CN108648495A - A kind of method and system of the intelligence real-time display bus degree of crowding - Google Patents
A kind of method and system of the intelligence real-time display bus degree of crowding Download PDFInfo
- Publication number
- CN108648495A CN108648495A CN201810583796.6A CN201810583796A CN108648495A CN 108648495 A CN108648495 A CN 108648495A CN 201810583796 A CN201810583796 A CN 201810583796A CN 108648495 A CN108648495 A CN 108648495A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- bus
- crowding
- degree
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/123—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams
Abstract
The invention discloses a kind of method and system of the intelligent real-time display bus degree of crowding, the method mainly captures real-time pictures in compartment by vehicle-mounted camera, it is uploaded to cloud server, go out number in compartment by being based on the trained model intelligent recognition of deep learning image recognition, to generate the real-time degree of crowding of current bus, it is sent in Message Display Terminal by data distribution module synchronization.Intelligent real-time display bus degree of crowding system involved by this method is mainly made of vehicle GPS locating module, data memory module, vehicle-mounted camera, data transmission module, cloud server and Message Display Terminal, this method uses deep learning image recognition algorithm, required appliance arrangement is less, degree of intelligence is high, recognition speed is fast, and bus scheduling central dispatching and passenger can greatly be facilitated to go on a journey.
Description
Technical field
The present invention relates to technical field of intelligent traffic, and in particular to a kind of side of the intelligence real-time display bus degree of crowding
Method and system.
Background technology
In today of urbanization fast development, traffic jam issue annoyings most cities, and various regions are all greatly developing
Public transport is as the important measure for alleviating congestion.However current bus run and service etc. there is it is many not
Foot, due to lacking real-time bus passenger flow information acquiring technology, most of public transport company is taken based on the warp of history passenger flow information
Scheduling method is tested, thereby determines that the departure frequency of bus, there is very big blindness, not for the Trip distribution during operation
Match, the case where public transport passenger inside the vehicle is crowded, peak period especially on and off duty, vehicle occupant suffers people frequent occurrence, traditional public transport
Scheduling mode far can not meet the needs of citizen comfortably go on a journey.Simultaneously as passenger can not understand public transport vehicle-mounted passenger traffic in real time
Capable situation, cause often to occur previous vehicle it is extremely crowded in the case of, passenger still up squeezes, and and then next vehicle
It is that passenger seldom even the case where empty wagons occurs.Therefore, it can be provided by the intelligent real-time display bus degree of crowding, one
Congestion information is referred to for bus scheduling center, and two can in real time show in platform electronic display and hand-held intelligent terminal
Show the public bus network vehicle congested conditions, reasonably selects vehicle ride for passenger, improve riding comfort.
Invention content
In view of the deficiencies of the prior art, it is an object of the present invention to provide a kind of intelligent real-time display bus degree of crowding
Method, the method judges the degree of crowding of bus using deep learning image recognition algorithm by less equipment,
So that bus scheduling center and passenger refer to.
Another object of the present invention is to provide a kind of systems of the intelligent real-time display bus degree of crowding.
The purpose of the present invention can be achieved through the following technical solutions：
A kind of method of the intelligence real-time display bus degree of crowding, the described method comprises the following steps：
S1, vehicle-mounted camera obtain compartment real-time pictures：When bus closed door, left for from a website next
When a website, vehicle GPS locating module sends out signal and is transmitted to vehicle-mounted camera, and the camera positioned at compartment front and rear part is opened
Begin shooting, shooting duration of video 20s, rear camera be automatically stopped work, and by data transmission module by the video of shooting, vehicle
Location information and vehicle characteristic information are uploaded to cloud server；
S2, the cloud server intelligent recognition compartment degree of crowding：After cloud server receives video data, video is carried out
Frame is divided, and reads video image frame later, carries out image preprocessing, and the image handled well is incoming based on the knowledge of deep learning image
In the other trained model of algorithm, model intelligent recognition goes out the compartment degree of crowding, and by congestion information and vehicle characteristic information,
Location information is transmitted to data distribution module；
S3, distribution bus congestion information：Data distribution module is by bus positional information, vehicle characteristic information and is somebody's turn to do
The degree of crowding of bus is distributed to Message Display Terminal.
