[go: up one dir, main page]

CN113066106A - Vehicle speed measuring method based on aerial robot mobile vision - Google Patents

Vehicle speed measuring method based on aerial robot mobile vision Download PDF

Info

Publication number
CN113066106A
CN113066106A CN202110414461.3A CN202110414461A CN113066106A CN 113066106 A CN113066106 A CN 113066106A CN 202110414461 A CN202110414461 A CN 202110414461A CN 113066106 A CN113066106 A CN 113066106A
Authority
CN
China
Prior art keywords
vehicle
distance
image
aerial robot
images
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
Application number
CN202110414461.3A
Other languages
Chinese (zh)
Other versions
CN113066106B (en
Inventor
王鹏
张红生
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN202110414461.3A priority Critical patent/CN113066106B/en
Publication of CN113066106A publication Critical patent/CN113066106A/en
Application granted granted Critical
Publication of CN113066106B publication Critical patent/CN113066106B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
    • G08G1/054—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed photographing overspeeding vehicles
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06T2207/30236—Traffic on road, railway or crossing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/08—Detecting or categorising vehicles

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

本发明公开了一种基于空中机器人移动视觉的车速测量方法，通过获取空中机器人所拍摄的视频图像画面，可以获取到车辆的运动轨迹；同时，通过检测车辆所在道路上的车道线，计算出图上距离，求得实际距离和图上距离的比值，算出车辆的实际运动距离；从而最终计算出画面内所有车辆的瞬时速度。本发明方法克服了传统固定测速设备灵活性不强、造价高昂、安装困难等缺点，摆脱了现有空中车辆测速方案对空中机器人飞行姿态的高要求，机动性强，对飞行姿态和飞手的要求不高，能够实时地监测地面车辆的速度，使用方便。The invention discloses a vehicle speed measurement method based on the moving vision of an aerial robot. The motion trajectory of the vehicle can be obtained by acquiring the video image pictures taken by the aerial robot; at the same time, by detecting the lane lines on the road where the vehicle is located, a graph is calculated. The upper distance is obtained, the ratio of the actual distance and the distance on the map is obtained, and the actual moving distance of the vehicle is calculated; thus the instantaneous speed of all vehicles in the screen is finally calculated. The method of the invention overcomes the shortcomings of the traditional fixed speed measuring equipment, such as inflexibility, high cost, difficult installation, etc., and gets rid of the high requirements for the flying attitude of the aerial robot in the existing aerial vehicle speed measuring scheme, and has strong maneuverability. The requirements are not high, the speed of the ground vehicle can be monitored in real time, and the use is convenient.

