CN113257003A - Traffic lane-level traffic flow counting system, method, device and medium thereof - Google Patents
Traffic lane-level traffic flow counting system, method, device and medium thereof Download PDFInfo
- Publication number
- CN113257003A CN113257003A CN202110518814.4A CN202110518814A CN113257003A CN 113257003 A CN113257003 A CN 113257003A CN 202110518814 A CN202110518814 A CN 202110518814A CN 113257003 A CN113257003 A CN 113257003A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- image data
- module
- traffic flow
- counting
- 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
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a traffic road lane-level traffic flow counting system, which comprises: the remote data fetching module: acquiring image data from various video sources, and pushing the image data into a corresponding global image data queue; a vehicle detection module: detecting and identifying vehicles and vehicle types in each frame of image in the global image data queue; a vehicle tracking module: carrying out vehicle multi-target tracking aiming at the detection identification data of the vehicle detection module; the traffic flow counting module: and carrying out track analysis based on the tracking data of each vehicle, and counting the traffic flow at the current moment. The invention also provides a method, equipment and medium for counting the lane-level traffic flow of the traffic lane. According to the invention, through the traffic road camera data, the deep learning target detection and tracking method is used, the automatic traffic flow statistics is realized, the statistics granularity is improved, the deployment is simple, and the cost is low.
Description
Technical Field
The invention relates to the technical field of intelligent urban traffic, in particular to a traffic lane-level traffic flow counting system, a method, equipment and a medium thereof, and particularly relates to a traffic lane-level traffic flow counting system based on an image video algorithm, a method, equipment and a medium thereof.
Background
The automatic traffic flow statistics has wide and profound application in the intelligent traffic field, and currently, there are two general implementation schemes: one is based on hardware devices such as high-precision sensors or laser radars, and the hardware devices and matched infrastructures need to be deployed on both sides of a road, so that the cost is high, the hardware devices cannot be borne by many places, and the universality is not high. The other is based on computer vision, and uses the existing traffic camera data, but also has the disadvantages of insufficient statistical granularity, difficult deployment and high cost.
Through retrieval, patent document CN110009023A discloses a traffic flow statistical method in intelligent traffic, which is composed of two parts, namely a vehicle detection method and a vehicle tracking method, by using SSD and ResNet in a neural network and a traditional target tracking CamShift algorithm. The vehicle detection method comprises the steps of firstly establishing an SSD network, obtaining a plurality of feature maps with different sizes by the SSD, adopting different aspect ratios for default frames on the same feature layer, enhancing the robustness of the default frames to the shape of an object, and performing SSD training while regressing the position and the target type. The vehicle tracking method is to adopt a continuous self-adaptive expected movement algorithm to track the vehicle identified by the first frame of the single-time mobile visual network detector. The disadvantages of the prior art are that the SSD network is adopted with slower speed and not high enough precision, and the precision of the camshift method is not high.
Patent document CN110599781A discloses an intelligent traffic flow statistics and identification system, which includes a video source, a vehicle statistics system and a traffic signal, wherein the video source is video information accessed by traffic video, and the vehicle statistics system includes AI identification of vehicles, image processing and generation of traffic flow information data reports. The vehicle flow statistical system of the intelligent city is combined with the modern artificial intelligence technology and the image processing technology to develop a set of vehicle statistical algorithm which is realized by a traffic camera at an intersection, and the purpose of obtaining and obtaining vehicle information at the intersection according to images can be realized; the vehicles passing through are counted and counted by the vehicle counting module, the time for the vehicles to pass in and out is calculated to reach the time for effectively controlling the traffic lights, and the data result is returned to the signal machine for self-adaptive control and adjustment of the signal lamps at the intersection. The prior art has the disadvantages that only a target detection-based technology is used, the same vehicle is easily subjected to repeated statistics, and the precision is low; and this prior art requires additional signals, requires additional installation equipment, and is costly.
