CN106971544A - A kind of direct method that vehicle congestion is detected using still image - Google Patents
A kind of direct method that vehicle congestion is detected using still image Download PDFInfo
- Publication number
- CN106971544A CN106971544A CN201710339106.8A CN201710339106A CN106971544A CN 106971544 A CN106971544 A CN 106971544A CN 201710339106 A CN201710339106 A CN 201710339106A CN 106971544 A CN106971544 A CN 106971544A
- Authority
- CN
- China
- Prior art keywords
- image
- vehicle
- information
- convolutional neural
- neural networks
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0133—Traffic data processing for classifying traffic situation
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of direct method that vehicle congestion is detected using still image, it includes step：Data prediction；Multitask convolutional neural networks are trained；Detection image；Vehicle queue length and congestion judge.Image aspects, vehicle number and vehicle closeness are determined by data prediction.Gray image and illumination invariant image after sample image is changed by original RGB image, as the input of multitask convolutional neural networks, to train multitask convolutional neural networks, are trained parallel to image aspects, vehicle number and vehicle closeness.The multitask convolutional neural networks that image to be detected is inputted after training, obtain the image aspects corresponding with image to be detected, vehicle number and vehicle closeness or its combined information.According to image aspects, vehicle number and vehicle closeness comprehensive descision vehicle congestion situation.The present invention need not recognize vehicle moving process, be that can obtain the judgement that number of vehicles realizes congestion detection directly using still image.
Description
Technical field
The present invention relates to a kind of vehicle congestion detection method, and in particular to one kind directly detects vehicle using still image
The method of congestion.
Background technology
With the increase of urban population and urban traffic flow, city, which is particularly metropolitan traffic problems, commonly turns into Jiao
Point problem.The congested in traffic blockage problem of urban road has turned into restriction economic development, reduction people's living standard, has weakened economy
One of bottleneck of vigor.So, how quickly and efficiently detection congestion in road is the need for complying with social development.
Traffic congestion, for traveler, mainly to time and the sensation of speed, i.e. vehicle in road or intersection
Upper queuing is slow mobile.The Ministry of Public Security of China sets forth definition to intersection and congested link：Vehicle is in no signal
It is obstructed at the intersection of lamp control on driveway and queue length is more than 250 meters, or vehicle is in the crossroad of Signalized control
Mouthful, No. 3 green lights are shown is not defined as intersection by the state at crossing；Congested link be then defined as vehicle on driveway by
Resistance and state of the queue length more than 1 kilometer.
From prior art, a traditional methods are started with itself from vehicle, are detected, utilized using sensor
GPS positioning technology carries out congestion detection, in addition with means such as ground induction coil detection, Data mining, microwave detections.But
All existed in these traditional detection methods in place of some shortcomings, cost of investment height, the physical equipment for example detected holds
It is fragile, the shortcomings of be difficult to safeguard.
Another kind of is to utilize video image technology for detection traffic congestion, using video image processing technology, in video
Vehicle is identified and tracked, and combines some speed, azimuth information, and congestion is detected.Such video detection technology
It there are many shortcomings, vehicle serious shielding each other, the track following based on single unit vehicle tends to failure.In addition,
The height of traffic shooting, angle is also different, and some vehicles are obscured very much very in video or image, and this also gives detection band
Carry out extreme difficulties, cause the inaccurate of result.
The content of the invention
The present invention proposes a kind of direct method that vehicle congestion is detected using still image, it is not necessary to recognize that vehicle is moved
Process is that can obtain the judgement that number of vehicles realizes congestion detection.
The present invention solution be：A kind of direct method that vehicle congestion is detected using still image, suitable in meter
Calculate in machine equipment and perform, it comprises the following steps：
Data prediction：Every width sample image in training sample image data is divided into m × n subgraph, people is taken
The form of work demarcation assigns its two attribute information for each subgraph：Vehicle number information and vehicle closeness information；Original image
Image aspects information is assigned according to shooting grease head highness visual angle；
Multitask convolutional neural networks are trained：Multitask convolutional neural networks are multilayer convolution, and sample image is passed through into original
Gray image and illumination invariant image after the conversion of beginning RGB image are more to train as the input of multitask convolutional neural networks
Task convolutional neural networks, are trained parallel to image aspects information, vehicle number information and vehicle closeness information；
Detection image：The multitask convolutional neural networks that image to be detected is inputted after training, are obtained and image to be detected
Corresponding image aspects information, vehicle number information and vehicle closeness information or its combined information；
Vehicle queue length and congestion judge：It is comprehensive according to image aspects information, vehicle number information and vehicle closeness information
Conjunction judges vehicle congestion situation.