Description

Vehicle speed measuring method based on aerial robot mobile vision
Technical Field
The invention belongs to the technical field of computer vision, and particularly relates to a vehicle speed measuring method.
Background
The existing vehicle speed measurement is generally based on bayonet speed measurement mostly, the speed measurement is carried out by combining a speed measuring radar basically, the equipment can only shoot and obtain evidence for the vehicle running at an overspeed aiming at a specific position or a specific interval, and the equipment has great limitation.
Because the traditional method for measuring the vehicle speed based on the fixed speed measuring camera and the speed measuring instrument has the defects of limited snapshot area, complex equipment deployment, high equipment manufacturing cost, difficulty in timely repair after damage and the like, and the positions of most speed measuring equipment are known, a vehicle driver can change the vehicle speed in a targeted manner so as to avoid monitoring of the speed measuring equipment. Therefore, the vehicle speed measurement by using the vision of the aerial robot has great practical value and commercial value.
Vehicle speed measurement based on vision of an aerial robot is a challenging and significant task in current intelligent traffic management. When the aerial robot operates, real-time monitoring pictures can be provided for ground personnel such as traffic police and the like, so that the ground personnel can find out drivers violating regulations in a large area. Although the aerial robots can help traffic police to find out dangerous drivers and give warnings and punishments, the aerial robots are less applied to traffic monitoring and management scenes for vehicle speed measurement at present, and the snapshot and evidence collection of illegal violation phenomena such as overspeed and the like still mainly depend on a fixed speed measurement camera and a speed measurement instrument.
The patent [ a method and a monitoring system for monitoring overspeed driving by using an unmanned aerial vehicle, and an unmanned aerial vehicle, CN 111583670 a ] proposes a method and a monitoring system for monitoring overspeed driving by using an unmanned aerial vehicle, and an unmanned aerial vehicle. The displacement of the unmanned aerial vehicle is obtained through the ground speed of the unmanned aerial vehicle within a period of time, meanwhile, the displacement of the target vehicle in a picture within the period of time is calculated, the relative displacement of the vehicle and the unmanned aerial vehicle is calculated according to the ground angle of a pod on the unmanned aerial vehicle and the height of the unmanned aerial vehicle, so that the ground displacement of the target vehicle is calculated, and finally the speed of the vehicle is calculated. This method requires that the control of the flight crew of the drone is required to ensure that the drone has a fixed flying speed, flying height and pod-to-ground angle while the drone is in operation, which puts high demands on the control level of the flight crew. Therefore, the method has great limitation and cannot adapt to complicated and variable environments.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a vehicle speed measuring method based on aerial robot mobile vision, which can acquire the motion trail of a vehicle by acquiring the video image picture shot by the aerial robot; meanwhile, the distance on the graph is calculated by detecting the lane line on the road where the vehicle is located, the ratio of the actual distance to the distance on the graph is obtained, and the actual movement distance of the vehicle is calculated; so as to finally calculate the instantaneous speed of all vehicles in the picture. The method overcomes the defects of low flexibility, high manufacturing cost, difficult installation and the like of the traditional fixed speed measuring equipment, overcomes the high requirement of the existing speed measuring scheme of the aerial vehicle on the flight attitude of the aerial robot, has strong maneuverability and low requirements on the flight attitude and the flying hand, can monitor the speed of the ground vehicle in real time, and is convenient to use.
The technical scheme adopted by the invention for solving the technical problem comprises the following steps:
step 1: collecting data;
acquiring images of different vehicles and roads by using an aerial robot under mobile vision, and labeling the vehicles and the roads in the acquired images;
performing data augmentation on the marked image;
then, a training set is formed by the image with augmented data;
step 2: model training