Therefore, it is necessary to develop and design a system and a method capable of automatically counting the number of vehicles entering and exiting at different time intervals and analyzing the traffic conditions at intersections and road sections.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a traffic lane level traffic flow counting system, a method, equipment and a medium thereof, which are suitable for real-time counting of traffic flow at the level of a traffic lane at an intersection, can automatically count the number of various vehicles entering and exiting at different time periods, analyze the traffic conditions of the intersection and a road section, and provide reference basis for traffic scheduling and road condition optimization.
The invention provides a traffic road lane-level traffic flow counting system, which comprises:
the remote data fetching module: acquiring image data from various video sources, and pushing the image data into a corresponding global image data queue;
a vehicle detection module: detecting and identifying vehicles and vehicle types in each frame of image in the global image data queue;
a vehicle tracking module: carrying out vehicle multi-target tracking aiming at the detection identification data of the vehicle detection module;
the traffic flow counting module: and carrying out track analysis based on the tracking data of each vehicle, and counting the traffic flow at the current moment.
Preferably, the remote data fetching module starts a corresponding method for different video sources to obtain image data, processes the obtained image data, and pushes the image data into a corresponding global image data queue according to the resolution of the image.
Preferably, the vehicle detection module detects and identifies the vehicle and the vehicle type in each frame of image using the YOLOv3 model.
Preferably, the vehicle detection module comprises a first sub-module and a second sub-module; the first sub-module acquires images from the image data queue, packs the image data in batches and pushes the image data into an internal data exchange queue according to the Batch size of the detection model; the second sub-module obtains a batch of images from the internal data exchange queue for processing.
Preferably, the vehicle tracking module uses the Deep Sort model to assign the same ID to the same vehicle in the image data queue, so as to know the position of the certain vehicle at a certain moment in real time.
Preferably, the traffic flow counting module counts the traffic flow of the road level and the lane level in real time according to different requirements, and counts the number of each vehicle type and the traffic flow of each direction of the intersection in real time.
The invention provides a method for counting lane-level traffic flow of a traffic lane, which comprises the following steps:
step S1: acquiring image data from various video sources, and pushing the image data into a corresponding global image data queue;
step S2: detecting and identifying vehicles and vehicle types in each frame of image in the global image data queue;
step S3: carrying out vehicle multi-target tracking aiming at the detection identification data of the vehicle detection module;
step S4: and carrying out track analysis based on the tracking data of each vehicle, and counting the traffic flow at the current moment.
Preferably, step S3 includes the steps of:
step S3.1: acquiring 128-dimensional features of each vehicle detection frame by using a vehicle weight recognition model;
step S3.2: predicting the position of the vehicle at the moment based on the vehicle track by using Kalman filtering;
step S3.3: and associating the vehicle of the current frame with the vehicle of the previous frame according to the Mahalanobis distance and the appearance characteristics.
According to the present invention, a computer-readable storage medium is provided, in which a computer program is stored, which, when being executed by a processor, carries out the above-mentioned method steps.
According to the invention, the traffic lane-level traffic flow counting device comprises the traffic lane-level traffic flow counting system or the computer readable storage medium storing the computer program.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the invention, through the traffic road camera data, the deep learning target detection and tracking method is used, the automatic traffic flow statistics is realized, the statistics granularity is improved, the deployment is simple, and the cost is low.
2. According to the invention, the traffic flow is automatically counted by using the Yolov3 target detection algorithm, so that the measuring speed is higher and the measuring precision is higher.
3. The invention adopts a multi-process concurrency technology and a multi-image simultaneous reasoning technology, and can support more image data acquisition heads at the same time.
4. The invention does not need additional installation equipment, has low cost and is expandable, and comprises the expansion of vehicle types, the replacement of the number of model types and the expansion of the number of models.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic view of the working process of the traffic lane level traffic flow counting system of the present invention;
FIG. 2 is a schematic diagram of the remote data fetch module according to the present invention;
FIG. 3 is a schematic flow chart of the vehicle detection module of the present invention;
FIG. 4 is a schematic flow chart of the vehicle tracking module of the present invention;
fig. 5 is a schematic view of the work flow of the traffic flow counting module according to the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
As shown in fig. 1, the present invention provides a traffic lane-level traffic flow counting system, comprising:
as shown in fig. 2, the remote data fetching module obtains image data from a plurality of video sources, and pushes the image data into a corresponding global image data queue; and the remote data fetching module starts corresponding methods for different video sources to acquire image data, processes the acquired image data and pushes the processed image data into a corresponding global image data queue according to the resolution of the image. The method comprises the steps of obtaining local videos, local pictures, remote video streams, remote images and the like, finally simply processing the obtained picture data, pushing the picture data into a corresponding global picture data queue according to the resolution of the pictures, and enabling the picture data to be consumed and used by a following module.