As the further improvement of such scheme, vehicle closeness information is determined by vehicle number information.
Further, 4 ranks of vehicle closeness information definition：The corresponding vehicle of vehicle number information of 0-5 car is intensive
It is 0 to spend information；The corresponding vehicle closeness information of vehicle number information of 6-10 car is 1；The vehicle number information pair of 11-15 car
The vehicle closeness information answered is 2；The corresponding vehicle closeness information of vehicle number information of 16 cars is 3.
As the further improvement of such scheme, image aspects information by road camera setting height(from bottom) and shooting visual angle
Determine.
Further, image aspects information is divided into basic, normal, high by three according to the setting height(from bottom) and shooting visual angle of road camera
Plant visual angle.
As the further improvement of such scheme, multitask convolutional neural networks carry out convolution to both direction：One side
To for image aspects information, other direction is vehicle number information and vehicle closeness information.
Further, the loss function defined in multitask convolutional neural networks：
First, vehicle count loss function：Vehicle count is regarded as depth regression problem, input is whole image, and output is
The vehicle number of different subgraphs, minimizes Euclidean lossUsing formulaWherein, k is the rope of subgraph
Draw, Gk and Pk are the true of kth subgraph and prediction vehicle number；
2nd, vehicle closeness loss function：Vehicle closeness is defined according to vehicle number, the loss of vehicle closenessAdopt
Use formulaWherein k is the index of subgraph, and pkm is the areal densities grade of kth subgraph；
3rd, image aspects loss function：Image aspects are lostUsing
Wherein, C is the sum of image aspects classification, yic
And picIt is prediction and true picture visual angle respectively；
By above three loss function, the multitask convolutional neural networks after training obtain convolutional neural networks model：Wherein, N is the quantity of training sample image, and λ is used to balance different loss letters
Number, W and b are model parameters, and the multitask convolutional neural networks that λ is set to after 0 training will not consider image aspects.
As the further improvement of such scheme, multitask convolutional neural networks share 3 convolutional layers.
The present invention also provides a kind of storage device, wherein storing a plurality of instruction, the instruction is loaded and held suitable for processor
OK, a plurality of instruct is：
Data prediction：Every width sample image in training sample image data is divided into m × n subgraph, people is taken
The form of work demarcation assigns its two attribute information for each subgraph：Vehicle number information and vehicle closeness information；Original image
Image aspects information is assigned according to shooting grease head highness visual angle；
Multitask convolutional neural networks are trained：Multitask convolutional neural networks are multilayer convolution, and sample image is passed through into original
Gray image and illumination invariant image after the conversion of beginning RGB image are more to train as the input of multitask convolutional neural networks
Task convolutional neural networks, are trained parallel to image aspects information, vehicle number information and vehicle closeness information；
Detection image：The multitask convolutional neural networks that image to be detected is inputted after training, are obtained and image to be detected
Corresponding image aspects information, vehicle number information and vehicle closeness information or its combined information；
Vehicle queue length and congestion judge：It is comprehensive according to image aspects information, vehicle number information and vehicle closeness information
Conjunction judges vehicle congestion situation.
The present invention also provides a kind of mobile communication terminal, and it includes：
Processor, is adapted for carrying out various instructions；And
Storage device, suitable for storing a plurality of instruction, the instruction is loaded and performed suitable for processor；
It is described it is a plurality of instruction be：
Data prediction：Every width sample image in training sample image data is divided into m × n subgraph, people is taken
The form of work demarcation assigns its two attribute information for each subgraph：Vehicle number information and vehicle closeness information；Original image
Image aspects information is assigned according to shooting grease head highness visual angle；
Multitask convolutional neural networks are trained：Multitask convolutional neural networks are multilayer convolution, and sample image is passed through into original
Gray image and illumination invariant image after the conversion of beginning RGB image are more to train as the input of multitask convolutional neural networks
Task convolutional neural networks, are trained parallel to image aspects information, vehicle number information and vehicle closeness information；
Detection image：The multitask convolutional neural networks that image to be detected is inputted after training, are obtained and image to be detected
Corresponding image aspects information, vehicle number information and vehicle closeness information or its combined information；
Vehicle queue length and congestion judge：It is comprehensive according to image aspects information, vehicle number information and vehicle closeness information
Conjunction judges vehicle congestion situation.