Setting training parameters by adopting a neural network model, and training the neural network model by utilizing a training set;
the trained neural network model can detect vehicles and roads in the input image;
and step 3: measuring the vehicle speed;
step 3-1: intercepting an image every N frames of videos collected by the aerial robot, inputting the images into the neural network model trained in the step 2, and detecting vehicles and roads in the images;
step 3-2: calculating the distance between two adjacent lane lines in the image, and calculating the average value d of the distances between the lane lines in the imagespic；
Step 3-3: obtaining the actual distance d between two adjacent lane linesactual；
Step 3-4: calculating the ratio of the average distance of the lane line in the image to the actual distance of the lane line
Figure BDA0003025305160000021
Step 3-5: calculating the interval time of every N frames of video
Figure BDA0003025305160000022
f is the video frame rate;
step 3-6: calculating the moving distance s of the vehicle on the adjacent N frames of imagespicAnd calculating the actual distance of the vehicle moving in the interval time delta t
Figure BDA0003025305160000023
Step 3-7: calculating a speed of movement of a vehicle
Figure BDA0003025305160000024
And 4, step 4: and 3, measuring the speed of the vehicle in the video by adopting the step 3, and marking the vehicle as red to warn an operator when the speed of the vehicle exceeds a set value.
Preferably, the data augmentation method includes translation, rotation, flipping, cropping, scaling, adjusting hue, adjusting lightness, adjusting saturation, perspective transformation, image fusion, and image occlusion.
Preferably, the neural network model is YOLOv 5.
Preferably, N ═ 10.
The invention has the following beneficial effects:
1. the method overcomes the defects of low flexibility, high manufacturing cost, difficult installation and the like of the traditional fixed speed measuring equipment, and gets rid of the high requirement of the existing speed measuring scheme of the aerial vehicle on the flight attitude of the aerial robot.
2. The method only depends on the vision of the aerial robot, has strong maneuverability and low requirements on flight attitude and flyers, can monitor the speed of the ground vehicle in real time, and is convenient to use.
Detailed Description
The invention is further illustrated by the following examples.
The invention provides a method for measuring the speed of a ground vehicle by using an aerial robot to visually measure the speed of the ground vehicle, which can detect the vehicle and a lane line through a neural network, calculate the distance on a lane line graph and calculate the ratio of the distance on the graph to the actual distance to solve the actual movement distance of the ground vehicle, and finally calculate the speed of the ground vehicle and track and monitor the speed of the ground vehicle.
A vehicle speed measuring method based on aerial robot mobile vision comprises the following steps:
step 1: collecting data;
acquiring images of different vehicles and roads by using an aerial robot under mobile vision, and labeling the vehicles and the roads in the acquired images;
performing data augmentation on the marked image;
then, a training set is formed by the image with augmented data;
step 2: model training
Setting training parameters by adopting a neural network model, and training the neural network model by utilizing a training set; the method comprises the following specific steps:
(1) setting a neural network; initializing model weight, an optimization function and a loss function, and setting initial training parameters of a neural network;
(2) pre-training a neural network; and the classification performance of the neural network on the images is improved by utilizing an open-source image data set pre-training model.
(3) Importing a training set; importing the data subjected to data preprocessing into a neural network model;
(4) training a neural network; optimizing and updating the network weight through a loss function and an optimization function by using the marked training data, updating the model parameters, and continuously iterating;
(5) modifying the hyper-parameters and optimizing the model; and (3) modifying the hyper-parameters, adjusting the threshold value, continuously comparing the experimental results and selecting the neural network model with the best effect.
The trained neural network model can detect vehicles and roads in the input image;
and step 3: measuring the vehicle speed;
step 3-1: intercepting an image every N frames of videos collected by the aerial robot, inputting the images into the neural network model trained in the step 2, and detecting vehicles and roads in the images;