Another object of the present invention can be achieved through the following technical solutions：
A kind of system of the intelligence real-time display bus degree of crowding, the system comprises vehicle GPS locating module, data
Memory module, vehicle-mounted camera, data transmission module, cloud server, data distribution module and Message Display Terminal, vehicle
GPS positioning module determines bus positional information, and vehicle-mounted camera enabling signal is conveyed after vehicle leaves website, makes vehicle-mounted
Camera works, while bus positional information is transferred to data memory module；Vehicle-mounted camera is located at compartment front and rear part, often
It is automatically stopped work after task 20s, and by the transmission of video of shooting to data memory module；It is stored in data memory module
Vehicle line information and vehicle characteristic information, and after receiving the data that vehicle GPS locating module and vehicle-mounted camera transmit, pass through
Data transmission module can locate all data transmissions to cloud server, cloud server the data received automatically
Reason is to judge the degree of crowding in compartment, and by data distribution module by bus positional information, vehicle characteristic information
And the degree of crowding of the bus is distributed to Message Display Terminal.
Further, described information display terminal includes bus scheduling center, the bus vehicle line residue website
Electronic display and hand-held intelligent terminal (include but not limited to smart mobile phone, tablet computer etc.).
Further, the cloud server includes degree of crowding intelligent identifying system, the degree of crowding intelligent recognition
After system receives video data, the video image frame in the video of shooting is read, is input an image into after image preprocessing
In trained deep learning image recognition model, deep learning image recognition model automatically identifies in the image in compartment
The degree of crowding, and the degree of crowding of bus positional information, vehicle characteristic information and the bus is transmitted to data distribution mould
Block.
Further, the deep learning image recognition model is based on deep learning image recognition algorithm, using convolution god
Through network, data set is the bus compartment internal image marked, and label is that not crowded, moderate is crowded and crowded three kinds of severe
Situation is divided into training set, verification collection and test set, is trained to convolutional neural networks with big data, adjusting parameter, final to instruct
Perfect the deep learning image recognition model that discrimination reaches 99.7%.
Further, the display information of described information display terminal is the distance or station that each circuit bus most reaches soon
Number of units, the font representation that is respectively adopted different colours is not crowded, moderate is crowded and the crowded three kinds of situations of severe.
Compared with prior art, the present invention having the following advantages that and advantageous effect：
The system of the intelligence real-time display bus degree of crowding of the present invention, occupancy bus inner space is less, institute
It needs equipment less, can be integrated in a device, be mounted in the middle part of compartment, and use deep learning image recognition algorithm,
With convolutional neural networks training pattern, recognition speed is fast, and discrimination is high, can be that bus scheduling center and passenger's trip carry
For effectively referring to.
Description of the drawings
Fig. 1 is the structure diagram of intelligence real-time display bus degree of crowding system of the embodiment of the present invention.
Fig. 2 is the structure chart of bottleneck residual error study modules in the embodiment of the present invention.
Fig. 3 is the residual error neural network complete structure figure of deep learning image recognition model of the embodiment of the present invention.
Specific implementation mode
Present invention will now be described in further detail with reference to the embodiments and the accompanying drawings, but embodiments of the present invention are unlimited
In this.