The method that vehicle congestion is directly detected using still image of the present invention, utilizes multitask convolutional neural networks
Mode, multitask convolutional neural networks model is obtained by training, therefore in actually detected, is only captured by a road
Image is with regard to that can detect the situation of congestion in road.
Brief description of the drawings
Fig. 1 directly detects the flow chart of the method for vehicle congestion for the present invention using still image.
When Fig. 2 trains for multitask convolutional neural networks in Fig. 1, the training method figure of multitask convolutional neural networks.
Fig. 3 be Fig. 1 in data prediction when, the acquisition of image aspects information, vehicle number information and vehicle closeness information
Exemplary plot.
When Fig. 4 trains for multitask convolutional neural networks in Fig. 1, the input mode exemplary plot of sample image.
When Fig. 5 trains for multitask convolutional neural networks in Fig. 1, the exemplary plot of the multitask convolutional neural networks of use.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
The method that vehicle congestion is directly detected using still image of the present invention, suitable for being performed in computer equipment,
The form of software installation bag can be made in actual applications, can also make APP form, such as mobile phone A PP.
Referring to Fig. 1, directly detecting that the method for vehicle congestion is mainly included the following steps that using still image：
Data prediction：Every width sample image in training sample image data is divided into m × n subgraph, people is taken
The form of work demarcation assigns its two attribute information for each subgraph：Vehicle number information and vehicle closeness information；Original image
Image aspects information is assigned according to shooting grease head highness visual angle；
Multitask convolutional neural networks are trained：Multitask convolutional neural networks are multilayer convolution, and sample image is passed through into original
Gray image and illumination invariant image after the conversion of beginning RGB image are more to train as the input of multitask convolutional neural networks
Task convolutional neural networks, are trained parallel to image aspects information, vehicle number information and vehicle closeness information；
Detection image：The multitask convolutional neural networks that image to be detected is inputted after training, are obtained and image to be detected
Corresponding image aspects information, vehicle number information and vehicle closeness information or its combined information；
Vehicle queue length and congestion judge：It is comprehensive according to image aspects information, vehicle number information and vehicle closeness information
Conjunction judges vehicle congestion situation.
Incorporated by reference to Fig. 2, in the present case, multitask convolutional neural networks carry out convolution to both direction：One direction is figure
As Viewing-angle information, other direction is vehicle number information and vehicle closeness information.
The present invention obtains multitask convolutional neural networks mould using the mode of multitask convolutional neural networks by training
Type, in actually detected, the image only captured by a road is with regard to that can detect the situation of congestion in road.From current existing side
Case is seen, does not find also only to carry out the scheme that congestion in road detection judges by picture.
Subsequently, give one example and explanation is explained in detail.
1) data prediction
Every width original image in training sample image data is divided into 4 × 4 subgraphs, each subgraph, which has, there can be 2 kinds
Attribute information：Vehicle number and vehicle closeness, wherein, vehicle closeness can be determined by vehicle number.
4 ranks of vehicle closeness definable：The corresponding vehicle closeness information of vehicle number information of 0-5 car is 0；6-
The corresponding vehicle closeness information of vehicle number information of 10 cars is 1；The corresponding vehicle of vehicle number information of 11-15 car is intensive
It is 2 to spend information；The corresponding vehicle closeness information of vehicle number information of 16 cars is 3.
Image aspects information determines that image aspects information is according to road by the setting height(from bottom) and shooting visual angle of road camera
The setting height(from bottom) and shooting visual angle of camera are divided into basic, normal, high three kinds of visual angles.As shown in Figure 3, whole image also has a figure
As Viewing-angle information, height and visual angle depending on road camera are divided into basic, normal, high three kinds.
2) multitask convolutional neural networks are trained
The training of multitask convolutional neural networks, which can be divided into, to be described in detail：Multichannel is inputted, multitask convolutional Neural net
Network.
Multichannel is inputted：Each training sample figure takes multichannel input mode, as shown in Figure 4.Multichannel is by original
The gray image and illumination invariant image composition of beginning RGB image conversion.Illumination invariant image can describe object well, it is easy to
Detection.
Multitask convolutional neural networks：Multitask convolutional neural networks framework is as shown in figure 5, vehicle count and image aspects
First share 3 convolutional layers, then to the two directions carry out convolution, the Training strategy of multitask convolutional neural networks and other
As the training of neutral net, when loss function is restrained, obtained model will be trained to be tested on checking collection, for inspection
The result of sniffing by mistake is analyzed, and is focused to find out the image of some corresponding types in supplemental training according to the type of the image of mistake
It is added in training set, proceeds training to network.In the present embodiment, it is defined respectively as loss function.
Vehicle technology loss function：Vehicle count can regard depth regression problem as, and input is whole image, and output is not
With the vehicle number of subgraph.Therefore, it is intended that minimizing Euclidean loss
Wherein k is the index of subgraph, and Gk and Pk are the true of kth subgraph and prediction vehicle number.
Vehicle closeness loss function：Vehicle closeness can be defined according to vehicle number, but can be with this network
Directly show that vehicle closeness is lost by neutral net by defining loss function
Wherein, k is the index of subgraph, pkmAreal densities for kth subgraph etc.
Level.The recurrence of vehicle number and traffic density horizontal classification share all deconvolution parameters, simply different in interior lamination parameter.
Image aspects loss function：Image aspects are lost
Herein, C is the sum (being set to 3 as described above) of image aspects classification, yicAnd picRespectively be prediction and
True picture visual angle.
Solved finally by majorized function, form convolutional neural networks model.
Herein, N is the quantity of training image, and λ is used to balance different loss functions.W and b are model parameters.λ is set
For 0, the network will not consider image aspects.
3) detection image
Image to be detected is inputted, number of vehicles can be obtained using only vehicle count flow and judges vehicle congestion situation.
4) vehicle queue length and congestion judge
Vehicle congestion situation is judged according to image aspects information, vehicle number information and vehicle closeness informix.
Congestion is determined with a variety of methods, such as number of vehicles threshold value is set respectively according to no view level, more than threshold value
As congestion, or congestion can be judged with the ratio or index information of number of vehicles and Viewing-angle information.It is long as vehicle queue
Degree, in general, vehicle more multiple-length is longer, is not that say have to a definite length.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
Any modifications, equivalent substitutions and improvements made within refreshing and principle etc., should be included in the scope of the protection.
Claims (10)
1. a kind of direct method that vehicle congestion is detected using still image, suitable for being performed in computer equipment, its feature
It is：It comprises the following steps：
Data prediction：Every width sample image in training sample image data is divided into m × n subgraph, artificial mark is taken
Fixed form is that each subgraph assigns its two attribute information：Vehicle number information and vehicle closeness information；Original image according to
Image grease head highness visual angle and assign image aspects information；
Multitask convolutional neural networks are trained：Multitask convolutional neural networks are multilayer convolution, and sample image is passed through into original RGB
Gray image and illumination invariant image after image conversion, as the input of multitask convolutional neural networks, to train multitask
Convolutional neural networks, are trained parallel to image aspects information, vehicle number information and vehicle closeness information；
Detection image：The multitask convolutional neural networks that image to be detected is inputted after training, obtain relative with image to be detected
Image aspects information, vehicle number information and the vehicle closeness information answered or its combined information；
Vehicle queue length and congestion judge：Sentenced according to image aspects information, vehicle number information and vehicle closeness informix
Disconnected vehicle congestion situation.
2. the method as claimed in claim 1 that vehicle congestion is directly detected using still image, it is characterised in that：Vehicle is close
Intensity information is determined by vehicle number information.
3. the method as claimed in claim 2 that vehicle congestion is directly detected using still image, it is characterised in that：Vehicle is close
4 ranks of intensity information definition：The corresponding vehicle closeness information of vehicle number information of 0-5 car is 0；The vehicle of 6-10 car
The corresponding vehicle closeness information of number information is 1；The corresponding vehicle closeness information of vehicle number information of 11-15 car is 2；16
The corresponding vehicle closeness information of vehicle number information of car is 3.
4. the method as claimed in claim 1 that vehicle congestion is directly detected using still image, it is characterised in that：Image is regarded
Angle information is determined by the setting height(from bottom) and shooting visual angle of road camera.
5. the method as claimed in claim 4 that vehicle congestion is directly detected using still image, it is characterised in that：Image is regarded
Angle information is divided into basic, normal, high three kinds of visual angles according to the setting height(from bottom) and shooting visual angle of road camera.
6. the method as claimed in claim 1 that vehicle congestion is directly detected using still image, it is characterised in that：Multitask
Convolutional neural networks carry out convolution to both direction：One direction is image aspects information, other direction be vehicle number information and
Vehicle closeness information.
7. the method as claimed in claim 6 that vehicle congestion is directly detected using still image, it is characterised in that：At many
Loss function defined in business convolutional neural networks：
First, vehicle count loss function：Vehicle count is regarded as depth regression problem, input is whole image, and output is different
The vehicle number of subgraph, minimizes Euclidean lossUsing formulaWherein, k is the index of subgraph, Gk
It is the true of kth subgraph and prediction vehicle number with Pk；
2nd, vehicle closeness loss function：Vehicle closeness is defined according to vehicle number, the loss of vehicle closenessUsing public affairs
FormulaWherein k is the index of subgraph, and pkm is the areal densities grade of kth subgraph；
3rd, image aspects loss function：Image aspects are lostUsing
Wherein, C is the sum of image aspects classification, yicAnd pic
It is prediction and true picture visual angle respectively；
By above three loss function, the multitask convolutional neural networks after training obtain convolutional neural networks model：Wherein, N is the quantity of training sample image, and λ is used to balance different loss letters
Number, W and b are model parameters, and the multitask convolutional neural networks that λ is set to after 0 training will not consider image aspects.
8. the method as claimed in claim 1 that vehicle congestion is directly detected using still image, it is characterised in that：Multitask
Convolutional neural networks share 3 convolutional layers.
9. a kind of storage device, wherein storing a plurality of instruction, the instruction is loaded and performed suitable for processor, it is characterised in that：
It is described it is a plurality of instruction be：
Data prediction：Every width sample image in training sample image data is divided into m × n subgraph, artificial mark is taken
Fixed form is that each subgraph assigns its two attribute information：Vehicle number information and vehicle closeness information；Original image according to
Image grease head highness visual angle and assign image aspects information；
Multitask convolutional neural networks are trained：Multitask convolutional neural networks are multilayer convolution, and sample image is passed through into original RGB
Gray image and illumination invariant image after image conversion, as the input of multitask convolutional neural networks, to train multitask
Convolutional neural networks, are trained parallel to image aspects information, vehicle number information and vehicle closeness information；
Detection image：The multitask convolutional neural networks that image to be detected is inputted after training, obtain relative with image to be detected
Image aspects information, vehicle number information and the vehicle closeness information answered or its combined information；
Vehicle queue length and congestion judge：Sentenced according to image aspects information, vehicle number information and vehicle closeness informix
Disconnected vehicle congestion situation.
10. a kind of mobile communication terminal, it includes：
Processor, is adapted for carrying out various instructions；And
Storage device, suitable for storing a plurality of instruction, the instruction is loaded and performed suitable for processor；
It is characterized in that：It is described it is a plurality of instruction be：
Data prediction：Every width sample image in training sample image data is divided into m × n subgraph, artificial mark is taken
Fixed form is that each subgraph assigns its two attribute information：Vehicle number information and vehicle closeness information；Original image according to
Image grease head highness visual angle and assign image aspects information；
Multitask convolutional neural networks are trained：Multitask convolutional neural networks are multilayer convolution, and sample image is passed through into original RGB
Gray image and illumination invariant image after image conversion, as the input of multitask convolutional neural networks, to train multitask
Convolutional neural networks, are trained parallel to image aspects information, vehicle number information and vehicle closeness information；
Detection image：The multitask convolutional neural networks that image to be detected is inputted after training, obtain relative with image to be detected
Image aspects information, vehicle number information and the vehicle closeness information answered or its combined information；
Vehicle queue length and congestion judge：Sentenced according to image aspects information, vehicle number information and vehicle closeness informix
Disconnected vehicle congestion situation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710339106.8A CN106971544B (en) | 2017-05-15 | 2017-05-15 | A kind of direct method that vehicle congestion is detected using still image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710339106.8A CN106971544B (en) | 2017-05-15 | 2017-05-15 | A kind of direct method that vehicle congestion is detected using still image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106971544A true CN106971544A (en) | 2017-07-21 |
| CN106971544B CN106971544B (en) | 2019-07-16 |
Family
ID=59331648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710339106.8A Active CN106971544B (en) | 2017-05-15 | 2017-05-15 | A kind of direct method that vehicle congestion is detected using still image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106971544B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107808140A (en) * | 2017-11-07 | 2018-03-16 | 浙江大学 | A kind of monocular vision Road Recognition Algorithm based on image co-registration |
| CN108039044A (en) * | 2017-12-05 | 2018-05-15 | 安徽大学 | The system and method that Vehicular intelligent based on multiple dimensioned convolutional neural networks is lined up |
| CN108550259A (en) * | 2018-04-19 | 2018-09-18 | 何澜 | Congestion in road judgment method, terminal device and computer readable storage medium |
| CN109255374A (en) * | 2018-08-27 | 2019-01-22 | 中共中央办公厅电子科技学院 | A kind of aesthetic properties evaluation method based on intensive convolutional network and multitask network |
| CN109858372A (en) * | 2018-12-29 | 2019-06-07 | 浙江零跑科技有限公司 | A kind of lane class precision automatic Pilot structured data analysis method |
| CN109887276A (en) * | 2019-01-30 | 2019-06-14 | 北京同方软件股份有限公司 | The night traffic congestion detection method merged based on foreground extraction with deep learning |
| CN110390822A (en) * | 2019-05-31 | 2019-10-29 | 东南大学 | Bridge statistical method of traffic flow based on FBG sensor and convolutional neural networks |
| CN112489456A (en) * | 2020-12-01 | 2021-03-12 | 山东交通学院 | Signal lamp regulation and control method and system based on urban trunk line vehicle queuing length |
| CN113570858A (en) * | 2021-07-22 | 2021-10-29 | 吉林大学 | System and method for assisting vehicle to identify traffic jam condition by unmanned aerial vehicle |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036323A (en) * | 2014-06-26 | 2014-09-10 | 叶茂 | Vehicle detection method based on convolutional neural network |
| CN106023605A (en) * | 2016-07-15 | 2016-10-12 | 姹ゅ钩 | Traffic signal lamp control method based on deep convolution neural network |
| CN106297297A (en) * | 2016-11-03 | 2017-01-04 | 成都通甲优博科技有限责任公司 | Traffic jam judging method based on degree of depth study |
| CN106599869A (en) * | 2016-12-22 | 2017-04-26 | 安徽大学 | Vehicle attribute identification method based on multi-task convolutional neural network |
| CN106611162A (en) * | 2016-12-20 | 2017-05-03 | 西安电子科技大学 | Method for real-time detection of road vehicle based on deep learning SSD frame |
| CN106650913A (en) * | 2016-12-31 | 2017-05-10 | 中国科学技术大学 | Deep convolution neural network-based traffic flow density estimation method |
-
2017
- 2017-05-15 CN CN201710339106.8A patent/CN106971544B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036323A (en) * | 2014-06-26 | 2014-09-10 | 叶茂 | Vehicle detection method based on convolutional neural network |
| CN106023605A (en) * | 2016-07-15 | 2016-10-12 | 姹ゅ钩 | Traffic signal lamp control method based on deep convolution neural network |
| CN106297297A (en) * | 2016-11-03 | 2017-01-04 | 成都通甲优博科技有限责任公司 | Traffic jam judging method based on degree of depth study |
| CN106611162A (en) * | 2016-12-20 | 2017-05-03 | 西安电子科技大学 | Method for real-time detection of road vehicle based on deep learning SSD frame |
| CN106599869A (en) * | 2016-12-22 | 2017-04-26 | 安徽大学 | Vehicle attribute identification method based on multi-task convolutional neural network |
| CN106650913A (en) * | 2016-12-31 | 2017-05-10 | 中国科学技术大学 | Deep convolution neural network-based traffic flow density estimation method |
Non-Patent Citations (2)
| Title |
|---|
| YINGYING ZHANG: "Single-Image Crowd Counting via Multi-Column Convolutional Neural Network", 《2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)》 * |
| 刘张: "城市交通运行状态分析与预测关键技术研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107808140A (en) * | 2017-11-07 | 2018-03-16 | 浙江大学 | A kind of monocular vision Road Recognition Algorithm based on image co-registration |
| CN107808140B (en) * | 2017-11-07 | 2020-07-31 | 浙江大学 | Monocular vision road recognition algorithm based on image fusion |
| CN108039044B (en) * | 2017-12-05 | 2021-06-01 | 安徽大学 | Vehicle intelligent queuing system and method based on multi-scale convolutional neural network |
| CN108039044A (en) * | 2017-12-05 | 2018-05-15 | 安徽大学 | The system and method that Vehicular intelligent based on multiple dimensioned convolutional neural networks is lined up |
| CN108550259A (en) * | 2018-04-19 | 2018-09-18 | 何澜 | Congestion in road judgment method, terminal device and computer readable storage medium |
| CN108550259B (en) * | 2018-04-19 | 2020-05-12 | 何澜 | Road congestion judging method, terminal device and computer readable storage medium |
| CN109255374A (en) * | 2018-08-27 | 2019-01-22 | 中共中央办公厅电子科技学院 | A kind of aesthetic properties evaluation method based on intensive convolutional network and multitask network |
| CN109858372A (en) * | 2018-12-29 | 2019-06-07 | 浙江零跑科技有限公司 | A kind of lane class precision automatic Pilot structured data analysis method |
| CN109887276A (en) * | 2019-01-30 | 2019-06-14 | 北京同方软件股份有限公司 | The night traffic congestion detection method merged based on foreground extraction with deep learning |
| CN110390822A (en) * | 2019-05-31 | 2019-10-29 | 东南大学 | Bridge statistical method of traffic flow based on FBG sensor and convolutional neural networks |
| CN112489456A (en) * | 2020-12-01 | 2021-03-12 | 山东交通学院 | Signal lamp regulation and control method and system based on urban trunk line vehicle queuing length |
| CN112489456B (en) * | 2020-12-01 | 2022-01-28 | 山东交通学院 | Signal lamp regulation and control method and system based on urban trunk line vehicle queuing length |
| WO2022116361A1 (en) * | 2020-12-01 | 2022-06-09 | 山东交通学院 | Traffic light control method and system based on urban trunk line vehicle queuing length |
| CN113570858A (en) * | 2021-07-22 | 2021-10-29 | 吉林大学 | System and method for assisting vehicle to identify traffic jam condition by unmanned aerial vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106971544B (en) | 2019-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106971544A (en) | A kind of direct method that vehicle congestion is detected using still image | |
| CN104751634B (en) | The integrated application method of freeway tunnel driving image acquisition information | |
| CN104778834B (en) | Urban road traffic jam judging method based on vehicle GPS data | |
| CN105185140B (en) | A kind of auxiliary driving method and system | |
| CN109410601A (en) | Method for controlling traffic signal lights, device, electronic equipment and storage medium | |
| Petrovska et al. | Traffic congestion analysis visualisation tool | |
| CN103903465B (en) | A kind of congestion in road reason real-time release method and system | |
| CN107301776A (en) | Track road conditions processing and dissemination method based on video detection technology | |
| CN109615864A (en) | Vehicle congestion analysis method, system, terminal and storage medium based on video structural | |
| CN106530730A (en) | Traffic violation detection method and system | |
| CN102945603A (en) | Method for detecting traffic event and electronic police device | |
| CN205665896U (en) | Intersection signal lamp state recognition device | |
| CN109115237A (en) | A kind of position recommended method and server of riding | |
| CN107240264B (en) | A kind of non-effective driving trace recognition methods of vehicle and urban road facility planing method | |
| CN106781582A (en) | Traffic lights auxiliary display method, device and car-mounted terminal | |
| CN107578620B (en) | Method, system and device for calculating road traffic capacity based on lane level system | |
| CN105469599B (en) | Vehicle tracing and vehicle behavior prediction method | |
| CN105741566A (en) | Traffic information display system controlled based on intelligent traffic management system | |
| CN103177595A (en) | Dynamic routing optimization system and method based on real-time traffic information | |
| CN107590999A (en) | A kind of traffic state judging method based on bayonet socket data | |
| CN106846819A (en) | A kind of taking photo by plane based on small aircraft and realizes system at penalty note generation method | |
| CN107480653A (en) | passenger flow volume detection method based on computer vision | |
| CN105912979B (en) | A kind of detection method and device of vehicle road occupying | |
| CN106558217A (en) | A kind of method of acquisition parking lay-by information, device and server | |
| Zhou et al. | Method for judging parking status based on yolov2 target detection algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 |