As shown in fig. 3, the vehicle detection module detects and identifies the vehicle and the vehicle type in each frame of image in the global image data queue; the vehicle and the vehicle type in each frame of image are detected and identified, and a model of YOLOv3 which is a model with both accuracy and running speed is used. It can be divided into two sub-modules:
acquiring images from the image data queue, and packaging and pushing the image data into an internal data exchange queue in batches according to the Batch size of the detection model; and the second sub-module acquires a batch of images from the internal data exchange queue for processing, and finally writes the result into the global detection data queue.
As shown in fig. 4, the vehicle tracking module performs vehicle multi-target tracking on the detection identification data of the vehicle detection module; the vehicle tracking module uses a Deep Sort model to assign the same ID to the same vehicle in the image data queue, and knows the position of the certain vehicle at a certain moment in real time. Based on which further vehicle behaviour analysis can be made.
The vehicle tracking module first obtains inspection data for the image from the global inspection data queue and then processes the data. And finally writing the tracking result into a global tracking data queue.
As shown in fig. 5, the traffic flow counting module performs trajectory analysis based on the tracking data of each vehicle, and counts the traffic flow at the current time. The traffic flow counting module counts the traffic flow of road levels and lane levels in real time according to different requirements, and counts the number of each vehicle type and the traffic flow of each direction of the intersection in real time.
During processing, firstly, a mask file of each road camera is read, the file contains an area needing attention on a road, and a cross-reference line is set; it stores the trajectory of each vehicle present in these zones in a hash table based on the vehicle ID, and increments the number of vehicles when there are vehicles crossing the line.
The invention provides a method for counting lane-level traffic flow of a traffic lane, which comprises the following steps:
step S1: and acquiring image data from various video sources, and pushing the image data into a corresponding global image data queue.
Step S2: and detecting and identifying vehicles and vehicle types in each frame of image in the global image data queue.
Step S2.1: and (4) carrying out image preprocessing, and carrying out processing such as resize, channel conversion, normalization and the like on the image.
Step S2.2: and picture reasoning, wherein the TensorRT is used for reasoning on the GPU for the image.
Step S2.3: and (4) performing prediction post-processing, namely analyzing the predicted result to obtain the results of the detection frames.
Step S3: and carrying out vehicle multi-target tracking aiming at the detection identification data of the vehicle detection module.
Step S3.1: the 128-dimensional features of each vehicle detection frame are obtained using a vehicle weight recognition model.
Step S3.2: the position of the vehicle at this moment is predicted based on its trajectory using kalman filtering.
Step S3.3: and associating the vehicle of the current frame with the vehicle of the previous frame according to the Mahalanobis distance and the appearance characteristics.
Step S4: and carrying out track analysis based on the tracking data of each vehicle, and counting the traffic flow at the current moment.
According to the present invention, a computer-readable storage medium is provided, in which a computer program is stored, which, when being executed by a processor, carries out the above-mentioned method steps.
According to the invention, the traffic lane-level traffic flow counting device comprises the traffic lane-level traffic flow counting system or the computer readable storage medium storing the computer program.
The invention can realize the following steps by the arrangement:
1. the accuracy is high: under the scene of 200 multi-path cameras, the traffic flow statistical accuracy rate exceeds 97%.
2. And (3) real-time monitoring: based on the counting of target detection and tracking, one camera can process 15 frames of data per second, and the requirement of real-time property is completely met.
3. Lane grade: the traffic flow can be counted for each lane in real time according to different masks, and the number of vehicles in each direction at the intersection can be mastered at any time.