Embodiment：
Present embodiments provide a kind of system of the intelligent real-time display bus degree of crowding, the structure diagram of the system
As shown in Figure 1, including vehicle GPS locating module, data memory module, vehicle-mounted camera, data transmission module, cloud service
Device, data distribution module and Message Display Terminal, vehicle GPS locating module determines bus positional information, and is left in vehicle
Vehicle-mounted camera enabling signal is conveyed after website, so that vehicle-mounted camera is worked, while bus positional information is transferred to data
Memory module；Vehicle-mounted camera is located at compartment front and rear part, work is automatically stopped after every task 20s, and the video of shooting is passed
It is defeated to arrive data memory module；It is stored with vehicle line information and vehicle characteristic information in data memory module, and receives vehicle GPS
After the data that locating module and vehicle-mounted camera transmit, by data transmission module by all data transmissions to cloud server,
Cloud server can automatically process the data received to judge the degree of crowding in compartment, and pass through data
The degree of crowding of bus positional information, vehicle characteristic information and the bus is distributed to Message Display Terminal by distribution module.
Wherein, data distribution module receives the data that cloud server transmission comes, according to vehicle characteristic information, public transport parking stall
Confidence breath determines circuit residing for bus and remaining circuit site information, and the degree of crowding of the vehicle is distributed to bus scheduling
On center, circuit residue site electrical display screen and hand-held intelligent terminal, bus scheduling center is according to vehicle congestion information
Bus and change departure frequency are sent to take whether to add, and passenger can be according to bus stop electronic display and hand-held intelligence
Vehicle congestion information that can be in terminal determines the choice for traveling of oneself.
Described information display terminal (i.e. bus scheduling center terminal, platform electronic display and hand-held intelligent terminal)
Platform electronic display screen displays scheme uses LED screen, and display information is each circuit vehicle fastest to up to distance or platform number, word
Body indicates that severe is crowded, moderate is crowded and not crowded three kinds of situations respectively using three kinds of colors of red, yellow, and green, can either show line
Road information can intuitively reflect the crowding of bus again.
The present embodiment additionally provides a kind of method of the intelligent real-time display bus degree of crowding, and the method includes following
Step：
S1, vehicle-mounted camera obtain compartment real-time pictures：When bus closed door, left for from a website next
When a website, vehicle GPS locating module sends out signal and is transmitted to vehicle-mounted camera, and the camera positioned at compartment front and rear part is opened
Begin shooting, shooting duration of video 20s, rear camera be automatically stopped work, and by data transmission module by the video of shooting, vehicle
Location information and vehicle characteristic information are uploaded to cloud server；
S2, the cloud server intelligent recognition compartment degree of crowding：After cloud server receives video data, video is carried out
Frame is divided, and reads video image frame later, carries out image preprocessing, and the image handled well is incoming based on the knowledge of deep learning image
In the other trained model of algorithm, model intelligent recognition goes out the compartment degree of crowding, and by congestion information and vehicle characteristic information,
Location information is transmitted to data distribution module；
Specifically, after cloud server receives video data, video frame segmentation is carried out by the frequency of 4s/ frames, to ensure
It can clearly reflect interior situation, read video image frame later, carry out image preprocessing, the image handled well is passed to
Based in the trained model of deep learning image recognition algorithm, intelligent recognition goes out the compartment degree of crowding, wherein image preprocessing
It is specific as follows：
Picture frame is denoted as pk, k expression image frame numbers, in a video sequence, k is assigned successively：1、2、
3 ... sampled images size is indicated with M × N, is taken weighted mean method to carry out gray proces collected video image frame and is obtained
To grey-level image frame pk, as shown in formula (1)：
pk(x, y)=0.2989Rk(x,y)+0.5870Gk(x,y)+0.1140Bk(x,y) (1)
Wherein pk(x, y) indicates gray level image pkGray value at coordinate (x, y), Rk(x,y)、Gk(x,y)、Bk(x,y)
Video image frame p is indicated respectivelykRed color component value, green component values at coordinate (x, y) and blue color component value.
Video image color space conversion is carried out later, and the color of the video image frame of acquisition is shown in the RGB model spaces
In, perception of the HSV color models than RGB color model closer to us to color, therefore color video frequency image is transformed into HSV
Hsv color spatial image is obtained in color space.
