CN110376585A - Compartment crowding detection method and device, system based on 3D radar scanning - Google Patents
Compartment crowding detection method and device, system based on 3D radar scanning Download PDFInfo
- Publication number
- CN110376585A CN110376585A CN201910666603.8A CN201910666603A CN110376585A CN 110376585 A CN110376585 A CN 110376585A CN 201910666603 A CN201910666603 A CN 201910666603A CN 110376585 A CN110376585 A CN 110376585A
- Authority
- CN
- China
- Prior art keywords
- compartment
- target
- dimensional character
- passenger
- character data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/415—Identification of targets based on measurements of movement associated with the target
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Train Traffic Observation, Control, And Security (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The embodiment of the invention provides a kind of compartment crowding detection method and device, system based on 3D radar scanning, obtain the point cloud data of the passenger getting on/off for the 3D radar scanning being mounted on above compartment car door, two dimensional character data are obtained from point cloud data, then mark the corresponding target two dimensional character data of visitor of starting a work shift from two dimensional character data by identification model trained in advance.Track the direction of travel for the passenger that each target two dimensional character data obtain corresponding to it, and then the number got on the bus and got off at current station is counted according to the direction of travel of each passenger, when calculating train and sailing out of the station, passengers quantity in compartment, next station is sent by passengers quantity, is shown at next station.Before passenger reaches next station, the passengers quantity in each compartment of statistics is shown, the passenger to clamp on is enable to select platform according to the passengers quantity in each compartment, avoids blindly improving the carrying capacity of train to vehicle.
Description
Technical field
The present invention relates to train transport power technical fields, examine more particularly, to a kind of compartment crowding based on 3D radar scanning
Survey method and device, system.
Background technique
In the process of running, before train gets to the station, the passenger on station can not learn in each compartment train at present
Passengers quantity.Thus when waiting since the congested conditions of every array carriage can not be predicted, can only blindness waiting.Especially peak
Period, the vehicle parking time is short, and passenger can not predict the degree of crowding of every array carriage, may cause some compartments passenger and gathers around very much
It squeezes, and some cabin spaces are very loose, cause space waste, reduce the carrying capacity of train.It, can also be when passenger is excessive
Interior passenger's delay of standing is caused to a certain extent, influences normal operation and management in arriving at a station.
In actual application, inventor has found in existing train travelling process that passenger, which can not learn, to be reached
Train in each compartment passengers quantity situation, cause blindly to vehicle, influence train transport power.
Summary of the invention
The embodiment of the present invention provides a kind of compartment crowding detection method and device, system based on 3D radar scanning, uses
To solve in train travelling process in the prior art, passenger can not learn the ridership in each compartment in the train that will be reached
Situation is measured, the problem of blindly influencing train transport power to vehicle is caused.
Against the above technical problems, in a first aspect, the embodiment provides a kind of vehicles based on 3D radar scanning
Compartment crowding detection method, comprising:
To the either objective compartment of train, the passenger obtained by 3D radar scanning is obtained at current station by the mesh
The point cloud data that the car door in mark compartment is got on or off the bus, and two dimensional character data are obtained from the point cloud data；
The scanning for being scanned and being formed to passenger is marked from two dimensional character data by identification model trained in advance
Point tracks each target two dimensional character data as target two dimensional character data, obtains that target two dimensional character data are corresponding to be multiplied
The direction of travel of visitor；
Enter first passengers quantity in the target compartment at the current station according to the direction of travel of each passenger statistics
With the second passengers quantity for walking out the target compartment, obtains having in target compartment when train sails out of a station and multiply
Objective quantity calculates train according to first passengers quantity, second passengers quantity and the existing passengers quantity and sails out of institute
Target passengers quantity when current station in the target compartment is stated, the target passengers quantity is sent, with aobvious at next station
Show the target passengers quantity；
Wherein, the identification model from two dimensional character data for marking the data for meeting humanoid feature, as right
Passenger is scanned the target two dimensional character data to be formed；What two dimensional character data referred to intercepting from point cloud data includes sweeping
The set of described point and the scanning element being parallel on the two-dimensional surface of target compartment car door.
Optionally, the training of the identification model includes:
The point cloud data that is scanned in advance to the passenger to get on or off the bus is obtained, by two-dimentional peak-seeking by the point cloud that obtains
The two dimensional character data that data obtain, and the data to be formed will be scanned to passenger in each two dimensional character data and marked
Note, using the two dimensional character data before being marked as the input parameter of deep learning, the two dimension after being marked is special
Desired output of the data as deep learning is levied, using the model trained by deep learning as the identification model.
Optionally, each target two dimensional character data of tracking, obtain the corresponding passenger's of target two dimensional character data
Direction of travel, comprising:
To each target two dimensional character data, the target two dimensional character data are obtained in upper primary scanning or next time
The first position in obtained point cloud data is scanned, and obtains the point that the target two dimensional character data present scan obtains
The second position in cloud data determines the target two dimension according to first position and second position
The direction of travel of the corresponding passenger of characteristic.
Optionally, described to send the target passengers quantity, to show the target passengers quantity at next station, packet
It includes:
The mapping relations between the passengers quantity in the preset compartment degree of crowding and compartment are obtained, are reflected according to described
It penetrates relationship and the target passengers quantity determines the degree of crowding in target compartment when train sails out of the current station, as
The target degree of crowding sends passenger information system PIS for the target passengers quantity and the target degree of crowding, to pass through
The display equipment at next station shows the target passengers quantity and the target degree of crowding；
Wherein, the target degree of crowding is indicated by preset color corresponding with the target degree of crowding.
Optionally, the mapping relations include:
When the passengers quantity in compartment is less than the first preset quantity, the compartment degree of crowding is the seat compartment Nei You；
When the passengers quantity in compartment is greater than or equal to first preset quantity and when less than the second preset quantity, compartment
The degree of crowding be compartment in without seat but loosely；
When the passengers quantity in compartment is greater than or equal to second preset quantity and is less than third preset quantity, compartment
The degree of crowding is more crowded in compartment；
When the passengers quantity in compartment is greater than or equal to the third preset quantity, the compartment degree of crowding is non-in compartment
It is often crowded；
Wherein, first preset quantity is equal to the amount of seats configured in compartment.
It is optionally, described that two dimensional character data are obtained from the point cloud data, comprising:
It is intercepted from the point cloud data by two-dimentional peak-seeking and contains the summit formed by scanning element and be parallel to described
The two-dimensional surface of target compartment car door, using the set of the scanning element on the two-dimensional surface of interception as two dimensional character data.