step 3-2: calculating the distance between two adjacent lane lines in the image, and calculating the average value d of the distances between the lane lines in the imagespic；
Step 3-3: obtaining the actual distance d between two adjacent lane linesactual；
Step 3-4: calculating the ratio of the average distance of the lane line in the image to the actual distance of the lane line
Figure BDA0003025305160000041
Step 3-5: calculating the interval time of every N frames of video
Figure BDA0003025305160000042
f is the video frame rate;
step 3-6: calculating the moving distance s of the vehicle on the adjacent N frames of imagespicAnd calculating the actual distance of the vehicle moving in the interval time delta t
Figure BDA0003025305160000043
Step 3-7: calculating a speed of movement of a vehicle
Figure BDA0003025305160000044
And 4, step 4: and 3, measuring the speed of the vehicle in the video by adopting the step 3, and marking the vehicle as red to warn an operator when the speed of the vehicle exceeds a set value.
The specific embodiment is as follows:
1. collecting data
The aerial robot is used for shooting road surface images of expressways, provincial roads and the like, and a large amount of image data of various roads based on the moving vision of the aerial robot is obtained. And the labelme software is used for carrying out frame selection and marking on the vehicles and the lane lines in the image.
2. Data pre-processing
Because vehicle occlusion, vehicle lane changing and lane line pressing, inconsistency of flight direction and road direction of the aerial robot, illumination on a shot picture and the like can occur frequently, data amplification needs to be carried out on data before the neural network training is started in order to ensure the accuracy and precision of model training. The method adopts the means of translation, rotation, turnover, cutting, scaling, adjustment of hue, brightness, saturation, perspective transformation, image fusion and image shielding, and combines the means to process the image so as to simulate the road conditions under different conditions and increase data. And forming a training set after data augmentation.
3. Model training
The method of YOLOv5 was used to train and detect vehicles and lane markings. And setting the detection threshold value to be 0.5, namely, the detected frame and the real frame IOU are greater than 0.5, namely, the vehicle and the lane line can be detected. During training, the sizes of all training image data are adjusted to 448 multiplied by 448 to pre-train the features to extract the neural network, then the size of the input training image data is increased to 640 multiplied by 640, the marked training data is utilized, network weights are optimized and updated through a loss function and an optimization function, model parameters are updated, iteration is carried out continuously, and then the neural network is subjected to fine tuning by using a detection data set. Continuously modifying the hyper-parameters, optimizing the model, adjusting the threshold value by modifying the hyper-parameters, continuously comparing the experimental results and selecting the neural network model with the best effect.
4. Vehicle speed measurement
(1) The trained neural network model can detect the lane lines and vehicles in the road image shot by the aerial robot.
(2) Intercepting one image frame every 10 frames to obtain a video imageCalculating the distance on the image of the adjacent lane lines and calculating the average value d of the distance on the imagepic。
(3) Looking up the national standard or provincial standard requirements of the roads of the flying ground of the aerial robot to obtain the actual distance d between the adjacent lane linesactual。
(4) Calculating the ratio of the distance on the lane line graph to the actual distance
Figure BDA0003025305160000051
(5) Reading the video frame rate information to obtain the frame rate f of video shooting, and calculating the interval time of every 10 frames of the video
Figure BDA0003025305160000052
(6) Calculating the distance s of the vehicle moving on the adjacent 10 frames of imagespicDetermining the actual distance of the vehicle moving within Δ t
Figure BDA0003025305160000053
(7) Calculating a speed of movement of a vehicle
Figure BDA0003025305160000054
5. By using the method, the position of the vehicle is tracked in the video frame, the motion track of the vehicle in the picture within a certain period of time is obtained, the vehicle is tracked and monitored, and when the speed of the vehicle in the picture exceeds a set value, the vehicle is marked red to warn an operator.