4. Vehicle type grade: the traffic flow data statistics of three motor vehicles of a truck, a bus and a car and two non-motor vehicles of a bicycle and an electric vehicle can be provided, and 5 compartments are counted.
5. Easy deployment: the service is packaged into a container by using a docker virtualization technology, and the service can be deployed on any machine after a configuration file is simply modified.
6. The cost is low: the yolov3 target detection method and the deepsort multi-target tracking method which are high in efficiency and accuracy in deep learning are used, the operation efficiency of the model is further improved by using model pruning and the English giant TensrT technology, and real-time traffic flow statistics of up to 16 cameras can be supported on one 1080 video card, so that the same equipment can support more camera services.
7. And (3) expandable: the sub-modules are decoupled, so that the overall system is very easily scalable, including the extension of vehicle types, the replacement of the number of model types, and the extension of the number of models.
8. Support for multiple resolutions: the system can count different resolutions of the accessed camera, and provide corresponding resources for each resolution, thereby achieving the optimal performance state.
Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, units provided by the present invention as pure computer readable program code, the system and its various devices, modules, units provided by the present invention can be fully implemented by logically programming method steps in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system and various devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units included in the system for realizing various functions can also be regarded as structures in the hardware component; means, modules, units for performing the various functions may also be regarded as structures within both software modules and hardware components for performing the method.
In the description of the present application, it is to be understood that the terms "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience in describing the present application and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present application.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A traffic lane-level traffic flow counting system, comprising:
the remote data fetching module: acquiring image data from various video sources, and pushing the image data into a corresponding global image data queue;
a vehicle detection module: detecting and identifying vehicles and vehicle types in each frame of image in the global image data queue;
a vehicle tracking module: carrying out vehicle multi-target tracking aiming at the detection identification data of the vehicle detection module;
the traffic flow counting module: and carrying out track analysis based on the tracking data of each vehicle, and counting the traffic flow at the current moment.
2. The system according to claim 1, wherein the remote data-fetching module starts corresponding methods for different video sources to obtain image data, processes the obtained image data, and pushes the processed image data into the corresponding global image data queue according to the resolution of the image.
3. The system of claim 1, wherein the vehicle detection module detects and identifies vehicles and vehicle types in each frame of image using a YOLOv3 model.
4. The system of claim 3, wherein the vehicle detection module includes a first sub-module and a second sub-module;
the first sub-module acquires images from the image data queue, packs the image data in batches and pushes the image data into an internal data exchange queue according to the Batch size of the detection model;
and the second sub-module acquires a batch of images from the internal data exchange queue for processing.
5. The system of claim 1, wherein the vehicle tracking module uses Deep Sort model to assign the same ID to the same vehicle in the image data queue, and knows the position of the vehicle at a certain time in real time.
6. The system of claim 1, wherein the traffic flow counting module counts the traffic flow at road level and lane level in real time, and counts the number of each vehicle type and the traffic flow in each direction of the intersection in real time according to different requirements.
7. A method for counting lane-level traffic flow of a traffic lane is characterized by comprising the following steps:
step S1: acquiring image data from various video sources, and pushing the image data into a corresponding global image data queue;
step S2: detecting and identifying vehicles and vehicle types in each frame of image in the global image data queue;
step S3: carrying out vehicle multi-target tracking aiming at the detection identification data of the vehicle detection module;
step S4: and carrying out track analysis based on the tracking data of each vehicle, and counting the traffic flow at the current moment.
8. The method according to claim 7, wherein the step S3 includes the steps of:
step S3.1: acquiring 128-dimensional features of each vehicle detection frame by using a vehicle weight recognition model;
step S3.2: predicting the position of the vehicle at the moment based on the vehicle track by using Kalman filtering;
step S3.3: and associating the vehicle of the current frame with the vehicle of the previous frame according to the Mahalanobis distance and the appearance characteristics.
9. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any one of claims 7 to 8.
10. A road-to-road lane-level traffic counting apparatus comprising the road-to-road lane-level traffic counting system of any one of claims 1 to 6 or the computer-readable storage medium of claim 9 having a computer program stored thereon.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110518814.4A CN113257003A (en) | 2021-05-12 | 2021-05-12 | Traffic lane-level traffic flow counting system, method, device and medium thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110518814.4A CN113257003A (en) | 2021-05-12 | 2021-05-12 | Traffic lane-level traffic flow counting system, method, device and medium thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113257003A true CN113257003A (en) | 2021-08-13 |
Family
ID=77223210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110518814.4A Pending CN113257003A (en) | 2021-05-12 | 2021-05-12 | Traffic lane-level traffic flow counting system, method, device and medium thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113257003A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114119648A (en) * | 2021-11-12 | 2022-03-01 | 史缔纳农业科技(广东)有限公司 | Pig counting method for fixed channel |
| CN114241786A (en) * | 2021-12-21 | 2022-03-25 | 中国科学院大气物理研究所 | System and method for recognizing vehicle types of multiple roads and monitoring traffic flow |
| CN117437792A (en) * | 2023-12-20 | 2024-01-23 | 中交第一公路勘察设计研究院有限公司 | Real-time road traffic state monitoring method, device and system based on edge calculation |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130088600A1 (en) * | 2011-10-05 | 2013-04-11 | Xerox Corporation | Multi-resolution video analysis and key feature preserving video reduction strategy for (real-time) vehicle tracking and speed enforcement systems |
| CN103402044A (en) * | 2013-08-07 | 2013-11-20 | 重庆大学 | Target recognition and tracking system based on multi-source video integration |
| CN108898842A (en) * | 2018-07-02 | 2018-11-27 | 武汉大学深圳研究院 | A kind of high efficiency encoding method and its system of multi-source monitor video |
| CN208271388U (en) * | 2017-12-06 | 2018-12-21 | 西安市交通信息中心 | A kind of volume of traffic acquisition system based on video monitoring |
| CN109919072A (en) * | 2019-02-28 | 2019-06-21 | 桂林电子科技大学 | Fine vehicle type recognition and flow statistics method based on deep learning and trajectory tracking |
| CN110472496A (en) * | 2019-07-08 | 2019-11-19 | 长安大学 | A kind of traffic video intelligent analysis method based on object detecting and tracking |
| CN111145213A (en) * | 2019-12-10 | 2020-05-12 | ***股份有限公司 | Target tracking method, device and system and computer readable storage medium |
| CN111554105A (en) * | 2020-05-29 | 2020-08-18 | 浙江科技学院 | Intelligent traffic identification and statistics method for complex traffic intersection |
| CN111667514A (en) * | 2020-05-22 | 2020-09-15 | 浙江工贸职业技术学院 | Quick and accurate vehicle tracking system |
| CN112101433A (en) * | 2020-09-04 | 2020-12-18 | 东南大学 | Automatic lane-dividing vehicle counting method based on YOLO V4 and DeepsORT |
| CN112750150A (en) * | 2021-01-18 | 2021-05-04 | 西安电子科技大学 | Vehicle flow statistical method based on vehicle detection and multi-target tracking |
-
2021
- 2021-05-12 CN CN202110518814.4A patent/CN113257003A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130088600A1 (en) * | 2011-10-05 | 2013-04-11 | Xerox Corporation | Multi-resolution video analysis and key feature preserving video reduction strategy for (real-time) vehicle tracking and speed enforcement systems |
| CN103402044A (en) * | 2013-08-07 | 2013-11-20 | 重庆大学 | Target recognition and tracking system based on multi-source video integration |
| CN208271388U (en) * | 2017-12-06 | 2018-12-21 | 西安市交通信息中心 | A kind of volume of traffic acquisition system based on video monitoring |
| CN108898842A (en) * | 2018-07-02 | 2018-11-27 | 武汉大学深圳研究院 | A kind of high efficiency encoding method and its system of multi-source monitor video |
| CN109919072A (en) * | 2019-02-28 | 2019-06-21 | 桂林电子科技大学 | Fine vehicle type recognition and flow statistics method based on deep learning and trajectory tracking |
| CN110472496A (en) * | 2019-07-08 | 2019-11-19 | 长安大学 | A kind of traffic video intelligent analysis method based on object detecting and tracking |
| CN111145213A (en) * | 2019-12-10 | 2020-05-12 | ***股份有限公司 | Target tracking method, device and system and computer readable storage medium |
| CN111667514A (en) * | 2020-05-22 | 2020-09-15 | 浙江工贸职业技术学院 | Quick and accurate vehicle tracking system |
| CN111554105A (en) * | 2020-05-29 | 2020-08-18 | 浙江科技学院 | Intelligent traffic identification and statistics method for complex traffic intersection |
| CN112101433A (en) * | 2020-09-04 | 2020-12-18 | 东南大学 | Automatic lane-dividing vehicle counting method based on YOLO V4 and DeepsORT |
| CN112750150A (en) * | 2021-01-18 | 2021-05-04 | 西安电子科技大学 | Vehicle flow statistical method based on vehicle detection and multi-target tracking |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114119648A (en) * | 2021-11-12 | 2022-03-01 | 史缔纳农业科技(广东)有限公司 | Pig counting method for fixed channel |
| CN114241786A (en) * | 2021-12-21 | 2022-03-25 | 中国科学院大气物理研究所 | System and method for recognizing vehicle types of multiple roads and monitoring traffic flow |
| CN117437792A (en) * | 2023-12-20 | 2024-01-23 | 中交第一公路勘察设计研究院有限公司 | Real-time road traffic state monitoring method, device and system based on edge calculation |
| CN117437792B (en) * | 2023-12-20 | 2024-04-09 | 中交第一公路勘察设计研究院有限公司 | Real-time road traffic state monitoring method, device and system based on edge calculation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11941887B2 (en) | Scenario recreation through object detection and 3D visualization in a multi-sensor environment | |
| CN113257003A (en) | Traffic lane-level traffic flow counting system, method, device and medium thereof | |
| WO2019145018A1 (en) | System, device and method for detecting abnormal traffic events in a geographical location | |
| EP2709066A1 (en) | Concept for detecting a motion of a moving object | |
| CN112241969A (en) | Target detection tracking method and device based on traffic monitoring video and storage medium | |
| JP7115502B2 (en) | Object state identification device, object state identification method, computer program for object state identification, and control device | |
| US11900690B2 (en) | Apparatus, method, and computer program for identifying state of signal light, and controller | |
| CN116434159A (en) | Traffic flow statistics method based on improved YOLO V7 and Deep-Sort | |
| JP7226368B2 (en) | Object state identification device | |
| CN117152971A (en) | AI traffic signal optimization method based on high-altitude panoramic video | |
| CN113808418A (en) | Road condition information display system, method, vehicle, computer device and storage medium | |
| CN114627409A (en) | Method and device for detecting abnormal lane change of vehicle | |
| Weimer et al. | Gpu architecture for stationary multisensor pedestrian detection at smart intersections | |
| EP2709065A1 (en) | Concept for counting moving objects passing a plurality of different areas within a region of interest | |
| Nalavde et al. | Driver assistant services using ubiquitous smartphone | |
| CN113917453A (en) | Multi-sensor fusion method based on radar and video | |
| ElHakim et al. | Traffisense: A smart integrated visual sensing system for traffic monitoring | |
| Song et al. | Method of Vehicle Behavior Analysis for Real-Time Video Streaming Based on Mobilenet-YOLOV4 and ERFNET | |
| Wang et al. | Traffic light and vehicle signal recognition with high dynamic range imaging and deep learning | |
| Moayed et al. | Surveillance-based collision-time analysis of road-crossing pedestrians | |
| CN114898296B (en) | Bus lane occupation detection method based on millimeter wave radar and vision fusion | |
| Wan et al. | Road traffic monitoring using image processing—A survey of systems, techniques and applications | |
| US20240020964A1 (en) | Method and device for improving object recognition rate of self-driving car | |
| Das | Vision-Based Lane and Vehicle Detection: A First Step Toward Autonomous Unmanned Vehicle | |
| Mandal | Artificial intelligence enabled automatic traffic monitoring system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210813 |