It is wherein described to be based on the trained model of deep learning image recognition algorithm, i.e. deep learning image recognition model,
Based on deep learning image recognition algorithm, using convolutional neural networks, data set is the bus compartment internal image marked,
Label is that not crowded, moderate is crowded and the crowded three kinds of situations of severe, is divided into training set, verification collection and test set, with big data pair
Convolutional neural networks are trained, adjusting parameter, finally train the deep learning image recognition mould that discrimination reaches 99.7%
Type.
For the deep learning image recognition model that the present embodiment uses for residual error neural network (ResNet), which can
Easily to reach layers up to a hundred or even thousands of layers, and training can be completed within the acceptable time, to greatly improve figure
As the accuracy of identification.Residual error neural network is mainly stacked by multiple residual error study modules, which is
Bottleneck residual error study modules, the modular structure are illustrated in fig. 2 shown below, and carry out simplifying expression.
The non-linear partial of Bottleneck structures haves three layers altogether, includes the volume of two 1 × 1 one k × k of convolution sum
It is long-pending, k=1 in figure.Assuming that the data that one dimension of input is 256, first 1 × 1 convolution have primarily served the effect of dimensionality reduction,
Also realize the information fusion across channel to a certain extent simultaneously, input dimension is dropped to 64 by it, and second 1 × 1 volume
Product primarily serves a liter effect for dimension, and output dimension is risen back 256 by it again.Pass through the two 1 × 1 convolution so that middle k ×
The convolution of k inputs, and output dimension has all been reduced to 64 by the 256 of script, greatly reduces number of parameters, while also adding module
Depth, this bottleneck modules are the key that allow residual error network that can reach even thousands of layers of layers up to a hundred.
The core network architecture of deep learning image recognition model is such as in intelligent real-time display bus degree of crowding system
Shown in lower Fig. 3, by pretreated image, it is assumed that be 32 × 32 × 3 image data, by 3 × 3 convolution kernel, convolution
The number of core is 16, the data of output 32 × 32 × 16, then passes through 6n layers of convolution, since each residual error study module has two
Layer convolution, so a shared 3n residual error study module.It is that an overall situation is averaged pond after this is by 6n layers convolution
Layer then operates for dropout, is finally full articulamentum and softmax, m is zooming parameter.
Shown in the mathematic(al) representation of convolutional layer such as formula (2)：
Wherein：K(l)={ (u, v) ∈ N2|0<u<Kx,0<v<Ky}；KxAnd KyIt is the width and height of convolution kernel respectively；L tables
Show current layer number；It is the biasing of convolutional layer jth characteristic pattern epineural member；Mj(l-1)(u,v)It is and convolutional layer jth characteristic pattern
The set of the preceding layer characteristic pattern of opening relationships.
Pond layer simulation complex cell is that primary visual signature is screened to and is combined into more advanced and abstract vision spy
The process of sign is realized by sampling in a network, shown in mathematic(al) representation such as formula (3)：
Wherein down indicates that the function of maximum value sampling, this layer of operation do not include the weights that can learn and threshold value.
Full articulamentum can enhance the non-linear mapping capability of network, while limit the size of network size, be added one
Full articulamentum, each neuron of this layer and all neurons of preceding layer interconnect, and are not connected between same layer neuron,
Shown in mathematic(al) representation such as formula (4)：
Wherein：N is the neuron number of preceding layer；L indicates current layer number；It is this layer of neuron j and preceding layer nerve
The bonding strength of first i；It is the biasing of this layer of neuron j；F () indicates activation primitive.
S3, distribution bus congestion information：Data distribution module is by bus positional information, vehicle characteristic information and is somebody's turn to do
The degree of crowding of bus is distributed to Message Display Terminal.
The above, patent preferred embodiment only of the present invention, but the protection domain of patent of the present invention is not limited to
This, any one skilled in the art is in the range disclosed in patent of the present invention, according to the skill of patent of the present invention
Art scheme and its patent of invention design are subject to equivalent substitution or change, belong to the protection domain of patent of the present invention.
Claims (6)
1. a kind of method of the intelligence real-time display bus degree of crowding, which is characterized in that the described method comprises the following steps：
S1, vehicle-mounted camera obtain compartment real-time pictures：When bus closed door, next station is left for from a website
When point, vehicle GPS locating module sends out signal and is transmitted to vehicle-mounted camera, and the camera positioned at compartment front and rear part starts to clap
Take the photograph, shooting duration of video 20s, rear camera be automatically stopped work, and by data transmission module by the video of shooting, vehicle position
Confidence ceases and vehicle characteristic information is uploaded to cloud server；
S2, the cloud server intelligent recognition compartment degree of crowding：After cloud server receives video data, video frame point is carried out
It cuts, reads video image frame later, carry out image preprocessing, the image handled well is incoming based on the calculation of deep learning image recognition
In the trained model of method, model intelligent recognition goes out the compartment degree of crowding, and by congestion information and vehicle characteristic information, position
Information is transmitted to data distribution module；
S3, distribution bus congestion information：Data distribution module is by bus positional information, vehicle characteristic information and the public transport
The degree of crowding of vehicle is distributed to Message Display Terminal.
2. a kind of system of the intelligence real-time display bus degree of crowding, it is characterised in that：The system comprises vehicle GPS positioning
Module, data memory module, vehicle-mounted camera, data transmission module, cloud server, data distribution module and presentation of information are whole
End, vehicle GPS locating module determines bus positional information, and conveys vehicle-mounted camera to start letter after vehicle leaves website
Number, so that vehicle-mounted camera is worked, while bus positional information is transferred to data memory module；Vehicle-mounted camera is located at compartment
Front and rear part, per task 20s after be automatically stopped work, and by the transmission of video of shooting to data memory module；Data store mould
It is stored with vehicle line information and vehicle characteristic information in block, and receives vehicle GPS locating module and vehicle-mounted camera transmits
After data, by data transmission module by all data transmissions to cloud server, cloud server can be to the number that receives
According to being automatically processed to judge the degree of crowding in compartment, and by data distribution module by bus positional information,
The degree of crowding of vehicle characteristic information and the bus is distributed to Message Display Terminal.
3. a kind of system of intelligent real-time display bus degree of crowding according to claim 2, it is characterised in that：It is described
Message Display Terminal include bus scheduling center, the bus vehicle line residue website electronic display and hand-held intelligent
Terminal.
4. a kind of system of intelligent real-time display bus degree of crowding according to claim 2, it is characterised in that：It is described
Cloud server includes degree of crowding intelligent identifying system, after the degree of crowding intelligent identifying system receives video data,
The video image frame in the video of shooting is read, trained deep learning image is input an image into after image preprocessing and is known
In other model, deep learning image recognition model automatically identifies the degree of crowding in the image in compartment, and by public transport parking stall
The degree of crowding of confidence breath, vehicle characteristic information and the bus is transmitted to data distribution module.
5. a kind of system of intelligent real-time display bus degree of crowding according to claim 2, it is characterised in that：It is described
Deep learning image recognition model is based on deep learning image recognition algorithm, and using convolutional neural networks, data set is to mark
Bus compartment internal image, label is that not crowded, moderate is crowded and the crowded three kinds of situations of severe, is divided into training set, verification
Collection and test set, are trained convolutional neural networks with big data, adjusting parameter, finally train discrimination and reach 99.7%
Deep learning image recognition model.
6. a kind of system of intelligent real-time display bus degree of crowding according to claim 2, it is characterised in that：It is described
The display information of Message Display Terminal is the distance or platform number that each circuit bus most reaches soon, and different colours are respectively adopted
Font representation is crowded, moderate is crowded and the crowded three kinds of situations of severe.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810583796.6A CN108648495A (en) | 2018-06-08 | 2018-06-08 | A kind of method and system of the intelligence real-time display bus degree of crowding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810583796.6A CN108648495A (en) | 2018-06-08 | 2018-06-08 | A kind of method and system of the intelligence real-time display bus degree of crowding |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108648495A true CN108648495A (en) | 2018-10-12 |
Family
ID=63751808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810583796.6A Pending CN108648495A (en) | 2018-06-08 | 2018-06-08 | A kind of method and system of the intelligence real-time display bus degree of crowding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108648495A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109684952A (en) * | 2018-12-13 | 2019-04-26 | 北京航空航天大学 | A kind of Passenger's distribution detection system and method suitable for high-speed rail train intelligent air condition |
| CN110335460A (en) * | 2019-05-08 | 2019-10-15 | 上海电机学院 | A kind of public traffic information intelligent interactive system and method |
| CN110458056A (en) * | 2019-07-29 | 2019-11-15 | 交控科技股份有限公司 | Compartment crowding detection method and device |
| CN110458080A (en) * | 2019-08-05 | 2019-11-15 | 西藏宁算科技集团有限公司 | The pre-judging method and system of front pit-hole in a kind of running car |
| CN110782692A (en) * | 2019-10-31 | 2020-02-11 | 青岛海信网络科技股份有限公司 | Signal lamp fault detection method and system |
| CN110852155A (en) * | 2019-09-29 | 2020-02-28 | 深圳市深网视界科技有限公司 | Method, system, device and storage medium for detecting crowdedness of bus passengers |
| CN111291766A (en) * | 2018-12-06 | 2020-06-16 | 乐金信世股份有限公司 | Image recognition method and server using deep learning |
| CN112241688A (en) * | 2020-09-24 | 2021-01-19 | 厦门卫星定位应用股份有限公司 | Carriage congestion degree detection method and system |
| CN112396587A (en) * | 2020-11-20 | 2021-02-23 | 重庆大学 | Method for detecting crowding degree in bus compartment based on cooperative training and density map |
| CN112904775A (en) * | 2021-01-20 | 2021-06-04 | 深圳市蓝泰源信息技术股份有限公司 | Bus-mounted intelligent terminal |
| CN113291703A (en) * | 2020-08-31 | 2021-08-24 | 湖南长天自控工程有限公司 | Discharge opening blockage detection method and device |
| CN115273480A (en) * | 2022-09-28 | 2022-11-01 | 南京智慧交通信息股份有限公司 | Traffic early warning system and method based on big data analysis |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090112452A1 (en) * | 2007-10-25 | 2009-04-30 | Gm Global Technology Operations, Inc. | Vehicle navigation system with real time traffic image display |
| CN102333234A (en) * | 2011-10-28 | 2012-01-25 | 清华大学 | Binocular stereo video state information monitoring method and device |
| CN102724390A (en) * | 2011-03-29 | 2012-10-10 | 赵山山 | Method for detecting crowding degree in train compartment and system for guiding stream of people |
| CN104091179A (en) * | 2014-07-01 | 2014-10-08 | 北京工业大学 | Intelligent blumeria graminis spore picture identification method |
| CN104751187A (en) * | 2015-04-14 | 2015-07-01 | 山西科达自控股份有限公司 | Automatic meter-reading image recognition method |
| CN105654729A (en) * | 2016-03-28 | 2016-06-08 | 南京邮电大学 | Short-term traffic flow prediction method based on convolutional neural network |
| CN105740894A (en) * | 2016-01-28 | 2016-07-06 | 北京航空航天大学 | Semantic annotation method for hyperspectral remote sensing image |
| CN106372402A (en) * | 2016-08-30 | 2017-02-01 | 中国石油大学(华东) | Parallelization method of convolutional neural networks in fuzzy region under big-data environment |
| CN205983830U (en) * | 2016-05-16 | 2017-02-22 | 李政达 | Intelligence is mutual system for public transit information processing |
| CN106485923A (en) * | 2016-12-20 | 2017-03-08 | 武汉理工大学 | A kind of public transport crowding real-time status acquisition method and device |
| CN107145819A (en) * | 2017-03-14 | 2017-09-08 | 浙江宇视科技有限公司 | A kind of bus crowding determines method and apparatus |
| CN107239576A (en) * | 2017-06-30 | 2017-10-10 | 青岛海澄知识产权事务有限公司 | A kind of bus degree of crowding method for visualizing |
| CN108073886A (en) * | 2016-11-18 | 2018-05-25 | 南京行者易智能交通科技有限公司 | A kind of passenger on public transport amount evaluation method based on machine learning |
-
2018
- 2018-06-08 CN CN201810583796.6A patent/CN108648495A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090112452A1 (en) * | 2007-10-25 | 2009-04-30 | Gm Global Technology Operations, Inc. | Vehicle navigation system with real time traffic image display |
| CN102724390A (en) * | 2011-03-29 | 2012-10-10 | 赵山山 | Method for detecting crowding degree in train compartment and system for guiding stream of people |
| CN102333234A (en) * | 2011-10-28 | 2012-01-25 | 清华大学 | Binocular stereo video state information monitoring method and device |
| CN104091179A (en) * | 2014-07-01 | 2014-10-08 | 北京工业大学 | Intelligent blumeria graminis spore picture identification method |
| CN104751187A (en) * | 2015-04-14 | 2015-07-01 | 山西科达自控股份有限公司 | Automatic meter-reading image recognition method |
| CN105740894A (en) * | 2016-01-28 | 2016-07-06 | 北京航空航天大学 | Semantic annotation method for hyperspectral remote sensing image |
| CN105654729A (en) * | 2016-03-28 | 2016-06-08 | 南京邮电大学 | Short-term traffic flow prediction method based on convolutional neural network |
| CN205983830U (en) * | 2016-05-16 | 2017-02-22 | 李政达 | Intelligence is mutual system for public transit information processing |
| CN106372402A (en) * | 2016-08-30 | 2017-02-01 | 中国石油大学(华东) | Parallelization method of convolutional neural networks in fuzzy region under big-data environment |
| CN108073886A (en) * | 2016-11-18 | 2018-05-25 | 南京行者易智能交通科技有限公司 | A kind of passenger on public transport amount evaluation method based on machine learning |
| CN106485923A (en) * | 2016-12-20 | 2017-03-08 | 武汉理工大学 | A kind of public transport crowding real-time status acquisition method and device |
| CN107145819A (en) * | 2017-03-14 | 2017-09-08 | 浙江宇视科技有限公司 | A kind of bus crowding determines method and apparatus |
| CN107239576A (en) * | 2017-06-30 | 2017-10-10 | 青岛海澄知识产权事务有限公司 | A kind of bus degree of crowding method for visualizing |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111291766A (en) * | 2018-12-06 | 2020-06-16 | 乐金信世股份有限公司 | Image recognition method and server using deep learning |
| CN109684952A (en) * | 2018-12-13 | 2019-04-26 | 北京航空航天大学 | A kind of Passenger's distribution detection system and method suitable for high-speed rail train intelligent air condition |
| CN110335460A (en) * | 2019-05-08 | 2019-10-15 | 上海电机学院 | A kind of public traffic information intelligent interactive system and method |
| CN110458056A (en) * | 2019-07-29 | 2019-11-15 | 交控科技股份有限公司 | Compartment crowding detection method and device |
| CN110458080A (en) * | 2019-08-05 | 2019-11-15 | 西藏宁算科技集团有限公司 | The pre-judging method and system of front pit-hole in a kind of running car |
| CN110852155A (en) * | 2019-09-29 | 2020-02-28 | 深圳市深网视界科技有限公司 | Method, system, device and storage medium for detecting crowdedness of bus passengers |
| CN110782692A (en) * | 2019-10-31 | 2020-02-11 | 青岛海信网络科技股份有限公司 | Signal lamp fault detection method and system |
| CN113291703A (en) * | 2020-08-31 | 2021-08-24 | 湖南长天自控工程有限公司 | Discharge opening blockage detection method and device |
| CN112241688A (en) * | 2020-09-24 | 2021-01-19 | 厦门卫星定位应用股份有限公司 | Carriage congestion degree detection method and system |
| CN112396587A (en) * | 2020-11-20 | 2021-02-23 | 重庆大学 | Method for detecting crowding degree in bus compartment based on cooperative training and density map |
| CN112396587B (en) * | 2020-11-20 | 2024-01-30 | 重庆大学 | Method for detecting congestion degree in bus compartment based on collaborative training and density map |
| CN112904775A (en) * | 2021-01-20 | 2021-06-04 | 深圳市蓝泰源信息技术股份有限公司 | Bus-mounted intelligent terminal |
| CN115273480A (en) * | 2022-09-28 | 2022-11-01 | 南京智慧交通信息股份有限公司 | Traffic early warning system and method based on big data analysis |
| CN115273480B (en) * | 2022-09-28 | 2023-01-20 | 南京智慧交通信息股份有限公司 | Traffic early warning system and method based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108648495A (en) | A kind of method and system of the intelligence real-time display bus degree of crowding | |
| CN111079586B (en) | Automatic driving target detection system and method based on deep learning and binocular shooting | |
| CN106446949B (en) | A kind of vehicle model recognition methods and device | |
| CN110097109A (en) | A kind of road environment obstacle detection system and method based on deep learning | |
| CN109636397A (en) | Transit trip control method, device, computer equipment and storage medium | |
| CN109919140A (en) | Vehicle collision accident responsibility automatic judging method, system, equipment and storage medium | |
| CN109886210A (en) | A kind of traffic image recognition methods, device, computer equipment and medium | |
| CN105512623A (en) | Foggy-day driving visual enhancement and visibility early warning system and method based on multiple sensors | |
| CN105184271A (en) | Automatic vehicle detection method based on deep learning | |
| CN105844257A (en) | Early warning system based on machine vision driving-in-fog road denoter missing and early warning method | |
| CN105354530A (en) | Vehicle body color identification method and apparatus | |
| CN109685026B (en) | Real-time monitoring method for mobile phone call held by driver | |
| CN106529446A (en) | Vehicle type identification method and system based on multi-block deep convolutional neural network | |
| US20190197328A1 (en) | Method and apparatus for outputting information | |
| CN205665896U (en) | Intersection signal lamp state recognition device | |
| CN109635825A (en) | Vehicle attribute detection method, device and storage medium | |
| CN106485923B (en) | A kind of public transport crowding real-time status acquisition method and device | |
| CN106815574A (en) | Set up detection model, detect the method and apparatus for taking mobile phone behavior | |
| CN106682648B (en) | A kind of user takes mobile phone behavioral value method and apparatus | |
| CN111783589A (en) | Complex scene crowd counting method based on scene classification and multi-scale feature fusion | |
| CN110532925A (en) | Driver Fatigue Detection based on space-time diagram convolutional network | |
| CN112084928A (en) | Road traffic accident detection method based on visual attention mechanism and ConvLSTM network | |
| CN108960133A (en) | Method, electronic equipment, system and the storage medium of ridership monitoring | |
| CN112633149A (en) | Domain-adaptive foggy-day image target detection method and device | |
| CN103106410A (en) | Method and system for achieving intelligent bus stop board |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181012 |
|
| RJ01 | Rejection of invention patent application after publication |