Claims (4)

1.一种基于空中机器人移动视觉的车速测量方法，其特征在于，包括以下步骤：1. a vehicle speed measurement method based on aerial robot mobile vision, is characterized in that, comprises the following steps: 步骤1：收集数据；Step 1: Collect data; 利用空中机器人在移动视觉下采集不同车辆和道路的图像，并对采集到的图像中的车辆和道路进行标注；Using aerial robots to collect images of different vehicles and roads under mobile vision, and label the vehicles and roads in the collected images; 对标注后的图像进行数据增广；Perform data augmentation on the labeled images; 再用数据增广后的图像形成训练集；Then use the augmented images to form a training set; 步骤2：模型训练Step 2: Model Training 采用神经网络模型，设置训练参数，利用训练集对神经网络模型进行训练；Use the neural network model, set the training parameters, and use the training set to train the neural network model; 训练完成的神经网络模型能够检测输入图像中的车辆和道路；The trained neural network model is able to detect vehicles and roads in the input image; 步骤3：车速测量；Step 3: Vehicle speed measurement; 步骤3-1：对空中机器人采集的视频，每隔N帧，截取一幅图像，输入步骤2训练完成的神经网络模型，检测出图像中的车辆和道路；Step 3-1: For the video collected by the aerial robot, intercept an image every N frames, input the neural network model trained in step 2, and detect vehicles and roads in the image; 步骤3-2：计算出图像中两两相邻的车道线在图像中的距离，并求出多幅图像中车道线在图像中距离的平均值dpic；Step 3-2: Calculate the distance between the adjacent lane lines in the image in the image, and obtain the average value d pic of the distance between the lane lines in the images in the multiple images; 步骤3-3：获取两两相邻车道线的实际距离dactual；Step 3-3: Obtain the actual distance d actual between two adjacent lane lines; 步骤3-4：计算车道线在图像中距离的平均值和车道线实际距离的比值
Figure FDA0003025305150000011
Step 3-4: Calculate the ratio of the average distance of the lane line in the image to the actual distance of the lane line
Figure FDA0003025305150000011
 步骤3-5：计算视频每N帧的间隔时间
Figure FDA0003025305150000012
f为视频帧率；Step 3-5: Calculate the interval time of every N frames of the video
Figure FDA0003025305150000012
f is the video frame rate; 步骤3-6：计算车辆在相邻N帧图像上运动的距离spic，并求出车辆在间隔时间Δt内运动的实际距离
Figure FDA0003025305150000013
Step 3-6: Calculate the distance s pic that the vehicle moves on the adjacent N frame images, and obtain the actual distance the vehicle moves within the interval time Δt
Figure FDA0003025305150000013
 步骤3-7：计算车辆的运动速度
Figure FDA0003025305150000014
Step 3-7: Calculate the moving speed of the vehicle
Figure FDA0003025305150000014
 步骤4：对空中机器人采集的视频，采用步骤3，测量视频中的车辆速度，当车辆的速度超过设定值时，则将车辆标为红色以警示操作人员。Step 4: For the video collected by the aerial robot, use Step 3 to measure the speed of the vehicle in the video. When the speed of the vehicle exceeds the set value, the vehicle will be marked red to warn the operator. 2.根据权利要求1所述的一种基于空中机器人移动视觉的车速测量方法，其特征在于，所述数据增广的方法包括平移、旋转、翻转、裁切、放缩、调整色调、调整明度、调整饱和度、透视变换、图像融合和图像遮挡。2. a kind of vehicle speed measurement method based on aerial robot mobile vision according to claim 1, is characterized in that, the method of described data augmentation comprises translation, rotation, flip, crop, zoom, adjust color tone, adjust brightness , adjust saturation, perspective transformation, image fusion and image occlusion. 3.根据权利要求1所述的一种基于空中机器人移动视觉的车速测量方法，其特征在于，所述神经网络模型为YOLOv5。3. A kind of vehicle speed measurement method based on aerial robot mobile vision according to claim 1, is characterized in that, described neural network model is YOLOv5. 4.根据权利要求1所述的一种基于空中机器人移动视觉的车速测量方法，其特征在于，所述N＝10。4 . The vehicle speed measurement method based on the moving vision of an aerial robot according to claim 1 , wherein the N=10. 5 .
CN202110414461.3A 2021-04-16 2021-04-16 A vehicle speed measurement method based on aerial robot mobile vision Active CN113066106B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110414461.3A CN113066106B (en) 2021-04-16 2021-04-16 A vehicle speed measurement method based on aerial robot mobile vision

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110414461.3A CN113066106B (en) 2021-04-16 2021-04-16 A vehicle speed measurement method based on aerial robot mobile vision

Publications (2)

Publication Number Publication Date
CN113066106A true CN113066106A (en) 2021-07-02
CN113066106B CN113066106B (en) 2024-04-16

Family

ID=76567375

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110414461.3A Active CN113066106B (en) 2021-04-16 2021-04-16 A vehicle speed measurement method based on aerial robot mobile vision

Country Status (1)

Country Link
CN (1) CN113066106B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119049301A (en) * 2024-08-28 2024-11-29 山东交通学院 Vehicle instantaneous speed detection method and device based on unmanned aerial vehicle aerial video

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20000008234A (en) * 1998-07-10 2000-02-07 김영환 Road traffic situation information display device and method using a pilotless excessive speed gun
CN107067752A (en) * 2017-05-17 2017-08-18 北京联合大学 Automobile speedestimate system and method based on unmanned plane image
US20170330043A1 (en) * 2016-05-11 2017-11-16 Icatch Technology, Inc. Method and System for Synthesizing a Lane Image
CN110516633A (en) * 2019-08-30 2019-11-29 的卢技术有限公司 A kind of method for detecting lane lines and system based on deep learning
KR102147830B1 (en) * 2019-12-16 2020-08-26 (주)프리뉴 Intergrated control system and method of unmanned aerial vehicle using al based image processing
CN112149595A (en) * 2020-09-29 2020-12-29 爱动超越人工智能科技（北京）有限责任公司 Method for detecting lane line and vehicle violation by using unmanned aerial vehicle
CN112307921A (en) * 2020-10-22 2021-02-02 桂林电子科技大学 Vehicle-mounted end multi-target identification tracking prediction method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20000008234A (en) * 1998-07-10 2000-02-07 김영환 Road traffic situation information display device and method using a pilotless excessive speed gun
US20170330043A1 (en) * 2016-05-11 2017-11-16 Icatch Technology, Inc. Method and System for Synthesizing a Lane Image
CN107067752A (en) * 2017-05-17 2017-08-18 北京联合大学 Automobile speedestimate system and method based on unmanned plane image
CN110516633A (en) * 2019-08-30 2019-11-29 的卢技术有限公司 A kind of method for detecting lane lines and system based on deep learning
KR102147830B1 (en) * 2019-12-16 2020-08-26 (주)프리뉴 Intergrated control system and method of unmanned aerial vehicle using al based image processing
CN112149595A (en) * 2020-09-29 2020-12-29 爱动超越人工智能科技（北京）有限责任公司 Method for detecting lane line and vehicle violation by using unmanned aerial vehicle
CN112307921A (en) * 2020-10-22 2021-02-02 桂林电子科技大学 Vehicle-mounted end multi-target identification tracking prediction method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
赵晋;龙浩;张雅婷;: "基于无人机航拍影像处理的车速检测研究", 物联网技术, no. 01 *
高浩军;杜宇人;: "基于视频序列图像的车辆测速研究", 电子测量技术, no. 02, 22 April 2007 (2007-04-22) *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119049301A (en) * 2024-08-28 2024-11-29 山东交通学院 Vehicle instantaneous speed detection method and device based on unmanned aerial vehicle aerial video
CN119049301B (en) * 2024-08-28 2025-03-14 山东交通学院 A vehicle instantaneous speed detection method and device based on drone aerial video

Also Published As

Publication number Publication date
CN113066106B (en) 2024-04-16

Similar Documents

Publication Publication Date Title
CN111461088B (en) Rail transit obstacle avoidance system based on image processing and target recognition
CN104200657B (en) A kind of traffic flow parameter acquisition method based on video and sensor
CN104282020B (en) A kind of vehicle speed detection method based on target trajectory
CN101969548B (en) Active video acquisition method and device based on binocular camera
CN109829365B (en) Multi-scene adaptive driving deviation and turn warning method based on machine vision
CN114155511B (en) A method for collecting environmental information for autonomous vehicles on public roads
CN111967360B (en) Target vehicle posture detection method based on wheels
CN104881645B (en) The vehicle front mesh object detection method of feature based point mutual information and optical flow method
CN110379168B (en) A traffic vehicle information acquisition method based on Mask R-CNN
CN102393901B (en) Traffic flow information perception method based on hybrid characteristic and system thereof
CN110097762B (en) Road video image low visibility scale estimation method and system
CN104575003B (en) A kind of vehicle speed detection method based on traffic surveillance videos
CN107609486A (en) To anti-collision early warning method and system before a kind of vehicle
CN110222667A (en) A kind of open route traffic participant collecting method based on computer vision
CN108068817A (en) A kind of automatic lane change device and method of pilotless automobile
CN118015377B (en) A road vehicle trajectory extraction method based on multi-UAV images
CN105404857A (en) Infrared-based night intelligent vehicle front pedestrian detection method
CN108364466A (en) A kind of statistical method of traffic flow based on unmanned plane traffic video
CN105654073A (en) Automatic speed control method based on visual detection
CN105810023B (en) Airport undercarriage control automatic monitoring method
CN110826412A (en) Highway visibility detection system and method
CN103425764A (en) Vehicle matching method based on videos
CN102722894B (en) Intelligent video monitoring method based on automatic calibration of camera
CN117275216A (en) Multifunctional unmanned aerial vehicle expressway inspection system
CN117636268A (en) Unmanned aerial vehicle aerial natural driving data set construction method oriented to ice and snow environment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant