同庆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Information Science and Technology
Original Assignee
Nanjing University of Information Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Information Science and Technology filed Critical Nanjing University of Information Science and Technology
Priority to CN202211251732.9A priority Critical patent/CN115620228B/en
Publication of CN115620228A publication Critical patent/CN115620228A/en
Application granted granted Critical
Publication of CN115620228B publication Critical patent/CN115620228B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61K—AUXILIARY EQUIPMENT SPECIALLY ADAPTED FOR RAILWAYS, NOT OTHERWISE PROVIDED FOR
    • B61K13/00—Other auxiliaries or accessories for railways
    • B61K13/04—Passenger-warning devices attached to vehicles; Safety devices for preventing accidents to passengers when entering or leaving vehicles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/62—Extraction of image or video features relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
    • 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
    • G06V40/23—Recognition of whole body movements, e.g. for sport training
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Psychiatry (AREA)
  • Social Psychology (AREA)
  • Mechanical Engineering (AREA)
  • Train Traffic Observation, Control, And Security (AREA)

Abstract

The invention discloses a video analysis-based early warning method for door violation of passengers who close a subway shield door, which comprises the following steps: starting a passenger door-rushing early warning terminal system installed at a shielding door head at the moment of closing a door, and continuously acquiring video frames by a video acquisition module according to a given frame rate; the video analysis module continuously calculates the motion track of the pedestrian target in the front door area based on YOLOv5 and Deepsort; the door-rushing risk evaluation module calculates the total passenger door-rushing risk faced by the door based on the motion tracks of all the pedestrian targets; and performing secondary door-violation early warning treatment including acousto-optic alarm and door control linkage on the condition that the evaluation risk exceeds a given threshold value. The invention provides an effective scheme for reducing adverse effects of door-break behaviors of passengers on driving and passenger safety.

Description

Subway shield door passenger door-rushing early warning method based on video analysis
Technical Field
The invention relates to an urban rail transit intelligent station, in particular to a method for carrying out shielding door near-door passenger door-crossing early warning based on video analysis.
Background
Subways have become a main tool and a vigorously developed field of urban public transport due to the advantages of safety, punctuality, rapidness, comfort, environmental protection and the like. The intelligent subway can achieve the aims of improving the operation efficiency, reducing the operation risk, improving the passenger service satisfaction and the like by intelligently enabling various systems such as train vehicles, wire network stations, dispatching management, operation and maintenance support and the like. Development of intelligent subway technology and application based on technologies such as cloud computing, internet of things and artificial intelligence is vigorous. In recent years, systems such as face recognition non-inductive payment passing brake, voice recognition intelligent customer service, station operation based on intelligent video analysis and the like have been demonstrated and applied in domestic main cities and have achieved practical results.
In a smart station scene, intelligent video analysis is carried out through the CCTV connected with a signal system, so that abnormal scenes such as reverse running of an escalator, stair crowding, barrier passing, passenger falling, article omission and the like can be detected, and the initiative and timeliness of abnormal event discovery are improved. In the whole process of passenger taking a car, the behavior of the passenger running the door close to the door closing time is a high-risk behavior endangering the safety of the passenger and the driving safety. Usually, the shielding door and the train door are equipped with anti-holding functions, but the functions only have feedback on a rigid object with a large size, so that a large number of accidents of door rushing by passengers directly cause that the doors cannot be normally closed, the vehicles cannot normally run, and more serious accidents cause that passengers are clamped between the shielding door and the doors to cause death. In recent years, in some cities, a beneficial search for new equipment based on sensors such as visible light vision and laser radar is performed in the aspect of detecting foreign matters between the gaps of the screen door and the train door, but the problems of post emergency and linkage treatment of people and objects being clamped still can be solved, and active door-break early warning means are lacked all the time. CCTV monitoring configured on a station floor usually adopts a plurality of groups of cameras parallel to a train track, and the early warning under the condition that passengers break through doors and move quickly cannot be dealt with by unfavorable view field and station-level medium-high delay video analysis.
The invention provides a subway shield door passenger door-rushing early warning method based on video analysis based on the field computing capability of a camera with an oblique downward viewing angle and a passenger door-rushing early warning terminal system which are installed on each shield door head, so as to reduce the adverse influence of the door-rushing behavior of passengers on the safety of driving and passengers.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a subway shield door near-door passenger door-rushing early warning method based on video analysis, so that adverse effects of passenger door-rushing behaviors on driving and passenger safety are reduced through active door-rushing early warning.
The technical scheme is as follows: in order to solve the technical problems, the technical scheme adopted by the invention is as follows:
in a first aspect, a subway shield door passenger door-rushing early warning method based on video analysis is provided, and comprises the following steps:
step S1: continuously acquiring video frames I of a view field F according to a preset frame rate R when T seconds are left for closing the door 1 、I 2 、....I i 、....；
Step S2: for the current video frame I i Detecting a pedestrian target based on a YOLOv5 neural network and associating the pedestrian target with a detected target in a preamble video frame based on Deepsort to realize the target O 1 、O 2 、...、O n Wherein n is the number of all pedestrian targets, the motion track of the target in the visual field F is converted into the corresponding target continuous motion track Tr in the front door area H through perspective transformation 1 (t)、Tr 2 (t)、...Tr n (t)；
And step S3: using pedestrian objects O 1 、O 2 、...、O n Movement track Tr in the door front region H i (t) carrying out single-target door-opening risk assessment on the data, and calculating a continuous risk assessment value R of each target 1 (t)、R 2 (t)、...、R n (t), generating a total door-opening risk RH (t) of the facing passengers by integrating the single-target risks;
and step S4: and carrying out graded door-opening early warning treatment based on the door-opening risk RH (t) of the total passengers facing the door and a preset secondary threshold value.
In some embodiments, in step S1, the method for capturing video frames of the field of view F includes:
according to the time plan of a train running scheduling signal system, a passenger door-rushing early warning terminal system installed on a shielding door head starts a door-rushing early warning function for a passenger who is close to a door when the distance from the door-closing time is still T seconds, and a video acquisition module of the system acquires video frames by using a camera installed at an oblique downward visual angle right above the shielding door.
In some embodiments, the step S2 comprises:
s201: for the current video frame I collected in the step S1 i Detecting pedestrian target detection based on a YOLOv5 neural network, and detecting and outputting O 1 、O 2 、...、O n If no pedestrian target is detected, outputting NULL;
s202: if the output of S201 is NULL, the step S201 is executed circularly for the next frame I output by S1 i+1 Carrying out pedestrian target detection operation until the current time T = T;
if the pedestrian target is detected in step S201, according to the previous frame I i-1 Performing the correlative tracking operation of the Deepsort target on the recorded target detection result to form O 1 、O 2 、...、O n Motion track K of object in field of view F 1 (t)、K 2 (t)、...K n (t)；
S203: through a perspective transformation matrix calibrated in advance, K is converted 1 (t)、K 2 (t)、...K n (t) transformation into a trajectory Tr in plan view coordinates corresponding to the door front region H 1 (t)、Tr 2 (t)、...Tr n (t)。
In some embodiments, said step S3 comprises:
s301: generates motion trail Tr for all detected i Target O of (t) i Continuously estimating the angle u between the moving direction and the Y axis i The coordinate system of the door front region H is defined as a center line in a downward directionIs Y axis, and takes the right direction of the upper edge line of the H area as X axis;
s302: generating motion trail Tr for all detected i Target O of (t) i Continuously estimating the Y-axis component v of the motion velocity thereof i The coordinate system of the door front area H is defined to take the downward direction of the center line as the Y axis and the rightward direction of the upper edge line of the area H as the X axis;
s303: generating motion trail Tr for all detected i Target O of (t) i Recording object O i A distance D from the bottom line of the H region in the Y direction i ；
According to R i (t)＝max(cos(u i ),0)×max(1.5-v i ×(T-t)/D i 0) calculating the target O at the current time t i Risk of door violation;
s304: calculating the total passenger door-rushing risk of the door as follows: RH (t) = max (R) 1 (t),R 2 (t),...,R n (t))。
In some embodiments, the step S4 comprises:
the risk of total passenger breakthrough RH (t) exceeding a first threshold Th 1 The first-level response of the system is that passengers are reminded through an acousto-optic warning device arranged above the shielding door;
the risk RH (t) of door break-through of the total passengers exceeds a second threshold Th 2 The shielding door and the vehicle door control system are linked to keep the shielding door and the vehicle door open while the acousto-optic alarm is kept, and the shielding door and the vehicle door are closed by the shielding door and the vehicle door control system until the total passenger door-rushing risk RH (t) of the door returns to zero, wherein a second threshold Th 2 Greater than a first threshold Th 1 。
The step S4 further includes:
a video recording of a passenger breaching the door is recorded and generated and saved while a primary or secondary response is made.
In some embodiments, the first threshold Th 1 0.25, second threshold Th 2 Is 0.5.
In a second aspect, the invention provides a video analysis-based subway platform screen door passenger door-rushing early warning device, which comprises a processor and a storage medium, wherein the processor is used for processing the video analysis;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to the first aspect.
In a third aspect, the invention provides a storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of the first aspect.
Has the beneficial effects that: compared with the prior art, the invention has the following advantages:
1. the invention solves the problem that the prior station and train lack a passenger door-through early warning mechanism, and provides an effective scheme for reducing the door-through occurrence rate of passengers and the risk of clamping people and objects under extreme conditions;
2. compared with passenger abnormal behavior detection based on existing video monitoring of a station hall, the passenger door-rushing early warning system based on the intelligent computing unit installed on the shielding door head can provide passenger door-rushing early warning for each single door based on favorable view field and terminal real-time computing capability of the intelligent computing unit, and has higher real-time performance and reliability.
Drawings
Fig. 1 is a flowchart of a video analysis-based early warning method for door violation of a passenger who is closing a subway platform screen door according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the moving direction and speed of a pedestrian object in the front door area in the embodiment.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.
In the description of the present invention, the meaning of a plurality is one or more, the meaning of a plurality is two or more, and the above, below, exceeding, etc. are understood as excluding the present numbers, and the above, below, within, etc. are understood as including the present numbers. If the first and second are described for the purpose of distinguishing technical features, they are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
In the description of the present invention, reference to the description of the terms "one embodiment," "some embodiments," "an illustrative embodiment," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Example 1
A subway shield door passenger door-rushing early warning method based on video analysis comprises the following steps:
step S1: continuously acquiring video frames I of a view field F according to a preset frame rate R when T seconds are left before the door closing time 1 、I 2 、....I i 、....；
Step S2: for the current video frame I i Detecting a pedestrian target based on a YOLOv5 neural network and associating the pedestrian target with a detected target in a preamble video frame based on Deepsort to realize the target O 1 、O 2 、...、O n Wherein n is the number of all pedestrian targets, the motion track of the target in the visual field F is converted into the corresponding target continuous motion track Tr in the front door area H through perspective transformation 1 (t)、Tr 2 (t)、...Tr n (t)；
And step S3: using pedestrian objects O 1 、O 2 、...、O n Movement track Tr in the front door region H i (t) carrying out single-target door-opening risk assessment on the data, and calculating a continuous risk assessment value R of each target 1 (t)、R 2 (t)、...、R n (t), generating a total door-rushing risk RH (t) of the total passengers facing the door by integrating the single-target risks;
and step S4: and carrying out graded door-opening early warning treatment based on the door-opening risk RH (t) of the total passengers facing the door and a preset secondary threshold value.
In some embodiments, as shown in fig. 1, a method for warning about door violation of a passenger who is closing a subway platform screen door based on video analysis includes the steps of: starting a passenger door-rushing early warning terminal system installed at a shielding door head at the moment of closing a door, and continuously acquiring video frames by a video acquisition module according to a given frame rate; the video analysis module continuously calculates the motion track of the pedestrian target in the front door area based on YOLOv5 and Deepsort; the door-rushing risk evaluation module calculates the total passenger door-rushing risk faced by the door based on the motion tracks of all the pedestrian targets; and performing door-violation early warning treatment comprising acousto-optic alarm and door control linkage on the condition that the evaluation risk exceeds a given threshold value.
The specific steps in this example are as follows:
s1: according to the time plan of a train running scheduling signal system, a passenger door-rushing early warning terminal system installed at a shield door head is opened when the distance from the door-closing time is still T seconds (according to the specific configuration of opening door-rushing and early warning at a station, the T value is recommended to be set to be 10S), the passenger door-rushing early warning function is started, a video acquisition module of the system utilizes a camera installed at an oblique downward visual angle right above the shield door to continuously acquire a video frame I of a visual field F according to a frame rate R (for ensuring the real-time performance of motion detection, the recommended frame rate is not lower than 720@ 60fps) 1 、I 2 、....；
S2: for the current frame I i Detecting a pedestrian target based on a YOLOv5 neural network and associating the pedestrian target with a detected target in a preamble frame based on Deepsort to realize the target O 1 、O 2 、...、O n Wherein n is the number of all pedestrian targets, the motion trajectory of the target in the field of view F is converted into a continuous motion trajectory Tr corresponding to the target in the front door area H by perspective transformation (the perspective transformation matrix is calculated in advance based on parameters such as camera lens parameters and installation position, and the uniform perspective transformation matrix is shared by all lines while ensuring the installation accuracy) 1 (t)、Tr 2 (t)、...Tr n (t)；
S3: by pedestriansObject O 1 、O 2 、...、O n Movement track Tr in the front door region H i (t) carrying out single-target door-violation risk assessment on the data, and calculating a continuous risk assessment value R of each target 1 (t)、R 2 (t)、...、R n (t), generating a total door-opening risk RH (t) of the facing passengers by integrating the single-target risks;
s4: based on the door-rushing risk RH (t) of the general passengers and a set secondary threshold, graded door-rushing early warning treatment comprising sound-light alarm and door control linkage is carried out.
In this embodiment, the step S2 specifically includes the following steps:
s201: for the current frame I collected in the step S1 i Detecting pedestrian target detection based on a YOLOv5 neural network, and detecting and outputting O 1 、O 2 、...、O n N pedestrian targets, and outputting NULL if no pedestrian target is detected;
s202: if the output of S201 is NULL, the step S201 is executed circularly for the next frame I output by S1 i+1 Performing a pedestrian target detection operation until T = T; if the pedestrian target is detected in the step S201, according to the previous frame I i-1 Performing the associated tracking operation of the Deepsort target on the recorded target detection result to form O 1 、O 2 、...、O n Motion track K of object in field of view F 1 (t)、K 2 (t)、...K n (t)；
S203: through a perspective transformation matrix calibrated in advance, K is converted 1 (t)、K 2 (t)、...K n (t) into a trajectory Tr in top view coordinates corresponding to the door front region H 1 (t)、Tr 2 (t)、...Tr n (t)；
In this embodiment, the step S2 specifically includes the following steps:
s301: generating motion trail Tr for all detected i Target O of (t) i Continuously estimating the angle u between the moving direction and the Y axis i As shown in fig. 2, the coordinate system of the door front area H is defined by taking the downward direction of the center line as the Y axis and the rightward direction of the upper edge line of the H area as the X axis;
s302: for allDetects and generates a motion trail Tr i Target O of (t) i Continuously estimating the Y-axis component v of the motion velocity thereof i As shown in fig. 2, the coordinate system of the door front area H is defined by taking the downward direction of the center line as the Y axis and the rightward direction of the upper edge line of the H area as the X axis;
s303: generating motion trail Tr for all detected i Target O of (t) i Recording object O i A distance D from the bottom line of the H region in the Y direction i According to R i (t)＝max(cos(u i ),0)×max(1.5-v i ×(T-t)/D i 0) calculating the target O at the current time t i Risk of door break.
S304: the total risk of door break through for the passengers for calculating the door position is RH (t) = max (R) 1 (t),R 2 (t),...,R n (t))。
In this embodiment, step S4 specifically includes the following steps:
s401: the risk of making a break of the door RH (t) for the total passengers exceeds a threshold Th 1 (according to the sensitivity requirement of the door-violation risk assessment, th 1 The recommendation is set to 0.25), and reminding is carried out through an acousto-optic warning device arranged above the shielding door;
s402: the total passenger door-break risk RH (t) exceeds a specific threshold Th 1 Higher threshold Th 2 (according to the sensitivity requirement of the door-violation risk assessment, th 2 The suggestion is set to be 0.5), sound and light alarm is kept, the shielding door and the vehicle door control system are linked to keep the vehicle door open, and the shielding door and the vehicle door are linked to close the shielding door and the vehicle door after the total passenger door rushing risk RH (t) of the vehicle door returns to zero;
s403: a video recording of a passenger break-through is recorded and generated and saved while performing primary and secondary responses.
Example 2
In a second aspect, the embodiment provides a video analysis-based early warning device for a passenger who is close to a screen door of a subway and rushes to close the door, which comprises a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method of embodiment 1.
Example 3
In a third aspect, the present embodiment provides a storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of embodiment 1.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be appreciated by those skilled in the art that the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The embodiments disclosed above are therefore to be considered in all respects as illustrative and not restrictive. All changes which come within the scope of or equivalence to the invention are intended to be embraced therein.

Claims (9)

1. A subway shield door passenger door-rushing early warning method based on video analysis is characterized by comprising the following steps:
step S1: continuously acquiring video frames I of a view field F according to a preset frame rate R when T seconds are left before the door closing time 1 、I 2 、....I i 、....；
Step S2: for the current video frame I i Detecting a pedestrian target based on a YOLOv5 neural network and associating the pedestrian target with a detected target in a preamble video frame based on Deepsort to realize the target O 1 、O 2 、...、O n Wherein n is the number of all pedestrian targets, the motion track of the target in the visual field F is converted into the corresponding target continuous motion track Tr in the front door area H through perspective transformation 1 (t)、Tr 2 (t)、...Tr n (t)；
And step S3: using pedestrian objects O 1 、O 2 、...、O n Movement track Tr in the door front region H i (t) carrying out single-target door-opening risk assessment on the data, and calculating a continuous risk assessment value R of each target 1 (t)、R 2 (t)、...、R n (t), generating a total door-rushing risk RH (t) of the total passengers facing the door by integrating the single-target risks;
and step S4: and carrying out graded door-rushing early warning treatment based on the door-rushing risk RH (t) of the total passengers facing the door and a preset secondary threshold value.
2. The subway platform screen door passenger door-break early warning method based on video analysis as claimed in claim 1, wherein in step S1, the method for acquiring video frames of the field of view F comprises:
according to the time plan of a train running scheduling signal system, a passenger door-rushing early warning terminal system installed on a shielding door head starts a door-rushing early warning function for a passenger who is close to a door when the distance from the door-closing time is still T seconds, and a video acquisition module of the system acquires video frames by using a camera installed at an oblique downward visual angle right above the shielding door.
3. The subway shield door passenger door-break early warning method based on video analysis as claimed in claim 1, wherein said step S2 comprises:
s201: for the current video frame I collected in the step S1 i Detecting pedestrian target detection based on a YOLOv5 neural network, and detecting and outputting O 1 、O 2 、...、O n N pedestrian targets, and outputting NULL if no pedestrian target is detected;
s202: if the output of S201 is NULL, the step S201 is executed circularly for the next frame I output from S1 i+1 Carrying out pedestrian target detection operation until the current time T = T;
if the pedestrian object is detected in step S201, the pedestrian object is detected according to the previous frame I i-1 Performing the associated tracking operation of the Deepsort target on the recorded target detection result to form O 1 、O 2 、...、O n Motion track K of object in field of view F 1 (t)、K 2 (t)、...K n (t)；
S203: through a perspective transformation matrix calibrated in advance, K is converted 1 (t)、K 2 (t)、...K n (t) transformation into a trajectory Tr in plan view coordinates corresponding to the door front region H 1 (t)、Tr 2 (t)、...Tr n (t)。
4. The subway shield door passenger door-break early warning method based on video analysis as claimed in claim 1, wherein said step S3 comprises:
s301: generating motion trail Tr for all detected i Target O of (t) i Continuously estimating the angle u between the moving direction and the Y axis i The coordinate system of the door front area H is defined as taking the downward direction of a central line as an axis Y and taking the rightward direction of an upper sideline of the area H as an axis X;
s302: generating motion trail Tr for all detected i Target O of (t) i Continuously estimating the Y-axis component v of the motion velocity thereof i The coordinate system of the door front area H is defined to take the downward direction of the center line as the Y axis and the rightward direction of the upper edge line of the area H as the X axis;
s303: generating motion trail Tr for all detected i Target O of (t) i Recording object O i A distance D from the bottom line of the H region in the Y direction i ；
According to R i (t)＝max(cos(u i ),0)×max(1.5-v i ×(T-t)/D i 0) calculating the target O at the current time t i Risk of door violation;
s304: calculating the total passenger door-rushing risk of the door as follows: RH (t) = max (R) 1 (t),R 2 (t),...,R n (t))。
5. The subway platform screen door passenger door-break early warning method based on video analysis as claimed in claim 1, wherein the step S4 comprises:
the risk of total passenger breakthrough RH (t) exceeding a first threshold Th 1 The first-level response of the system is that passengers are reminded through an acousto-optic warning device arranged above the shielding door;
the risk of total passenger breakthrough RH (t) exceeds a second threshold Th 2 The shielding door and the vehicle door control system are linked to keep the shielding door and the vehicle door open while the acousto-optic alarm is kept, and the shielding door and the vehicle door are closed by the shielding door and the vehicle door control system until the total passenger door rushing risk RH (t) of the door returns to zero, wherein the second step is thatThreshold Th 2 Greater than a first threshold Th 1 。
6. The subway platform screen door passenger door-break early warning method based on video analysis as claimed in claim 5, wherein the step S4 further comprises:
a video recording of a passenger breaching the door is recorded and generated and saved while a primary or secondary response is made.
7. The subway shield door passenger door-rushing early warning method based on video analysis as claimed in claim 5, wherein the first threshold Th 1 0.25, second threshold Th 2 Is 0.5.
8. A subway shield door passenger door-rushing early warning device based on video analysis is characterized by comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any one of claims 1 to 7.
9. A storage medium having a computer program stored thereon, the computer program, when being executed by a processor, performing the steps of the method of any one of claims 1 to 7.
CN202211251732.9A 2022-10-13 2022-10-13 Subway shielding door close-door passenger door-opening early warning method based on video analysis Active CN115620228B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211251732.9A CN115620228B (en) 2022-10-13 2022-10-13 Subway shielding door close-door passenger door-opening early warning method based on video analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211251732.9A CN115620228B (en) 2022-10-13 2022-10-13 Subway shielding door close-door passenger door-opening early warning method based on video analysis

Publications (2)

Publication Number Publication Date
CN115620228A true CN115620228A (en) 2023-01-17
CN115620228B CN115620228B (en) 2023-05-23

Family

ID=84863445

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211251732.9A Active CN115620228B (en) 2022-10-13 2022-10-13 Subway shielding door close-door passenger door-opening early warning method based on video analysis

Country Status (1)

Country Link
CN (1) CN115620228B (en)

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103198605A (en) * 2013-03-11 2013-07-10 成都百威讯科技有限责任公司 Indoor emergent abnormal event alarm system
CN106219367A (en) * 2016-08-05 2016-12-14 沈阳聚德视频技术有限公司 A kind of elevator O&M based on intelligent vision light curtain monitoring method
CN206298233U (en) * 2016-11-07 2017-07-04 何桂尧 A kind of Elevator Monitoring is saved oneself disposal device
CN108001460A (en) * 2017-12-12 2018-05-08 徐薇 Subway shield door gauge area intelligent anti-clip detection warning device and method
CN110008867A (en) * 2019-03-25 2019-07-12 五邑大学 A kind of method for early warning based on personage's abnormal behaviour, device and storage medium
CN209765074U (en) * 2018-12-27 2019-12-10 上海仁童电子科技有限公司 Platform shield door foreign matter detection device
CN110895861A (en) * 2018-09-13 2020-03-20 杭州海康威视数字技术股份有限公司 Abnormal behavior early warning method and device, monitoring equipment and storage medium
CN112509190A (en) * 2021-02-08 2021-03-16 南京信息工程大学 Subway vehicle section passenger flow statistical method based on shielded gate passenger flow counting
CN113343960A (en) * 2021-08-06 2021-09-03 南京信息工程大学 Method for estimating and early warning passenger flow retained in subway station in real time
CN113658192A (en) * 2021-07-08 2021-11-16 华南理工大学 Multi-target pedestrian track acquisition method, system, device and medium
CN114663390A (en) * 2022-03-22 2022-06-24 平安普惠企业管理有限公司 Intelligent anti-pinch method, device, equipment and storage medium for automatic door
CN115063730A (en) * 2022-07-13 2022-09-16 山东建筑大学 Video track analysis-based method and system for early warning of intrusion of workers into borderline area
CN115131821A (en) * 2022-06-29 2022-09-30 大连理工大学 Improved YOLOv5+ Deepsort-based campus personnel crossing warning line detection method

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103198605A (en) * 2013-03-11 2013-07-10 成都百威讯科技有限责任公司 Indoor emergent abnormal event alarm system
CN106219367A (en) * 2016-08-05 2016-12-14 沈阳聚德视频技术有限公司 A kind of elevator O&M based on intelligent vision light curtain monitoring method
CN206298233U (en) * 2016-11-07 2017-07-04 何桂尧 A kind of Elevator Monitoring is saved oneself disposal device
CN108001460A (en) * 2017-12-12 2018-05-08 徐薇 Subway shield door gauge area intelligent anti-clip detection warning device and method
CN110895861A (en) * 2018-09-13 2020-03-20 杭州海康威视数字技术股份有限公司 Abnormal behavior early warning method and device, monitoring equipment and storage medium
CN209765074U (en) * 2018-12-27 2019-12-10 上海仁童电子科技有限公司 Platform shield door foreign matter detection device
CN110008867A (en) * 2019-03-25 2019-07-12 五邑大学 A kind of method for early warning based on personage's abnormal behaviour, device and storage medium
CN112509190A (en) * 2021-02-08 2021-03-16 南京信息工程大学 Subway vehicle section passenger flow statistical method based on shielded gate passenger flow counting
CN113658192A (en) * 2021-07-08 2021-11-16 华南理工大学 Multi-target pedestrian track acquisition method, system, device and medium
CN113343960A (en) * 2021-08-06 2021-09-03 南京信息工程大学 Method for estimating and early warning passenger flow retained in subway station in real time
CN114663390A (en) * 2022-03-22 2022-06-24 平安普惠企业管理有限公司 Intelligent anti-pinch method, device, equipment and storage medium for automatic door
CN115131821A (en) * 2022-06-29 2022-09-30 大连理工大学 Improved YOLOv5+ Deepsort-based campus personnel crossing warning line detection method
CN115063730A (en) * 2022-07-13 2022-09-16 山东建筑大学 Video track analysis-based method and system for early warning of intrusion of workers into borderline area

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
LING F等: "Infrared multi-target tracking based on Deep-Sort optimization algorithm", 《2021 INTERNATIONAL CONFERENCE ON CONTROL, AUTOMATION AND INFORMATION SCIENCES (ICCAIS)》 *
张梦华: "基于Yolov5和DeepSort的视频行人识别与跟踪探究", 《现代信息科技》 *
汪晓臣等: "物联网设备的深度学习故障预测方法", 《小型微型计算机系统》 *

Also Published As

Publication number Publication date
CN115620228B (en) 2023-05-23

Similar Documents

Publication Publication Date Title
Ijjina et al. Computer vision-based accident detection in traffic surveillance
Chen et al. Surrogate safety analysis of pedestrian‐vehicle conflict at intersections using unmanned aerial vehicle videos
CN102945603B (en) Method for detecting traffic event and electronic police device
CN104809887B (en) A kind of retrograde detection method of vehicle on expressway and autoalarm
US11380105B2 (en) Identification and classification of traffic conflicts
Ki et al. A traffic accident recording and reporting model at intersections
US20130093895A1 (en) System for collision prediction and traffic violation detection
CN110473402A (en) A kind of accident detection early warning system based on target abnormal behaviour trajectory analysis
WO2018096371A1 (en) Passenger transport monitoring system
CN105744232A (en) Method for preventing power transmission line from being externally broken through video based on behaviour analysis technology
KR101101860B1 (en) Monitoring system for traffic condition
CN110288738B (en) Intelligent comprehensive management and control system and method for bicycle lane
Nakashima et al. Passenger counter based on random forest regressor using drive recorder and sensors in buses
US20160241839A1 (en) System for traffic behaviour surveillance
CN108289203A (en) A kind of video monitoring system for rail traffic
CN112306051A (en) Robot system for unmanned traffic police vehicle on highway
CN114494998B (en) Intelligent analysis method and system for vehicle data
CN115909223A (en) Method and system for matching WIM system information with monitoring video data
Ki Accident detection system using image processing and MDR
Zheng Developing a traffic safety diagnostics system for unmanned aerial vehicles usingdeep learning algorithms
CN111062238A (en) Escalator flow monitoring method and system based on human skeleton information and multi-target tracking
CN115620228B (en) Subway shielding door close-door passenger door-opening early warning method based on video analysis
CN113538968B (en) Method and apparatus for outputting information
Bahloul et al. Adding technological solutions for safety improvement at level crossings: a functional specification
CN118135768B (en) Safety monitoring method for highway toll station personnel under dynamic operation environment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant