CN114419555A - Road traffic target tracking method based on deep convolutional neural network - Google Patents
Road traffic target tracking method based on deep convolutional neural network Download PDFInfo
- Publication number
- CN114419555A CN114419555A CN202210062436.8A CN202210062436A CN114419555A CN 114419555 A CN114419555 A CN 114419555A CN 202210062436 A CN202210062436 A CN 202210062436A CN 114419555 A CN114419555 A CN 114419555A
- Authority
- CN
- China
- Prior art keywords
- neural network
- convolutional neural
- target
- road
- monitoring terminal
- 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
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000012544 monitoring process Methods 0.000 claims abstract description 45
- 230000005540 biological transmission Effects 0.000 claims abstract description 4
- 238000001514 detection method Methods 0.000 claims description 14
- 230000000694 effects Effects 0.000 abstract description 2
- 230000033001 locomotion Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 206010000117 Abnormal behaviour Diseases 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000474 nursing effect Effects 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Abstract
The invention belongs to the technical field of road traffic, and particularly relates to a road traffic target tracking method based on a deep convolutional neural network, which comprises the following operation steps of: s1: and (3) inputting the license plate and the characteristics of the target vehicle to be searched through the monitoring terminal, S2: the monitoring terminal screens and checks the road camera in multiple angles and multiple positions through the internal convolutional neural network transmission module, the invention can search and accurately capture target vehicles in road traffic with huge flow, and can quickly and effectively capture and track the target vehicles through the intelligence of the convolutional neural network, thereby increasing the timeliness and speed of locking, preventing the target vehicles from being lost, detecting the road camera and each module, ensuring the normal operation of the device, preventing the equipment from being damaged and being incapable of monitoring the road condition in time, and facilitating the effect of maintaining the equipment in time.
Description
Technical Field
The invention relates to the technical field of road traffic, in particular to a road traffic target tracking method based on a deep convolutional neural network.
Background
Road traffic refers to traffic for short, and the traffic retention of vehicles and pedestrians on roads sometimes also include parking vehicles, the service of enterprises for the flow of people and the movement of people for goods, and sometimes also include the storage of goods, and is called transportation, such as passenger transportation, goods transportation, road transportation, railway transportation, air transportation, waterway transportation, and the like, and the two are collectively called transportation or traffic and transportation, and are also often referred to as traffic for short or transportation for short in china.
The convolutional neural network is a feedforward neural network containing convolutional calculation and having a deep structure, and is one of representative algorithms for deep learning.
The convolutional neural network is constructed by imitating a visual perception mechanism of a living being, and can perform supervised learning and unsupervised learning, and the parameter sharing of convolution kernels in hidden layers and the sparsity of interlayer connection enable the convolutional neural network to learn lattice characteristics such as pixels and audio with small calculation amount, have stable effect and have no additional characteristic engineering on data.
The intelligent multi-target tracking system adopts advanced image detection, identification and tracking technologies and is matched with a precise motion control system to realize continuous and rapid tracking capture of a plurality of moving targets in a large scene;
the system integrates the tracking detection and video analysis functions of multiple targets in a large scene into an independent system, intelligently analyzes video information acquired by a front-end camera, automatically acquires and classifies data of abnormal behaviors and events and gives an alarm in a linkage manner, meanwhile, a background can see the analyzed data and video records in real time, and video extraction and evidence collection can be carried out through event retrieval afterwards.
One of the biggest advantages of the system is that different behavior modes of a plurality of targets in the same scene can be recognized and monitored simultaneously, the system can be widely applied to various large public places, including important places such as airports, stations, prisons, ports, mines, oil fields, nursing homes, streets, communities and markets, and is used for detecting, classifying, tracking and recording passing pedestrians, vehicles and other suspicious objects, judging whether abnormal behaviors exist or not and giving an alarm.
The existing road traffic has huge vehicle flow, when a vehicle needs to be searched, the vehicle cannot be accurately, quickly and effectively captured and tracked through an intelligent multi-target tracking system, so that the vehicle cannot be locked in time, and therefore, a road traffic target tracking method based on a deep convolutional neural network is provided.
Disclosure of Invention
The present invention has been made in view of the above and/or other problems occurring in the prior art, and/or a road traffic target tracking method based on a deep convolutional neural network.
Therefore, the invention aims to provide a road traffic target tracking method based on a deep convolutional neural network, which can solve the existing problems provided by the prior art, wherein a monitoring terminal inputs license plates and characteristics of target vehicles to be searched, so that the convolutional neural network captures commands for road cameras, and real-time positioning and real-time images and videos of the target vehicles are transmitted to the monitoring terminal for tracking and displaying through mutual cooperation of a plurality of groups of road cameras.
To solve the above technical problem, according to an aspect of the present invention, the present invention provides the following technical solutions:
a road traffic target tracking method based on a deep convolutional neural network comprises the following steps: the method comprises the following operation steps:
s1: the license plate and the characteristic of the target vehicle to be searched are input through the monitoring terminal;
s2: the monitoring terminal screens and checks the road camera in multiple angles and multiple positions through the internal convolutional neural network transmission module;
s3: carrying out target vehicle identification on a road camera through a convolutional neural network, and reading and checking vehicles with similar characteristics and similar license plates to the target vehicle;
s4: transmitting the checked target vehicle to a monitoring terminal for judgment;
s5: after the target vehicle is determined, the target vehicle is captured and locked through the target capture module and the multiple groups of road cameras, and the target vehicle is ensured to be locked in the field of view of the road cameras;
s6: after a target vehicle is captured, the captured images of the road camera are displayed on a monitoring terminal through a tracking switching module, a plurality of groups of images and a plurality of groups of images at different positions are presented through the driving of the vehicle, and the images are switched in real time;
s7: the method comprises the steps that when a target vehicle is captured and tracked, the real-time positioning of a road camera at the moment is sent to a monitoring terminal through a convolutional neural network, so that the moving track of the vehicle is displayed on the monitoring terminal;
s8: after the tracking and capturing task is finished, each module and the camera are routinely detected through the convolutional neural network and the detection module, and the next task can be smoothly implemented.
As a preferred scheme of the road traffic target tracking method based on the deep convolutional neural network, the method comprises the following steps: in S1, the monitoring terminal is used for inputting vehicle information and displaying the target vehicle.
As a preferred scheme of the road traffic target tracking method based on the deep convolutional neural network, the method comprises the following steps: the convolutional neural network in S2 is used to transmit and command the device through the intelligent network.
As a preferred scheme of the road traffic target tracking method based on the deep convolutional neural network, the method comprises the following steps: if the determination of the monitoring terminal in S4 is yes, the process continues to step S5, and if the determination of the monitoring terminal is no, the process returns to S1 to continue to screen the vehicles, and the target vehicle screened for the first time is intelligently discharged.
As a preferred scheme of the road traffic target tracking method based on the deep convolutional neural network, the method comprises the following steps: the S5 target capturing module is used for capturing a dynamic view angle, and the target capturing module is used for dynamically capturing a target vehicle and determining a vehicle traveling route.
As a preferred scheme of the road traffic target tracking method based on the deep convolutional neural network, the method comprises the following steps: and the tracking switching module in the S6 switches video images, locks the target vehicle through a plurality of groups of road cameras, and transmits the picture to the monitoring terminal.
As a preferred scheme of the road traffic target tracking method based on the deep convolutional neural network, the method comprises the following steps: and the positioning in the S7 is set as a GPS/Beidou satellite, and the positioning can be carried out aiming at the position feedback of the road camera and transmits the position information of the target vehicle to the monitoring terminal.
As a preferred scheme of the road traffic target tracking method based on the deep convolutional neural network, the method comprises the following steps: the detection module compares the operation data of the camera and each module with the normal operation data through the neural network, and sends the detection result to the monitoring terminal.
Compared with the prior art: the method comprises the steps that license plates and features of target vehicles needing to be searched are input through a monitoring terminal, a road camera is subjected to a capture command through a convolutional neural network, the feature vehicles are screened and checked through the road camera to determine the target vehicles, then a plurality of groups of road cameras are matched with one another, real-time positioning and real-time images and videos of the target vehicles are transmitted to the monitoring terminal to be tracked and displayed, the monitoring terminal can determine and deny the target vehicles, positioning and tracking accuracy is improved, the target vehicles can be searched and accurately captured in the road traffic with huge flow, the target vehicles can be rapidly and effectively captured and tracked through the intelligence of the convolutional neural network, locking timeliness and locking speed are improved, and the target vehicles are prevented from being lost;
after the task is finished, the running data of the camera and each module is compared with the normal running data through the detection module and the convolutional neural network, the road camera and each module are detected, the normal running of the device is ensured, the device is prevented from being damaged, the road condition cannot be monitored in time, and the device is convenient to maintain in time.
Drawings
Fig. 1 is a schematic view of the overall flow structure provided by the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The invention provides a road traffic target tracking method based on a deep convolutional neural network, which has the advantages that a target vehicle can be searched and accurately captured in road traffic with huge flow, the target vehicle can be rapidly and effectively captured and tracked through the intelligence of the convolutional neural network, the timeliness and the speed of locking are increased, and the target vehicle is prevented from being lost, please refer to fig. 1, and the method comprises the following operation steps:
s1: the license plate and the characteristic of the target vehicle to be searched are input through the monitoring terminal;
the monitoring terminal is used for inputting vehicle information and displaying a target vehicle and is set as a computer;
s2: the monitoring terminal screens and checks the road camera in multiple angles and multiple positions through the internal convolutional neural network transmission module;
the convolutional neural network transmits and commands to the equipment through the intelligent network;
s3: carrying out target vehicle identification on a road camera through a convolutional neural network, and reading and checking vehicles with similar characteristics and similar license plates to the target vehicle;
s4: transmitting the checked target vehicle to a monitoring terminal for judgment, if the monitoring terminal judges that the vehicle is the right vehicle, continuing to step S5, if the monitoring terminal judges that the vehicle is the no vehicle, returning to S1 to continue screening the vehicle, and intelligently discharging the first screened target vehicle;
s5: after the target vehicle is determined, the target vehicle is captured and locked through the target capture module and the multiple groups of road cameras, and the target vehicle is ensured to be locked in the field of view of the road cameras;
the target capturing module is used for capturing a dynamic visual angle, and dynamically capturing a target vehicle to determine a vehicle travelling route;
the target capture module is a dynamic capture instrument, which can be called a dynamic capture system. The general motion capture system hardware comprises a capture camera, a connecting cable, hub hardware for power supply and data exchange, a system calibration kit, special capture clothes and a capture reflective ball, and the general system is provided with special motion capture software for system setting, capture process control, editing and processing of capture data, output and the like;
s6: after a target vehicle is captured, the captured images of the road camera are displayed on a monitoring terminal through a tracking switching module, a plurality of groups of images and a plurality of groups of images at different positions are presented through the driving of the vehicle, and the images are switched in real time;
the tracking switching module is used for switching video images, locks a target vehicle through a plurality of groups of road cameras and transmits pictures to the monitoring terminal;
s7: the method comprises the steps that when a target vehicle is captured and tracked, the real-time positioning of a road camera at the moment is sent to a monitoring terminal through a convolutional neural network, so that the moving track of the vehicle is displayed on the monitoring terminal;
the positioning is set as a GPS/Beidou satellite, and the positioning can be used for feeding back the position of a road camera and transmitting the position information of a target vehicle to a monitoring terminal;
s8: after the tracking and capturing task is finished, carrying out conventional detection on each module and the camera through the convolutional neural network and the detection module, and ensuring that the next task can be smoothly implemented;
the detection module compares the operation data of the camera and each module with normal operation data through a neural network and sends a detection result to the monitoring terminal;
the method comprises the steps that a license plate and characteristics of a target vehicle to be searched are input through a monitoring terminal, a road camera is captured by a convolutional neural network, the characteristic vehicle is screened and checked through the road camera to determine the target vehicle, then a plurality of groups of road cameras are matched with each other, real-time positioning and real-time images and videos of the target vehicle are transmitted to the monitoring terminal to be tracked and displayed, the monitoring terminal can determine and deny the target vehicle, positioning and tracking accuracy is improved, the target vehicle can be searched and accurately captured in the road traffic with huge traffic, the target vehicle can be rapidly and effectively captured and tracked through the intelligence of the convolutional neural network, locking timeliness and locking speed are increased, the target vehicle is prevented from being lost, and after a task is finished, the running data of the camera and each module is compared with normal running data through a detection module and the convolutional neural network, the road camera and each module are detected, the normal operation of the device is ensured, equipment is prevented from being damaged, the road condition cannot be monitored in time, and the equipment is conveniently and timely maintained.
While the invention has been described above with reference to an embodiment, various modifications may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In particular, the various features of the disclosed embodiments of the invention may be used in any combination, provided that no structural conflict exists, and the combinations are not exhaustively described in this specification merely for the sake of brevity and resource conservation. Therefore, it is intended that the invention not be limited to the particular embodiments disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.
Claims (8)
1. A road traffic target tracking method based on a deep convolutional neural network is characterized by comprising the following steps: the method comprises the following operation steps:
s1: the license plate and the characteristic of the target vehicle to be searched are input through the monitoring terminal;
s2: the monitoring terminal screens and checks the road camera in multiple angles and multiple positions through the internal convolutional neural network transmission module;
s3: carrying out target vehicle identification on a road camera through a convolutional neural network, and reading and checking vehicles with similar characteristics and similar license plates to the target vehicle;
s4: transmitting the checked target vehicle to a monitoring terminal for judgment;
s5: after the target vehicle is determined, the target vehicle is captured and locked through the target capture module and the multiple groups of road cameras, and the target vehicle is ensured to be locked in the field of view of the road cameras;
s6: after a target vehicle is captured, the captured images of the road camera are displayed on a monitoring terminal through a tracking switching module, a plurality of groups of images and a plurality of groups of images at different positions are presented through the driving of the vehicle, and the images are switched in real time;
s7: the method comprises the steps that when a target vehicle is captured and tracked, the real-time positioning of a road camera at the moment is sent to a monitoring terminal through a convolutional neural network, so that the moving track of the vehicle is displayed on the monitoring terminal;
s8: after the tracking and capturing task is finished, each module and the camera are routinely detected through the convolutional neural network and the detection module, and the next task can be smoothly implemented.
2. The method for tracking the road traffic target based on the deep convolutional neural network of claim 1, wherein the monitoring terminal in S1 inputs vehicle information and displays a target vehicle.
3. The method for tracking the road traffic target based on the deep convolutional neural network as claimed in claim 1, wherein the convolutional neural network in S2 is used for transmitting and commanding the device through an intelligent network.
4. The method for tracking the road traffic target based on the deep convolutional neural network as claimed in claim 1, wherein if the monitor terminal determines yes in S4, the method continues to step S5, and if the monitor terminal determines no, the method returns to S1 to continue to screen the vehicles, and intelligently discharge the target vehicles screened in the first pass.
5. The method for tracking the road traffic target based on the deep convolutional neural network as claimed in claim 1, wherein the S5 target capturing module is a dynamic perspective capturing module, and the target capturing module performs dynamic capturing on the target vehicle to determine the vehicle traveling route.
6. The method according to claim 1, wherein the tracking switching module in the step S6 switches video images, and the tracking switching module locks the target vehicle through a plurality of sets of road cameras and transmits a picture to the monitoring terminal.
7. The method for tracking the road traffic target based on the deep convolutional neural network as claimed in claim 1, wherein the positioning in S7 is set as a GPS/beidou satellite, and the positioning can transmit the position information of the target vehicle to the monitoring terminal for the position feedback of the road camera.
8. The method for tracking the road traffic target based on the deep convolutional neural network as claimed in claim 1, wherein the detection module compares the operation data of the camera and each module with normal operation data through a convolutional neural network, and sends the detection result to the monitoring terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210062436.8A CN114419555A (en) | 2022-01-19 | 2022-01-19 | Road traffic target tracking method based on deep convolutional neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210062436.8A CN114419555A (en) | 2022-01-19 | 2022-01-19 | Road traffic target tracking method based on deep convolutional neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114419555A true CN114419555A (en) | 2022-04-29 |
Family
ID=81274849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210062436.8A Pending CN114419555A (en) | 2022-01-19 | 2022-01-19 | Road traffic target tracking method based on deep convolutional neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114419555A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1897015A (en) * | 2006-05-18 | 2007-01-17 | 王海燕 | Method and system for inspecting and tracting vehicle based on machine vision |
| CN107105207A (en) * | 2017-06-09 | 2017-08-29 | 北京深瞐科技有限公司 | Target monitoring method, target monitoring device and video camera |
| CN108417047A (en) * | 2018-05-10 | 2018-08-17 | 杭州盈蝶科技有限公司 | A kind of vehicle location method for tracing and its system |
| CN111696365A (en) * | 2019-03-11 | 2020-09-22 | 杭州海康威视数字技术股份有限公司 | Vehicle tracking system |
| CN112132866A (en) * | 2020-09-22 | 2020-12-25 | 厦门大学 | Target object tracking method, device and equipment and computer readable storage medium |
| CN112418213A (en) * | 2020-11-06 | 2021-02-26 | 北京航天自动控制研究所 | Vehicle driving track identification method and device and storage medium |
| CN112752067A (en) * | 2019-10-30 | 2021-05-04 | 杭州海康威视系统技术有限公司 | Target tracking method and device, electronic equipment and storage medium |
-
2022
- 2022-01-19 CN CN202210062436.8A patent/CN114419555A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1897015A (en) * | 2006-05-18 | 2007-01-17 | 王海燕 | Method and system for inspecting and tracting vehicle based on machine vision |
| CN107105207A (en) * | 2017-06-09 | 2017-08-29 | 北京深瞐科技有限公司 | Target monitoring method, target monitoring device and video camera |
| CN108417047A (en) * | 2018-05-10 | 2018-08-17 | 杭州盈蝶科技有限公司 | A kind of vehicle location method for tracing and its system |
| CN111696365A (en) * | 2019-03-11 | 2020-09-22 | 杭州海康威视数字技术股份有限公司 | Vehicle tracking system |
| CN112752067A (en) * | 2019-10-30 | 2021-05-04 | 杭州海康威视系统技术有限公司 | Target tracking method and device, electronic equipment and storage medium |
| CN112132866A (en) * | 2020-09-22 | 2020-12-25 | 厦门大学 | Target object tracking method, device and equipment and computer readable storage medium |
| CN112418213A (en) * | 2020-11-06 | 2021-02-26 | 北京航天自动控制研究所 | Vehicle driving track identification method and device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110428522B (en) | Intelligent security system of wisdom new town | |
| KR101942491B1 (en) | Hybrid ai cctv mediation module device consisting of road traffic situation monitoring and real time traffic information analysis | |
| US10565680B2 (en) | Systems and methods for dewarping images | |
| US10692175B1 (en) | Systems and methods for dewarping images | |
| EP4105101A1 (en) | Monitoring system, monitoring method, and monitoring device for railway train | |
| US10438071B2 (en) | Distributed system for mining, correlating, and analyzing locally obtained traffic data including video | |
| CN104200671A (en) | Method and system for managing virtual gate based on big data platform | |
| KR102282800B1 (en) | Method for trackig multi target employing ridar and camera | |
| Sabri et al. | Low-cost intelligent surveillance system based on fast CNN | |
| US20220351515A1 (en) | Method for perceiving event tagging-based situation and system for same | |
| CN112257683A (en) | Cross-mirror tracking method for vehicle running track monitoring | |
| CN1960479A (en) | Method for tracking principal and subordinate videos by using single video camera | |
| Rezaee et al. | Deep-Transfer-learning-based abnormal behavior recognition using internet of drones for crowded scenes | |
| CN114419555A (en) | Road traffic target tracking method based on deep convolutional neural network | |
| Pavlidis et al. | Two examples of indoor and outdoor surveillance systems: motivation, design, and testing | |
| CN113076821A (en) | Event detection method and device | |
| CN112686136A (en) | Object detection method, device and system | |
| Modi | Counting Empty Parking Spots at Truck Stops Using Computer Vision. | |
| ElHakim et al. | Traffisense: A smart integrated visual sensing system for traffic monitoring | |
| Catani et al. | A framework based on vision sensors for the automatic management of exchange parking areas | |
| Rothkrantz | Crisis management using multiple camera surveillance systems. | |
| WO2023241536A1 (en) | A system and method for detecting traffic events | |
| Yang et al. | Real-time automated deep learning based detection and tracking near highway-rail grade crossing for vulnerable road users safety | |
| CN114220070A (en) | Camera system for locking object in image based on IPC | |
| Tedblad et al. | Flowity–Artificial Intelligence for traffic analysis |
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: 20220429 |
|
| RJ01 | Rejection of invention patent application after publication |