Optionally, further includes:
After receiving train position and car door opening state by Train Control and management system TCMS transmission, if judgement
Train is in main track operation and car door is in the open state, then sends to 3D radar and open prompt, so that 3D radar starts to sweep
It retouches and obtains the point cloud data by the passenger that the car door in the target compartment is got on or off the bus at current station.
Optionally, further includes:
After receiving train position and car door opening state by Train Control and management system TCMS transmission, if judgement
Train is located at terminus or car door is not opened, then does not send the unlatching prompt to 3D radar.
Second aspect, the embodiment provides a kind of compartment ridership amount detection device based on 3D radar scanning
It sets, comprising:
Module is obtained, for the either objective compartment to train, obtains the passenger obtained by 3D radar scanning current
The point cloud data that station is got on or off the bus by the car door in the target compartment, and two dimensional character number is obtained from the point cloud data
According to；
Processing module sweeps passenger for being marked from two dimensional character data by identification model trained in advance
The scanning element to be formed is retouched, as target two dimensional character data, each target two dimensional character data is tracked, obtains target two dimensional character
The direction of travel of the corresponding passenger of data；
Sending module, for entering the target compartment at the current station according to the direction of travel of each passenger statistics
First passengers quantity and the second passengers quantity for walking out the target compartment obtain train and sail out of a target carriage when station
Existing passengers quantity in compartment, according to first passengers quantity, second passengers quantity and the existing ridership meter
It calculates train and sails out of target passengers quantity when the current station in the target compartment, send the target passengers quantity to
Next station；
Wherein, the identification model from two dimensional character data for marking the data for meeting humanoid feature, as right
Passenger is scanned the target two dimensional character data to be formed；What two dimensional character data referred to intercepting from point cloud data includes sweeping
The set of described point and the scanning element being parallel on the two-dimensional surface of target compartment car door.
The third aspect, the compartment passengers quantity detection system based on 3D radar scanning that the embodiment provides a kind of
System, including data processing unit and the 3D radar being arranged in above each compartment car door of train；
Every 3D radar connection Train Control and management system TCMS and data processing unit, data processing unit connection
PIS；
Wherein, to the either objective compartment of train, the 3D radar above target compartment car door is set and is being received
Start to scan at current station after the unlatching for the passenger that the car door in the target compartment is got on or off the bus prompt, opens 3D radar and sweep
The passenger to get on or off the bus at current station by the car door in the target compartment is retouched, point cloud data is obtained；
The data processing unit is used to execute the compartment crowding inspection described in any of the above item based on 3D radar scanning
Survey method.
Fourth aspect the embodiment provides a kind of electronic equipment, including memory, processor and is stored in
On reservoir and the computer program that can run on a processor, the processor realize any of the above item institute when executing described program
The step of compartment crowding detection method based on 3D radar scanning stated.
5th aspect, the embodiment provides a kind of non-transient computer readable storage mediums, are stored thereon with
Computer program realizes the compartment based on 3D radar scanning described in any of the above item when the computer program is executed by processor
The step of crowding detection method.
The compartment crowding detection method and device that the embodiment provides a kind of based on 3D radar scanning are
System obtains the point cloud data of the passenger getting on/off for the 3D radar scanning being mounted on above compartment car door, obtains from point cloud data
Two dimensional character data, then the corresponding target two dimension of the visitor that starts a work shift is marked from two dimensional character data by identification model trained in advance
Characteristic.The direction of travel for the passenger that each target two dimensional character data obtain corresponding to it is tracked, and then according to each passenger
Direction of travel count the number getting on the bus and get off at current station, passenger when calculating train and sailing out of the station, in compartment
Passengers quantity is sent next station by quantity, shows at next station.Before passenger reaches next station, by statistics
The passengers quantity in each compartment is shown, so that the passenger to clamp on is selected platform according to the passengers quantity in each compartment, is avoided
Blindly to vehicle, the carrying capacity of train is improved.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to do one simply to introduce, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of stream of compartment crowding detection method based on 3D radar scanning provided by one embodiment of the present invention
Journey schematic diagram；
Fig. 2 is the schematic illustration for the compartment congestion state detection that another embodiment of the present invention provides；
Fig. 3 is that the passenger that another embodiment of the present invention provides passes in and out walking direction flow diagram；
Fig. 4 is demographics flow diagram in the compartment of another embodiment of the present invention offer；
Fig. 5 is a kind of compartment passengers quantity detection device based on 3D radar scanning that another embodiment of the present invention provides
Structural block diagram；
Fig. 6 is the state flow diagram that the 3D radar that another embodiment of the present invention provides opens and closes；
Fig. 7 is the structural block diagram for the electronic equipment that another embodiment of the present invention provides.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Fig. 1 is a kind of process signal of compartment crowding detection method based on 3D radar scanning provided in this embodiment
Figure, referring to Fig. 1, this method comprises:
101: to the either objective compartment of train, obtaining the passenger obtained by 3D radar scanning at current station by institute
The point cloud data that the car door in target compartment is got on or off the bus is stated, and obtains two dimensional character data from the point cloud data；
102: marking by identification model trained in advance to be scanned passenger from two dimensional character data to be formed
Scanning element tracks each target two dimensional character data as target two dimensional character data, and it is corresponding to obtain target two dimensional character data
Passenger direction of travel；
103: first passenger in the target compartment is entered at the current station according to the direction of travel of each passenger statistics
Quantity and the second passengers quantity for walking out the target compartment obtain train and sail out of when a station in the target compartment
There is passengers quantity, train is calculated according to first passengers quantity, second passengers quantity and the existing passengers quantity and is sailed
Target passengers quantity when from the current station in the target compartment sends the target passengers quantity, in next vehicle
It stands and shows the target passengers quantity；
Wherein, the identification model from two dimensional character data for marking the data for meeting humanoid feature, as right
Passenger is scanned the target two dimensional character data to be formed；What two dimensional character data referred to intercepting from point cloud data includes sweeping
The set of described point and the scanning element being parallel on the two-dimensional surface of target compartment car door.
Method provided in this embodiment is executed by being equipped with the equipment for executing the software of the above method, which can be meter
Calculation machine, dedicated processing equipment are integrated in same equipment with 3D radar, and the present embodiment is not particularly limited this.
3D radar is generally arranged at the top of compartment car door, and when car door opening, 3D radar carries out space below
Scanning, obtains the point cloud data of passenger getting on/off.What two dimensional character data referred to intercepting from point cloud data is parallel to target
The set of scanning element on the two-dimensional surface of compartment car door.In order to reduce operand, calculation can be passed through when intercepting two-dimensional surface
Method interception contains the two-dimensional surface of the scanning element with certain features, for example, can be intercepted by two-dimentional peak-seeking comprising by sweeping
The two-dimensional surface for the summit that described point is formed.Wherein, there are peaks for finding from three dimensional point cloud for this algorithm of two-dimentional peak-seeking
The position of value, and the two-dimensional surface of car door place plane is intercepted there are peak value and is parallel to, the point on the two-dimensional surface is two
Dimensional feature data.In general, by calling the two-dimentional peak-seeking function in Matlab software that can obtain two-dimentional spy from point cloud data
Levy data.Identification model be train in advance come can distinguish whether two dimensional character data are that the point to be formed is scanned to people
Model, can accurately mark the data for being scanned and being formed to passenger.
Two dimensional character data are that there are peak value and the set of point that is parallel on the two-dimensional surface of car door, target two dimensional characters
Data are to be scanned the set for the point to be formed in the two-dimensional surface to a certain passenger.That is, in a frame two dimensional character
The each target two dimensional character data marked in data correspond to a passenger.
Further, it is calculated according to first passengers quantity, second passengers quantity and the existing passengers quantity
Train sails out of target passengers quantity when the current station in the target compartment, comprising:
The difference of the existing passengers quantity Yu second passengers quantity is calculated, then calculates the difference and described first
The sum of passengers quantity, using the sum of calculating as the target passengers quantity.
A kind of compartment crowding detection method based on 3D radar scanning is present embodiments provided, acquisition is mounted on compartment vehicle
The point cloud data of the passenger getting on/off of the 3D radar scanning of door top, obtains two dimensional character data, then pass through from point cloud data
Trained identification model marks the corresponding target two dimensional character data of visitor of starting a work shift from two dimensional character data in advance.Track each mesh
Mark two dimensional character data obtain the direction of travel of the passenger corresponding to it, and then are counted according to the direction of travel of each passenger current
The number that station gets on the bus and gets off, when calculating train and sailing out of the station, passengers quantity in compartment sends passengers quantity to
Next station is shown at next station.Before passenger reaches next station, the passengers quantity in each compartment of statistics is shown
Show, the passenger to clamp on is enable to select platform according to the passengers quantity in each compartment, avoids blindly improving the delivery of train to vehicle
Ability.
Further, on the basis of the various embodiments described above, the training of the identification model includes:
The point cloud data that is scanned in advance to the passenger to get on or off the bus is obtained, by two-dimentional peak-seeking by the point cloud that obtains
The two dimensional character data that data obtain, and the data to be formed will be scanned to passenger in each two dimensional character data and marked
Note, using the two dimensional character data before being marked as the input parameter of deep learning, the two dimension after being marked is special
Desired output of the data as deep learning is levied, using the model trained by deep learning as the identification model.
Specifically, Fig. 2 is the schematic illustration of compartment congestion state provided in this embodiment detection, referring to fig. 2, by taking
It is downloaded to compartment door outside 3D radar, acquires the disengaging situation of Vehicular occupant, the spy of the number of people and shoulder is obtained by two-dimentional peak-seeking
Distribution of the point value on two-dimensional space is levied, collected characteristic is put into identification model trained in advance, so that mould
Type can by be transmitted through come character numerical value judge whether as people.After completing Human detection, the feelings that are obtained by 3D radar scanning
Condition, judges the direction of motion of the humanoid data of every frame, to obtain the number of disengaging.After having judged disengaging direction, by entering people
Number subtracts away number to count the number in compartment.Finally by the number in compartment, sent out by 3D radar scanning system
It is sent in the PIS display system of the next stop, passenger is allowed intuitively to see each compartment by the platform CCTV of the next stop
Crowding situation.
Further, the point cloud data for being trained to the identification model includes being scanned to child
Point cloud data, to exist mutually point cloud data that the parallel passenger blocked is scanned and to there are when shelter to passenger
The point cloud data being scanned.Wherein, shelter includes the objects such as school bag, luggage and passenger's knapsack.
As shown in Fig. 2, need to carry out the training and test of model before carrying out the detection of compartment congestion state, for into
The data as training sample acquired when row model training need type abundant, are related to a variety of data cases, such as child's 3D thunder
Up to scan data and the radar scanning data blocked etc. of parallel passenger.By two-dimentional peak-seeking, acquisition characteristics point data is carried out
Second order derivation and smoothing processing, and carry out characteristic and be labeled.The present embodiment carries out model using the method for deep learning
Training, for example, using Tensorflow deep learning frame, the data that will be collected are put into corresponding label (label)
Model training is carried out in model, obtains identification model.Further, it is carried out during training pattern using test the set pair analysis model
Adjustment, so that the Human detection rate of model is up to 92 or more percent.
A kind of compartment crowding detection method based on 3D radar scanning is present embodiments provided, deep learning training is passed through
The model that the corresponding data of passenger are identified in two dimensional character data is entered data into using deep learning algorithm out
It can accurately judge whether to be people in model trace, can effectively evade the interference of classes personage's product such as school bag, identification is accurate
Degree height, strong antijamming capability.
Further, on the basis of the various embodiments described above, each target two dimensional character data of tracking obtain target
The direction of travel of the corresponding passenger of two dimensional character data, comprising:
To each target two dimensional character data, the target two dimensional character data are obtained in upper primary scanning or next time
The first position in obtained point cloud data is scanned, and obtains the point that the target two dimensional character data present scan obtains
The second position in cloud data determines the target two dimension according to first position and second position
The direction of travel of the corresponding passenger of characteristic.
During judging direction of travel, in order to avoid having obscured the target two dimensional character obtained to different passenger scans
Data can add the step of whether the target two dimensional character data that verifying front and back twice sweep obtains correspond to same passenger.
For example, if target two dimensional character data are in the point cloud data of position and present scan in the upper point cloud data once scanned
Positional distance is smaller, then the target two dimensional character data of twice sweep belong to same passenger, can be according to target in twice sweep
The change in location of two dimensional character data judges passenger's direction of travel, and otherwise, by present scan and scanning next time obtains target
Two dimensional character data judge passenger's direction of travel.
For example, the point cloud for obtaining the target two dimensional character data and being scanned in upper primary scanning or next time
The first position in data, comprising:
If the peak value position of the target two dimensional character data of last time scanning and the target of present scan
The distance between peak value position of two dimensional character data is less than or equal to pre-determined distance, then the target two dimension of twice sweep is special
Sign data are the corresponding target two dimensional character data of same passenger, obtain the target two dimensional character data and primary scan upper
To point cloud data in the first position, otherwise, the target two dimensional character data of twice sweep are not that same passenger is corresponding
Target two dimensional character data, obtain first of the target two dimensional character data in the point cloud data scanned next time
Position.
During judging passenger's direction of travel, if last the first place for scanning the target two dimensional character data
Second position of target two dimensional character data described in position to present scan is directed toward in the target compartment, then the target
The corresponding passenger of two dimensional character data enters the target compartment.
If last time scanning the first position to the second position of present scan in the target compartment,
Then the corresponding passenger of the target two dimensional character data walks out the target compartment.
Fig. 3 is that passenger provided in this embodiment disengaging walking direction flow diagram after humanoid judgement, passes through referring to Fig. 3
Humanoid (target two dimensional character data) position in two continuous frames of label judges the direction of travel of passenger.For example, to mesh
The peak value of mark two dimensional character data is tracked, and is obtained the direction of motion of the peak value in two continuous frames and then is determined the disengaging of passenger
Direction.
A kind of compartment crowding detection method based on 3D radar scanning is present embodiments provided, by special to target two dimension
The tracking of sign data realizes the judgement that compartment is passed in and out to passenger, convenient for the passengers quantity in statistics disengaging compartment.
Further, described to send the target passengers quantity on the basis of the various embodiments described above, at next station
Show the target passengers quantity, comprising:
The mapping relations between the passengers quantity in the preset compartment degree of crowding and compartment are obtained, are reflected according to described
It penetrates relationship and the target passengers quantity determines the degree of crowding in target compartment when train sails out of the current station, as
The target degree of crowding sends passenger information system PIS for the target passengers quantity and the target degree of crowding, to pass through
The display equipment at next station shows the target passengers quantity and the target degree of crowding；
Wherein, the target degree of crowding is indicated by preset color corresponding with the target degree of crowding.
Further, when train is in the starting station and non-opening car door, by existing passengers quantity, the first passengers quantity and
Second passengers quantity is initialized as zero.
Fig. 4 is demographics flow diagram in compartment provided in this embodiment, and referring to fig. 4, system is from the train starting station
It rises and completes crowding state initialization, the number in number of people entering and compartment is set as zero.When train is run in section,
3D radar scanning system respectively adds up to disengaging compartment number after car door opening, by subtracting the number walked out into number
To obtain the effective strength in the compartment.When train reaches the end of run, TCMS is by location information and vehicle door status information
It is sent to 3D radar scanning system, system automatically resets the number in number of people entering and compartment, until train reaches real hair station
It runs again, system re-starts disengaging passenger and counts.
A kind of compartment crowding detection method based on 3D radar scanning is present embodiments provided, the compartment degree of crowding is passed through
Mapping relations between the passengers quantity in compartment divide compartment congestion state, convenient for passing through the display at next station
Equipment intuitively represents the degree of crowding in compartment.
Further, on the basis of the various embodiments described above, the mapping relations include:
When the passengers quantity in compartment is less than the first preset quantity, the compartment degree of crowding is the seat compartment Nei You；
When the passengers quantity in compartment is greater than or equal to first preset quantity and when less than the second preset quantity, compartment
The degree of crowding be compartment in without seat but loosely；
When the passengers quantity in compartment is greater than or equal to second preset quantity and is less than third preset quantity, compartment
The degree of crowding is more crowded in compartment；
When the passengers quantity in compartment is greater than or equal to the third preset quantity, the compartment degree of crowding is to gather around in compartment
It squeezes；
Wherein, first preset quantity is equal to the amount of seats configured in compartment.
The degree of crowding can also be indicated by different colors, intuitively to show the congestion state in each compartment, example
Such as, passengers quantity is greater than or equal to 310 people in compartment, then the degree of crowding is very crowded, returns the result to PIS as car number
And compartment number adds red；Passengers quantity is greater than or equal to 150 in compartment, and less than 310 people, then the degree of crowding is more crowded,
It returns the result to PIS as car number and compartment number plus yellow；Passengers quantity is greater than or equal to 40 in compartment, less than 150 people,
Then the degree of crowding is no seat but loose, is returned the result to PIS as car number and compartment number plus green；Ridership in compartment
Amount is less than 40 people, then the degree of crowding is to have seat, returns the result to PIS as car number and compartment number plus blue.
The degree of crowding and passengers quantity are sent to PIS system, and next station receives the congestion information in each compartment of train, and
Uniformly send result to progress result displaying in platform CCTV.Show that content includes the number in each compartment, each compartment
The number of crowding corresponding color, each compartment, when standing, interior train is sailed out of, and receives the train congestion information that will arrive at a station
When, PIS system completes display information with new.After withdrawal of train, system will no longer receive train congestion information, until next
Run the period starts, restarting.
A kind of compartment crowding detection method based on 3D radar scanning is present embodiments provided, to the degree of crowding in compartment
Detailed grade classification has been carried out, the degree of crowding in compartment is intuitively illustrated by color.
It is further, described that two dimensional character data are obtained from the point cloud data on the basis of the various embodiments described above,
Include:
It is intercepted from the point cloud data by two-dimentional peak-seeking and contains the summit formed by scanning element and be parallel to described
The two-dimensional surface of target compartment car door, using the set of the scanning element on the two-dimensional surface of interception as two dimensional character data.
A kind of compartment crowding detection method based on 3D radar scanning is present embodiments provided, is obtained by two-dimentional peak-seeking
Two dimensional character data, greater probability includes that the scanning element to be formed is scanned to passenger in the two dimensional character data of acquisition, is subtracted
Lack the calculation amount for identifying target two dimensional character data by identification model, improves computational efficiency.
Further, on the basis of the various embodiments described above, further includes:
After receiving train position and car door opening state by Train Control and management system TCMS transmission, if judgement
Train is in main track operation and car door is in the open state, then sends to 3D radar and open prompt, so that 3D radar starts to sweep
It retouches and obtains the point cloud data by the passenger that the car door in the target compartment is got on or off the bus at current station.
Further, on the basis of the various embodiments described above, further includes:
Receive by TCMS send train position and car door opening state after, if judge train be located at terminus or
Car door is not opened, then does not send the unlatching prompt to 3D radar.
It is provided with 3D radar above the compartment car door of each train, when the train position and car door that send according to TCMS are opened
It when the state of opening judges that train is in main track operation and car door in the open state, is sent to 3D radar and opens prompt, 3D radar exists
After receiving unlatching prompt, that is, starts to scan the passenger to get on or off the bus at current station by the car door in the target compartment, obtain
Otherwise point cloud data does not send to 3D radar and opens prompt.3D radar only can just start to mesh when receiving and opening prompt
The passenger that the car door in mark compartment is got on or off the bus is scanned.
For example, the algorithm integration for executing the above-mentioned compartment crowding detection method based on 3D radar scanning in the present embodiment exists
In 3D radar scanning system, then when 3D radar scanning system receives the location information and vehicle for the train that vehicle-mounted TCMS system is sent
When door state, if judging, train is in main track operation, and car door is in the open state, then sends to 3D radar and open prompt,
Start to scan to control 3D radar, obtains point cloud data.
A kind of compartment crowding detection method based on 3D radar scanning is present embodiments provided, by sending to 3D radar
It opens prompt triggering 3D radar to be scanned, realizes the acquisition of point cloud data.Simultaneously when not needing scanning, 3D radar is not opened
It opens, avoids unnecessary resource loss.
Compartment crowding detection method provided by the present application based on 3D radar scanning passes in and out passengers quantity by identification,
By calculating the number in each compartment, by the logical PIS system for being sent to next platform of data after identifying processing, pass through station
Platform CCTV display screen shows each compartment crowding situation, sets including the number in compartment number, compartment, and according to number
The crowding colouring information that threshold feedback is returned is set, the passenger that next station is waited selects according to the congestion state in compartment
Waiting station platform is selected, cabin space is rationally applied, improves train transport power.
Fig. 5 is the structural block diagram of the compartment passengers quantity detection device provided in this embodiment based on 3D radar scanning, ginseng
See Fig. 5, which includes obtaining module 501, processing module 502 and sending module 503, wherein
Module 501 is obtained, for the either objective compartment to train, the passenger obtained by 3D radar scanning is obtained and is working as
The point cloud data that preceding station is got on or off the bus by the car door in the target compartment, and two dimensional character number is obtained from the point cloud data
According to；
Processing module 502, for by identification model trained in advance marked from two dimensional character data to passenger into
The scanning element that row scanning is formed tracks each target two dimensional character data as target two dimensional character data, obtains target two dimension
The direction of travel of the corresponding passenger of characteristic；
Sending module 503, for entering the target carriage at the current station according to the direction of travel of each passenger statistics
First passengers quantity in compartment and the second passengers quantity for walking out the target compartment obtain train and sail out of a mesh when station
The existing passengers quantity in compartment is marked, according to first passengers quantity, second passengers quantity and the existing ridership
Amount calculates train and sails out of target passengers quantity when the current station in the target compartment, and the target passengers quantity is sent out
It is sent to next station；
Wherein, the identification model from two dimensional character data for marking the data for meeting humanoid feature, as right
Passenger is scanned the target two dimensional character data to be formed；What two dimensional character data referred to intercepting from point cloud data includes sweeping
The set of described point and the scanning element being parallel on the two-dimensional surface of target compartment car door.
Compartment passengers quantity detection device provided in this embodiment based on 3D radar scanning is mentioned suitable for above-described embodiment
The compartment crowding detection method based on 3D radar scanning supplied, details are not described herein.
A kind of compartment passengers quantity detection device based on 3D radar scanning is present embodiments provided, acquisition is mounted on compartment
The point cloud data of the passenger getting on/off of 3D radar scanning above car door, obtains two dimensional character data, then lead to from point cloud data
The corresponding target two dimensional character data of visitor of starting a work shift are marked from two dimensional character data after identification model trained in advance.It tracks each
Target two dimensional character data obtain the direction of travel of the passenger corresponding to it, and then are being worked as according to the direction of travel of each passenger statistics
The number that preceding station gets on the bus and gets off, when calculating train and sailing out of the station, passengers quantity in compartment sends passengers quantity
To next station, shown at next station.Before passenger reaches next station, the passengers quantity in each compartment of statistics is carried out
Display enables the passenger to clamp on to select platform according to the passengers quantity in each compartment, avoids blindly improving the fortune of train to vehicle
Loading capability.
The present embodiment additionally provides a kind of compartment passengers quantity detection system based on 3D radar scanning, including data processing
Unit and the 3D radar being arranged in above each compartment car door of train；
Every 3D radar connection Train Control and management system TCMS and data processing unit, data processing unit connection
PIS；
Wherein, to the either objective compartment of train, the 3D radar above target compartment car door is set and is being received
Start to scan at current station after the unlatching for the passenger that the car door in the target compartment is got on or off the bus prompt, opens 3D radar and sweep
The passenger to get on or off the bus at current station by the car door in the target compartment is retouched, point cloud data is obtained；
The data processing unit is used to execute the compartment crowding inspection described in any of the above item based on 3D radar scanning
Survey method.
3D radar is provided with above the car door in each compartment of train, 3D radar is for sweeping the passenger to get on or off the bus
It retouches to obtain point cloud data, data processing unit is to be integrated in 3D radar scanning system, for by described in the various embodiments described above
The method obtained point cloud data of processing 3D radar scanning and the functional module opened of control 3D radar, pass through data processing list
Member obtains the passengers quantity in each compartment, PIS system is sent by the passengers quantity in compartment and congestion state, at next station
It is shown, avoids the passenger at next station blindly to vehicle, promote train transport power.
Fig. 6 is the state flow diagram that 3D radar provided in this embodiment opens and closes, referring to Fig. 6, vehicle-mounted TCMS
Vehicle door status and location information can be sent in 3D radar scanning system.(i.e. train position is removing train main track operational process
When operation section other than terminus) in, after car door opening, 3D radar system is connected to Train door opening imformation, opens immediately
Scanning function, after closing of the door, TCMS sends location information and door closing information in 3D radar scanning system, closes
3D radar.
Methods, devices and systems provided in this embodiment have the advantage that (1) only from Vehicular door state and vehicle
Location information, so that it may which it is determined whether to enable crowding detection functions to be not necessarily to people without multisystem linkage and mass data interaction
Because the function such as automatic opening, automatic identification detection, automatic signal transmitting, the automatic display of the compartment CCTV crowding can be realized in intervention
Energy；(2) fusion for passing through 3D radar scanning and deep learning algorithm may make application scenarios root for multiplicity, and recognition result is more
Accurately；(3) crowding grade classification in compartment is more careful, include sit, be loose, more crowding, very crowded four grades,
More careful crowding display function can be provided for passenger, perfect PIS system display function can mention to a certain extent
The transport power of train is risen, the risk that passenger's large area is detained is reduced.
Fig. 7 is the structural block diagram for showing electronic equipment provided in this embodiment.
Referring to Fig. 7, the electronic equipment includes: processor (processor) 710, communication interface (Communications
Interface) 720, memory (memory) 730 and communication bus 740, wherein processor 710, communication interface 720, storage
Device 730 completes mutual communication by communication bus 740.Processor 710 can call the logical order in memory 730,
To execute following method: to the either objective compartment of train, obtaining the passenger obtained by 3D radar scanning and passed through at current station
The point cloud data that the car door in the target compartment is got on or off the bus is crossed, and obtains two dimensional character data from the point cloud data；Pass through
Trained identification model marks the scanning element for being scanned and being formed to passenger from two dimensional character data in advance, as target two
Dimensional feature data track each target two dimensional character data, obtain the direction of travel of the corresponding passenger of target two dimensional character data；
Enter first passengers quantity in the target compartment at the current station according to the direction of travel of each passenger statistics and walks out institute
Second passengers quantity in target compartment is stated, train is obtained and sails out of an existing passengers quantity when station in the target compartment,
It calculates train according to first passengers quantity, second passengers quantity and the existing passengers quantity and sails out of and described work as front truck
Target passengers quantity when standing in the target compartment sends the target passengers quantity, to show the mesh at next station
Mark passengers quantity；Wherein, the identification model is used to mark the data for meeting humanoid feature from two dimensional character data, as
The target two dimensional character data to be formed are scanned to passenger；What two dimensional character data referred to intercepting from point cloud data includes
The set of scanning element and the scanning element being parallel on the two-dimensional surface of target compartment car door.
In addition, the logical order in above-mentioned memory 730 can be realized by way of SFU software functional unit and conduct
Independent product when selling or using, can store in a computer readable storage medium.Based on this understanding, originally
Substantially the part of the part that contributes to existing technology or the technical solution can be in other words for the technical solution of invention
The form of software product embodies, which is stored in a storage medium, including some instructions to
So that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation of the present invention
The all or part of the steps of example the method.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. it is various
It can store the medium of program code.
The present embodiment provides a kind of non-transient computer readable storage mediums, are stored thereon with computer program, the calculating
Machine program is executed by processor following method: to the either objective compartment of train, obtaining the passenger obtained by 3D radar scanning
In the point cloud data that current station is got on or off the bus by the car door in the target compartment, and two dimension spy is obtained from the point cloud data
Levy data；The scanning for being scanned and being formed to passenger is marked from two dimensional character data by identification model trained in advance
Point tracks each target two dimensional character data as target two dimensional character data, obtains that target two dimensional character data are corresponding to be multiplied
The direction of travel of visitor；Enter first passenger in the target compartment at the current station according to the direction of travel of each passenger statistics
Quantity and the second passengers quantity for walking out the target compartment obtain train and sail out of when a station in the target compartment
There is passengers quantity, train is calculated according to first passengers quantity, second passengers quantity and the existing passengers quantity and is sailed
Target passengers quantity when from the current station in the target compartment sends the target passengers quantity, in next vehicle
It stands and shows the target passengers quantity；Wherein, the identification model meets humanoid spy for marking from two dimensional character data
The data of sign, as being scanned the target two dimensional character data to be formed to passenger；Two dimensional character data are referred to from a cloud number
What it is according to middle interception includes scanning element and the set of scanning element that is parallel on the two-dimensional surface of target compartment car door.
The present embodiment discloses a kind of computer program product, and the computer program product includes being stored in non-transient calculating
Computer program on machine readable storage medium storing program for executing, the computer program include program instruction, when described program instruction is calculated
When machine executes, computer is able to carry out method provided by above-mentioned each method embodiment, it may for example comprise: to any mesh of train
Compartment is marked, the point that the passenger obtained by 3D radar scanning gets on or off the bus at current station by the car door in the target compartment is obtained
Cloud data, and two dimensional character data are obtained from the point cloud data；By identification model trained in advance from two dimensional character number
The scanning element for being scanned and being formed to passenger is marked in, and as target two dimensional character data, it is special to track each target two dimension
Data are levied, the direction of travel of the corresponding passenger of target two dimensional character data is obtained；It is counted according to the direction of travel of each passenger in institute
First passengers quantity and second passengers quantity of walking out the target compartment of the current station into the target compartment are stated, is obtained
Train sails out of an existing passengers quantity when station in the target compartment, according to first passengers quantity, described second
Passengers quantity and the existing passengers quantity calculate train and sail out of target passenger when the current station in the target compartment
Quantity sends the target passengers quantity, to show the target passengers quantity at next station；Wherein, the identification model
It is two-dimentional as the target to be formed is scanned to passenger for marking the data for meeting humanoid feature from two dimensional character data
Characteristic；What two dimensional character data referred to intercepting from point cloud data includes scanning element and is parallel to target compartment vehicle
The set of scanning element on the two-dimensional surface of door.
The embodiments such as electronic equipment described above are only schematical, wherein it is described as illustrated by the separation member
Unit may or may not be physically separated, and component shown as a unit may or may not be object
Manage unit, it can it is in one place, or may be distributed over multiple network units.It can select according to the actual needs
Some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying wound
In the case where the labour for the property made, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above various embodiments is only to illustrate the technical solution of the embodiment of the present invention, rather than it is right
It is limited；Although the embodiment of the present invention is described in detail referring to foregoing embodiments, the ordinary skill of this field
Personnel are it is understood that it is still possible to modify the technical solutions described in the foregoing embodiments, or to part
Or all technical features are equivalently replaced；And these are modified or replaceed, it does not separate the essence of the corresponding technical solution
The range of each embodiment technical solution of the embodiment of the present invention.
Claims (10)
1. a kind of compartment crowding detection method based on 3D radar scanning characterized by comprising
To the either objective compartment of train, the passenger obtained by 3D radar scanning is obtained at current station by the target carriage
The point cloud data that the car door in compartment is got on or off the bus, and two dimensional character data are obtained from the point cloud data；
The scanning element for being scanned and being formed to passenger is marked from two dimensional character data by identification model trained in advance, is made
For target two dimensional character data, each target two dimensional character data are tracked, obtain the corresponding passenger's of target two dimensional character data
Direction of travel；
Enter first passengers quantity in the target compartment at the current station according to the direction of travel of each passenger statistics and walks
Second passengers quantity in the target compartment out obtains train and sails out of an existing ridership when station in the target compartment
Amount calculates train according to first passengers quantity, second passengers quantity and the existing passengers quantity and sails out of described work as
Target passengers quantity when preceding station in the target compartment sends the target passengers quantity, to show institute at next station
State target passengers quantity；
Wherein, the identification model from two dimensional character data for marking the data for meeting humanoid feature, as to passenger
It is scanned the target two dimensional character data to be formed；What two dimensional character data referred to intercepting from point cloud data includes scanning element
And it is parallel to the set of the scanning element on the two-dimensional surface of target compartment car door.
2. the compartment crowding detection method according to claim 1 based on 3D radar scanning, which is characterized in that the knowledge
The training of other model includes:
The point cloud data that is scanned in advance to the passenger to get on or off the bus is obtained, by two-dimentional peak-seeking by the point cloud data that obtains
Obtained two dimensional character data, and the data to be formed will be scanned to passenger in each two dimensional character data and be marked,
Two dimensional character number using the two dimensional character data before being marked as the input parameter of deep learning, after being marked
According to the desired output as deep learning, using the model trained by deep learning as the identification model.
3. the compartment crowding detection method according to claim 1 based on 3D radar scanning, which is characterized in that described to chase after
The each target two dimensional character data of track, obtain the direction of travel of the corresponding passenger of target two dimensional character data, comprising:
To each target two dimensional character data, obtains the target two dimensional character data and scanned in upper primary scanning or next time
The first position in obtained point cloud data, and obtain the point cloud number that the target two dimensional character data present scan obtains
The second position in determines the target two dimensional character according to first position and second position
The direction of travel of the corresponding passenger of data.
4. the compartment crowding detection method according to claim 1 based on 3D radar scanning, which is characterized in that the hair
The target passengers quantity is sent, to show the target passengers quantity at next station, comprising:
The mapping relations between the passengers quantity in the preset compartment degree of crowding and compartment are obtained, are closed according to the mapping
System and the target passengers quantity determine the degree of crowding in target compartment when train sails out of the current station, as target
The target passengers quantity and the target degree of crowding are sent passenger information system PIS by the degree of crowding, by next
The display equipment at station shows the target passengers quantity and the target degree of crowding；
Wherein, the target degree of crowding is indicated by preset color corresponding with the target degree of crowding.
5. the compartment crowding detection method according to claim 4 based on 3D radar scanning, which is characterized in that described to reflect
The relationship of penetrating includes:
When the passengers quantity in compartment is less than the first preset quantity, the compartment degree of crowding is the seat compartment Nei You；
When the passengers quantity in compartment is greater than or equal to first preset quantity and when less than the second preset quantity, compartment is crowded
Degree be compartment in without seat but loosely；
When the passengers quantity in compartment is greater than or equal to second preset quantity and is less than third preset quantity, compartment is crowded
Degree is more crowded in compartment；
When the passengers quantity in compartment is greater than or equal to the third preset quantity, the compartment degree of crowding is to gather around very much in compartment
It squeezes；
Wherein, first preset quantity is equal to the amount of seats configured in compartment.
6. the compartment crowding detection method according to claim 1 based on 3D radar scanning, which is characterized in that it is described from
Two dimensional character data are obtained in the point cloud data, comprising:
It is intercepted from the point cloud data by two-dimentional peak-seeking and contains the summit formed by scanning element and be parallel to the target
The two-dimensional surface of compartment car door, using the set of the scanning element on the two-dimensional surface of interception as two dimensional character data.
7. the compartment crowding detection method according to claim 1 based on 3D radar scanning, which is characterized in that also wrap
It includes:
After receiving train position and car door opening state by Train Control and management system TCMS transmission, if judging train
It is run in main track and car door is in the open state, then sent to 3D radar and open prompt, existed so that 3D radar starts scanning
Current station obtains the point cloud data by the passenger that the car door in the target compartment is got on or off the bus.
8. the compartment crowding detection method according to claim 7 based on 3D radar scanning, which is characterized in that also wrap
It includes:
After receiving train position and car door opening state by Train Control and management system TCMS transmission, if judging train
It is not opened positioned at terminus or car door, does not then send the unlatching prompt to 3D radar.
9. a kind of compartment crowding detection device based on 3D radar scanning characterized by comprising
Module is obtained, for the either objective compartment to train, obtains the passenger obtained by 3D radar scanning at current station
By the point cloud data that the car door in the target compartment is got on or off the bus, and two dimensional character data are obtained from the point cloud data；
Processing module is scanned shape to passenger for marking from two dimensional character data by identification model trained in advance
At scanning element track each target two dimensional character data as target two dimensional character data, obtain target two dimensional character data
The direction of travel of corresponding passenger；
Sending module, for entering the first of the target compartment at the current station according to the direction of travel of each passenger statistics
Passengers quantity and the second passengers quantity for walking out the target compartment obtain train and sail out of when a station in the target compartment
Existing passengers quantity, column are calculated according to first passengers quantity, second passengers quantity and the existing passengers quantity
Vehicle sails out of target passengers quantity when the current station in the target compartment, sends the target passengers quantity to next
Station；
Wherein, the identification model from two dimensional character data for marking the data for meeting humanoid feature, as to passenger
It is scanned the target two dimensional character data to be formed；What two dimensional character data referred to intercepting from point cloud data includes scanning element
And it is parallel to the set of the scanning element on the two-dimensional surface of target compartment car door.
10. a kind of compartment crowding detection system based on 3D radar scanning, which is characterized in that including data processing unit and set
Set the 3D radar above each compartment car door of train；
Every 3D radar connection Train Control and management system TCMS and data processing unit, data processing unit connect PIS；
Wherein, to the either objective compartment of train, the 3D radar above target compartment car door is set and is receiving beginning
It scans at current station after the unlatching for the passenger that the car door in the target compartment is got on or off the bus prompt, opens 3D radar scanning and exist
The passenger that current station is got on or off the bus by the car door in the target compartment, obtains point cloud data；
The data processing unit requires the described in any item compartment crowdings based on 3D radar scanning of 1-8 for perform claim
Detection method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910666603.8A CN110376585B (en) | 2019-07-23 | 2019-07-23 | Carriage congestion degree detection method, device and system based on 3D radar scanning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910666603.8A CN110376585B (en) | 2019-07-23 | 2019-07-23 | Carriage congestion degree detection method, device and system based on 3D radar scanning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110376585A true CN110376585A (en) | 2019-10-25 |
| CN110376585B CN110376585B (en) | 2022-02-15 |
Family
ID=68255188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910666603.8A Active CN110376585B (en) | 2019-07-23 | 2019-07-23 | Carriage congestion degree detection method, device and system based on 3D radar scanning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110376585B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561971A (en) * | 2020-12-16 | 2021-03-26 | 珠海格力电器股份有限公司 | People flow statistical method, device, equipment and storage medium |
| CN114782469A (en) * | 2022-06-16 | 2022-07-22 | 西南交通大学 | Method and device for identifying degree of congestion of public traffic, electronic device and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268506A (en) * | 2014-09-15 | 2015-01-07 | 郑州天迈科技股份有限公司 | Passenger flow counting detection method based on depth images |
| CN104517040A (en) * | 2014-12-31 | 2015-04-15 | 青岛海信网络科技股份有限公司 | Method for calculating in-carriage congestion degree of public traffic vehicle based on IC card data |
| CN106022460A (en) * | 2016-05-25 | 2016-10-12 | 重庆市勘测院 | Crowd density real-time monitoring method based on laser radar |
| CN106846297A (en) * | 2016-12-21 | 2017-06-13 | 深圳市镭神智能系统有限公司 | Pedestrian's flow quantity detecting system and method based on laser radar |
| CN206696862U (en) * | 2017-03-22 | 2017-12-01 | 西南交通大学 | Subway number identifies and delivery system |
| CN108269219A (en) * | 2017-12-25 | 2018-07-10 | 河南辉煌软件有限公司 | A kind of rail transit cars passengers quantity counting and displaying method |
| CN108710818A (en) * | 2018-03-09 | 2018-10-26 | 唐义 | A kind of real-time monitoring and statistics system and method for number based on three-dimensional laser radar |
-
2019
- 2019-07-23 CN CN201910666603.8A patent/CN110376585B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268506A (en) * | 2014-09-15 | 2015-01-07 | 郑州天迈科技股份有限公司 | Passenger flow counting detection method based on depth images |
| CN104517040A (en) * | 2014-12-31 | 2015-04-15 | 青岛海信网络科技股份有限公司 | Method for calculating in-carriage congestion degree of public traffic vehicle based on IC card data |
| CN106022460A (en) * | 2016-05-25 | 2016-10-12 | 重庆市勘测院 | Crowd density real-time monitoring method based on laser radar |
| CN106846297A (en) * | 2016-12-21 | 2017-06-13 | 深圳市镭神智能系统有限公司 | Pedestrian's flow quantity detecting system and method based on laser radar |
| CN206696862U (en) * | 2017-03-22 | 2017-12-01 | 西南交通大学 | Subway number identifies and delivery system |
| CN108269219A (en) * | 2017-12-25 | 2018-07-10 | 河南辉煌软件有限公司 | A kind of rail transit cars passengers quantity counting and displaying method |
| CN108710818A (en) * | 2018-03-09 | 2018-10-26 | 唐义 | A kind of real-time monitoring and statistics system and method for number based on three-dimensional laser radar |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561971A (en) * | 2020-12-16 | 2021-03-26 | 珠海格力电器股份有限公司 | People flow statistical method, device, equipment and storage medium |
| CN112561971B (en) * | 2020-12-16 | 2024-08-09 | 珠海格力电器股份有限公司 | People flow statistics method, device, equipment and storage medium |
| CN114782469A (en) * | 2022-06-16 | 2022-07-22 | 西南交通大学 | Method and device for identifying degree of congestion of public traffic, electronic device and storage medium |
| CN114782469B (en) * | 2022-06-16 | 2022-08-19 | 西南交通大学 | Method and device for identifying degree of congestion of public traffic, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110376585B (en) | 2022-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110362077B (en) | Unmanned vehicle emergency hedge decision making system, method and medium | |
| CN105678267A (en) | Scene recognition method and device | |
| CN103824066B (en) | A kind of licence plate recognition method based on video flowing | |
| CN109808700A (en) | System and method for mapping road interfering object in autonomous vehicle | |
| CN110487562A (en) | One kind being used for unpiloted road-holding ability detection system and method | |
| US10990820B2 (en) | Heterogeneous convolutional neural network for multi-problem solving | |
| CN109190444A (en) | A kind of implementation method of the lane in which the drivers should pay fees vehicle feature recognition system based on video | |
| CN110516518A (en) | A kind of illegal manned detection method of non-motor vehicle, device and electronic equipment | |
| CN105216797A (en) | Method of overtaking and system | |
| CN113688805B (en) | Unmanned aerial vehicle-based unlicensed muck vehicle identification method and system | |
| CN108765945A (en) | A kind of vehicle peccancy recognition methods and system | |
| CN107145819A (en) | A kind of bus crowding determines method and apparatus | |
| CN110376585A (en) | Compartment crowding detection method and device, system based on 3D radar scanning | |
| CN110490103A (en) | Track traffic for passenger flow Density Detection method and device | |
| CN113255444A (en) | Training method of image recognition model, image recognition method and device | |
| CN112183206B (en) | Traffic participant positioning method and system based on road side monocular camera | |
| CN117037084A (en) | Method and device for identifying abnormal behavior of toll station vehicle based on track reconstruction | |
| CN111079488B (en) | Deep learning-based bus passenger flow detection system and method | |
| CN110154896A (en) | A kind of method and equipment detecting barrier | |
| CN110696828B (en) | Forward target selection method and device and vehicle-mounted equipment | |
| CN111275008B (en) | Method and device for detecting abnormality of target vehicle, storage medium and electronic device | |
| CN113705549B (en) | Road rescue work node determination method and device and related equipment | |
| CN114972731A (en) | Traffic light detection and identification method and device, moving tool and storage medium | |
| CN115616557A (en) | Vehicle visibility detection method and system | |
| CN112164205A (en) | Method and device for sending alarm information |
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 |