CN106373430B - Intersection traffic early warning method based on computer vision - Google Patents
Intersection traffic early warning method based on computer vision Download PDFInfo
- Publication number
- CN106373430B CN106373430B CN201610735587.XA CN201610735587A CN106373430B CN 106373430 B CN106373430 B CN 106373430B CN 201610735587 A CN201610735587 A CN 201610735587A CN 106373430 B CN106373430 B CN 106373430B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- vehicles
- early warning
- moving target
- moving
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000010276 construction Methods 0.000 claims abstract description 33
- 239000000284 extract Substances 0.000 claims abstract description 15
- 230000002159 abnormal effect Effects 0.000 claims abstract description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 238000012544 monitoring process Methods 0.000 claims description 11
- 238000012706 support-vector machine Methods 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 2
- 238000010191 image analysis Methods 0.000 abstract 1
- 239000011159 matrix material Substances 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 4
- 206010039203 Road traffic accident Diseases 0.000 description 3
- 238000009434 installation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 101000827703 Homo sapiens Polyphosphoinositide phosphatase Proteins 0.000 description 1
- 102100023591 Polyphosphoinositide phosphatase Human genes 0.000 description 1
- 101100012902 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) FIG2 gene Proteins 0.000 description 1
- 101100233916 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) KAR5 gene Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000009430 construction management Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
本发明公开了一种基于计算机视觉的交叉路口通行预警方法，包括，S1实时采集交叉路口的视频图像，并抓拍车辆的特写图像；S2根据视频图像提取运动目标，生成运动目标行进信息及根据特写图像提取出完整的车辆牌照信息；S3对运动目标进行分类，得出分类结果；S4根据分类结果及行进信息，计算车辆的行进速度；S5存储通过交叉路口的车辆信息，根据车辆的行进速度及行进信息，预测所述运动目标下一时刻位置的质心坐标；S6根据行进信息、行进速度及该运动目标下一时刻位置的质心坐标，符合异常条件的情况，生成预警信号并显示预警信息。利用视频图像分析技术对施工路口通行异常情况进行分析并输出预警信号，稳定性强，准确率高。
The invention discloses a computer vision-based early warning method for intersection traffic, comprising: S1 collects video images of intersections in real time, and captures close-up images of vehicles; S2 extracts moving targets according to the video images, generates moving target travel information and The complete vehicle license plate information is extracted from the image; S3 classifies the moving objects and obtains the classification results; S4 calculates the speed of the vehicle according to the classification results and travel information; S5 stores the vehicle information passing through the intersection, and according to the speed and The traveling information predicts the coordinates of the center of mass of the moving object at the next moment; S6 generates an early warning signal and displays the early warning information if the abnormal conditions are met according to the traveling information, the traveling speed, and the coordinates of the center of mass of the moving object at the next moment. Using video image analysis technology to analyze the abnormal traffic conditions at construction intersections and output early warning signals, with strong stability and high accuracy.
Description
技术领域Technical Field
本发明涉及一种交通安全预警领域，具体涉及一种基于计算机视觉的交叉路口通行预警方法。The present invention relates to the field of traffic safety warning, and in particular to a computer vision-based intersection traffic warning method.
背景技术Background Art
在施工地段道路中，尤其是交叉路口，是道路网中最危险的区域，由于施工场地为临时开辟，信号灯和道路指示线稀少，行人不集中，施工车辆大多数为大型车辆，再加上外来车辆的驶入，成为了交通事故的多发点。行人、车辆行经交叉路口时，往往由于施工建筑物导致驾驶员与行人产生视觉盲区，外来车辆行驶至施工特定路段违反施工管理规定，以及车辆速度超过施工路段安全值，这些因素都有可能引发交通事故，在施工路段交通信号灯稀少，道路不规则的情况下，尤其需要注意交通安全隐患的产生。传统的交通预警系统并不完全适用于施工现场这样的地理条件，而且预警范围不够广，技术手段单一，为此我们提出一种基于计算机视觉的施工地段交叉路口通行预警系统及方法。The roads in the construction area, especially the intersection, are the most dangerous areas in the road network. Since the construction site is temporarily opened, there are few traffic lights and road signs, and there are no concentrated pedestrians. Most of the construction vehicles are large vehicles. In addition, the entry of foreign vehicles has become a high-incidence point for traffic accidents. When pedestrians and vehicles pass through the intersection, there is often a visual blind spot for drivers and pedestrians due to construction buildings. Foreign vehicles driving to specific construction sections violate construction management regulations, and vehicle speeds exceed the safety value of the construction section. These factors may cause traffic accidents. In the case of scarce traffic lights and irregular roads in the construction section, it is particularly necessary to pay attention to the generation of traffic safety hazards. The traditional traffic warning system is not fully applicable to geographical conditions such as construction sites, and the warning range is not wide enough and the technical means are single. For this reason, we propose a construction section intersection traffic warning system and method based on computer vision.
发明内容Summary of the invention
为了克服现有技术存在的缺点与不足，本发明提供一种基于计算机视觉的交叉口通行预警方法，具体应用在施工地段交叉路口通行的车辆与行人出现异常情况时进行预警，防止交通事故的发生。In order to overcome the shortcomings and deficiencies of the prior art, the present invention provides an intersection traffic warning method based on computer vision, which is specifically used to issue warnings when abnormal situations occur at vehicles and pedestrians passing through intersections in construction areas to prevent traffic accidents.
本发明采用如下技术方案：The present invention adopts the following technical solution:
一种基于计算机视觉的交叉路口通行预警方法，包括如下步骤：A computer vision-based intersection warning method comprises the following steps:
S1实时采集交叉路口的视频图像，并抓拍车辆的特写图像；S1 collects video images of intersections in real time and captures close-up images of vehicles;
S2根据视频图像提取运动目标，生成运动目标行进信息及根据特写图像提取出完整的车辆牌照信息；S2 extracts the moving target based on the video image, generates the moving target's travel information, and extracts the complete vehicle license plate information based on the close-up image;
S3对运动目标进行分类，得出分类结果；S3 classifies the moving target and obtains the classification result;
S4根据分类结果及行进信息，计算车辆的行进速度；S4 calculates the vehicle's travel speed according to the classification result and the travel information;
S5存储通过交叉路口的车辆信息，根据车辆的行进速度及行进信息，预测所述运动目标下一时刻位置的质心坐标；S5 stores the information of vehicles passing through the intersection, and predicts the centroid coordinates of the position of the moving target at the next moment according to the moving speed and moving information of the vehicles;
S6根据行进信息、行进速度及该运动目标下一时刻位置的质心坐标，对符合异常条件的情况，生成预警信号并显示预警信息。S6 generates a warning signal and displays warning information for situations that meet abnormal conditions based on the travel information, travel speed and the centroid coordinates of the moving target's position at the next moment.
所述运动目标包括车辆及行人，所述行进信息包括车辆或行人通过交叉路口的时间段、车辆或行人的质心坐标及行进方向。The moving targets include vehicles and pedestrians, and the travel information includes the time period when the vehicle or pedestrian passes through the intersection, the center of mass coordinates of the vehicle or pedestrian, and the travel direction.
所述根据视频图像提取运动目标，具体包括如下步骤：The method of extracting a moving target according to a video image specifically comprises the following steps:
S2.1对交叉路口场景的视频图像建立混合高斯模型，提取出前景目标，生成二值化的运动目标前景图；S2.1 establishes a mixed Gaussian model for the video image of the intersection scene, extracts the foreground target, and generates a binary moving target foreground image;
S2.2统计各个运动目标像素点个数和像素点的图像坐标，计算运动目标质心的图像坐标
其中M、N分别表示图像中某个运动目标的最大宽度和高度，i表示图像中像素点的横坐标，j表示图像中像素点的纵坐标，xij表示运动目标中像素点的横坐标值，yij表示运动目标像素点的纵坐标值；Where M and N represent the maximum width and height of a moving target in the image, i represents the horizontal coordinate of a pixel in the image, j represents the vertical coordinate of a pixel in the image, xij represents the horizontal coordinate value of a pixel in the moving target, and yij represents the vertical coordinate value of a pixel in the moving target;
S2.3根据相邻20帧图像中运动目标的质心的图像坐标计算运动目标的运动矢量，由运动矢量确定运动目标的行进方向，行进方向可依据交叉路口的不同进行统一划分。S2.3 calculates the motion vector of the moving target according to the image coordinates of the center of mass of the moving target in 20 adjacent frames of images, and determines the moving direction of the moving target by the motion vector. The moving direction can be uniformly divided according to different intersections.
S3对运动目标进行分类，得出分类结果，所述分类结果包括工程车、非工程车和行人，具体步骤包括：S3 classifies the moving targets to obtain classification results, which include engineering vehicles, non-engineering vehicles and pedestrians. The specific steps include:
S3.1提取运动目标的轮廓，并依据轮廓绘制包围轮廓的外接矩形，根据运动目标外接矩形的宽高比初步区分车辆目标与行人目标；S3.1 extract the contour of the moving target, and draw a circumscribed rectangle enclosing the contour based on the contour, and preliminarily distinguish between vehicle targets and pedestrian targets based on the aspect ratio of the circumscribed rectangle of the moving target;
如果运动目标外接矩形的宽高比小于设定的经验阈值，则认为此时的运动目标为行人；If the aspect ratio of the bounding rectangle of the moving target is less than the set empirical threshold, the moving target is considered to be a pedestrian.
如果运动目标外接矩形的宽高比大于设定的经验阈值，则认为此时的运动目标为工程车、非工程车或者是多个行人粘连形成的粘连行人目标；If the aspect ratio of the bounding rectangle of the moving target is greater than the set empirical threshold, the moving target is considered to be an engineering vehicle, a non-engineering vehicle, or a contiguous pedestrian target formed by the adhesion of multiple pedestrians;
S3.2根据运动目标的方向梯度直方图特征，利用支持向量机的算法建立分类器模型来区分工程车、非工程车与粘连行人目标。S3.2 Based on the directional gradient histogram features of the moving target, a classifier model is established using the support vector machine algorithm to distinguish engineering vehicles, non-engineering vehicles and attached pedestrian targets.
所述S3.2具体为：The S3.2 is specifically:
S3.2.1收集工程车、非工程车与粘连行人的图像，统一图像的尺寸大小，分别建立样本集，设工程车样本集为集合A、非工程车样本集为集合B、粘连行人样本集为集合C；S3.2.1 Collect images of engineering vehicles, non-engineering vehicles and attached pedestrians, unify the image sizes, and establish sample sets respectively. Let the engineering vehicle sample set be set A, the non-engineering vehicle sample set be set B, and the attached pedestrian sample set be set C;
S3.2.2将S3.2.1中的集合A与集合B作为正集、集合C作为负集，分别提取正集与负集的方向梯度直方图特征作为支持向量机算法的输入，生成分类器模型一，所述分类器模型一用于对车辆与行人进行二分类；S3.2.2: Set A and Set B in S3.2.1 are used as positive sets, and Set C is used as a negative set. The directional gradient histogram features of the positive set and the negative set are respectively extracted as inputs of the support vector machine algorithm to generate a classifier model 1, wherein the classifier model 1 is used to perform binary classification of vehicles and pedestrians.
S3.2.3将S3.2.1中的集合A作为正集、集合B作为负集，分别提取正集与负集的方向梯度直方图特征作为支持向量机算法的输入，生成分类器模型二，所述分类器模型二用于对工程车与非工程车进行二分类；S3.2.3: Set A in S3.2.1 is used as a positive set and set B is used as a negative set. The directional gradient histogram features of the positive set and the negative set are extracted as inputs of the support vector machine algorithm to generate a second classifier model, which is used to perform binary classification of engineering vehicles and non-engineering vehicles.
S3.2.4根据提取的各个运动目标的外接矩形提取各个运动目标的对应原始RGB图像，提取各个运动目标原始RGB图像的方向梯度直方图特征；S3.2.4 extract the corresponding original RGB image of each moving target according to the extracted bounding rectangle of each moving target, and extract the directional gradient histogram feature of the original RGB image of each moving target;
S3.2.5提取的各个运动目标原始RGB图像的方向梯度直方图特征依次作为分类器模型一的输入，如果输出结果为正，则认为此时的运动目标为车辆，如果输出结果为负，则认为此时的运动目标为粘连行人；The directional gradient histogram features of the original RGB images of each moving target extracted in S3.2.5 are sequentially used as the input of the classifier model 1. If the output result is positive, the moving target at this time is considered to be a vehicle. If the output result is negative, the moving target at this time is considered to be an adhered pedestrian.
S3.2.6将S3.2.5中分类结果为车辆的运动目标的方向梯度直方图特征作为分类器模型二的输入，如果输出结果为正，则认为此时的运动目标为工程车，如果输出结果为负，则认为此时的运动目标为非工程车。S3.2.6 uses the directional gradient histogram feature of the moving target classified as a vehicle in S3.2.5 as the input of classifier model 2. If the output result is positive, the moving target at this time is considered to be an engineering vehicle. If the output result is negative, the moving target at this time is considered to be a non-engineering vehicle.
所述S4中根据分类结果及行进信息，计算车辆的行进速度，具体为：In S4, the vehicle's travel speed is calculated according to the classification result and the travel information, specifically:
S4.1根据分类结果，选取车辆目标作为计算速度的对象，不计算行人的行进速度；S4.1 Based on the classification results, the vehicle target is selected as the object for calculating the speed, and the speed of pedestrians is not calculated;
S4.2：在视频图像中，按垂直于车辆目标行进方向设置两条虚拟检测线；再测定两条虚拟检测线对应实际道路上的距离ΔD；计算实时视频图像中车辆先后到达两条虚拟检测线的帧数F；S4.2: In the video image, two virtual detection lines are set perpendicular to the direction of travel of the vehicle target; then the distance ΔD between the two virtual detection lines on the actual road is measured; and the number of frames F in which the vehicle reaches the two virtual detection lines successively in the real-time video image is calculated;
S4.3：根据视频图像的采样频率f，两条虚拟检测线对应实际道路上的距离ΔD，车辆先后到达两条虚拟检测线的帧数F，计算出车辆的行进速度V：S4.3: According to the sampling frequency f of the video image, the distance ΔD between the two virtual detection lines on the actual road, and the number of frames F at which the vehicle reaches the two virtual detection lines one after another, the vehicle's travel speed V is calculated:
所述S5具体是采用卡尔曼滤波算法预测运动目标下一时刻位置的质心坐标。The S5 specifically uses the Kalman filter algorithm to predict the centroid coordinates of the moving target's position at the next moment.
所述异常条件包括如下情况：The abnormal conditions include the following:
车辆行进速度超出施工现场规定的速度值的情况生成预警信号Generates an early warning signal when the vehicle speed exceeds the speed value specified at the construction site
根据车辆的位置坐标以及所预测的车辆下一时刻位置的质心坐标，判断车辆是否存在位于施工现场的特殊区域或者已经驶入该特殊区域的情况，从而生成预警信号，所述特殊区域是指施工现场管理规定的非工程车车辆禁止驶入的路段；According to the position coordinates of the vehicle and the centroid coordinates of the predicted vehicle position at the next moment, it is determined whether the vehicle is located in a special area of the construction site or has entered the special area, thereby generating an early warning signal. The special area refers to a road section where non-engineering vehicles are prohibited from entering as stipulated by the construction site management regulations;
根据车辆与行人的行进信息，判断车辆与行人、车辆与车辆的通行是否存在构成视觉盲区的情况，从而生成预警信号。Based on the travel information of vehicles and pedestrians, it is determined whether the passage of vehicles and pedestrians, or vehicles and vehicles, constitutes a visual blind spot, thereby generating a warning signal.
所述预警信息包括显示超速车辆的监控图像并发出预警语音、驶入施工现场特殊区域的非工程车辆的监控图像并发出预警语音及显示车辆与行人、车辆与车辆之间有发生碰撞可能性时的场景图像并发出预警语音并显示提示标语。The warning information includes monitoring images showing speeding vehicles and issuing warning voices, monitoring images of non-engineering vehicles entering special areas of the construction site and issuing warning voices, and displaying scene images when there is a possibility of collision between vehicles and pedestrians or between vehicles, issuing warning voices and displaying prompt slogans.
全景摄像机采集交叉路口的视频图像，特写摄像机抓拍车辆的特写图像。The panoramic camera collects video images of the intersection, and the close-up camera captures close-up images of the vehicles.
本发明的有益效果：Beneficial effects of the present invention:
(1)能够有效区分工程车辆与非工程车辆，从而实行自动高效的监管，便于对不同的车辆采取不同的管制措施，为施工现场的安全管理提供便捷；(1) It can effectively distinguish between engineering vehicles and non-engineering vehicles, thereby implementing automatic and efficient supervision, facilitating the adoption of different control measures for different vehicles, and providing convenience for safety management of construction sites;
(2)能够对运动目标下一时刻的位置进行估计，从而对有闯入施工现场禁止区域可能性的车辆发出警报，从而避免危险情况的发生；(2) It can estimate the position of the moving target at the next moment, thereby issuing an alarm to vehicles that may enter the prohibited area of the construction site, thereby avoiding the occurrence of dangerous situations;
(3)能够检测出通行方向构成视觉盲区的车辆与行人，并显示安全提示信息，让驾驶员有充足时间采取减速或停车措施；(3) It can detect vehicles and pedestrians that form a visual blind spot in the direction of travel and display safety prompts, allowing the driver sufficient time to slow down or stop;
(4)对施工现场通行的车辆，不管是施工车辆还是外来车辆，实行信息化管理，建立数据库，存储其通过施工现场的时刻、车辆的种类，牌照等信息。(4) Implement information management for vehicles passing through the construction site, whether they are construction vehicles or external vehicles, and establish a database to store information such as the time they pass through the construction site, the type of vehicle, and the license plate.
(5)用于检测施工地段场景的高清摄像机架设在路面的上空，不需要将设备埋入地下，安装和维护费用较低。(5) The high-definition cameras used to detect construction site scenes are installed above the road surface. There is no need to bury the equipment underground, and the installation and maintenance costs are low.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1是本发明的结构示意图；Fig. 1 is a schematic structural diagram of the present invention;
图2是本发明实施例的结构现场安装图；FIG2 is a structural on-site installation diagram of an embodiment of the present invention;
图3是本发明实施例LED显示屏显示内容示意图；3 is a schematic diagram of the display content of the LED display screen according to an embodiment of the present invention;
图4是本发明的工作流程图。FIG. 4 is a flowchart of the present invention.
具体实施方式DETAILED DESCRIPTION
下面结合实施例及附图，对本发明作进一步地详细说明，但本发明的实施方式不限于此。The present invention will be further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.
实施例Example
如图1、图2及图3所示，一种实施本发明的基于计算机视觉的交叉路口通行预警系统，包括前端采集模块、视频信息处理模块、网络传输模块、监控模块及预警模块；As shown in FIG. 1 , FIG. 2 and FIG. 3 , a computer vision-based intersection traffic warning system implementing the present invention includes a front-end acquisition module, a video information processing module, a network transmission module, a monitoring module and a warning module;
所述前端采集模块包括全景摄像机2、特写摄像机1及视频编码设备，所述全景摄像机、特写摄像机分别与视频编码设备连接，在交叉路口特定区域设置虚拟线圈，特写摄像机的视场对准虚拟线圈7，全景摄像机的视场包括交叉路口各个方向视角；The front-end acquisition module includes a panoramic camera 2, a close-up camera 1 and a video encoding device. The panoramic camera and the close-up camera are respectively connected to the video encoding device. A virtual coil is set in a specific area of the intersection. The field of view of the close-up camera is aligned with the
所述视频信息处理模块，具体为工控机3，所述工控机通过网络传输模块分别与监控模块及预警模块连接，所述视频编码设备与工控机连接。The video information processing module is specifically an
所述监控模块由信息终端显示设备5及数据库服务器6构成；The monitoring module is composed of an information
所述预警模块包括LED显示屏8。The early warning module includes an
所述工控机的型号为IPC610。The model of the industrial computer is IPC610.
所述LED显示屏的型号为CSD-P6-SMD3535，双备份电源线，分辨率在720P以上。The model of the LED display is CSD-P6-SMD3535, with dual backup power lines and a resolution of 720P or above.
所述网络传输模块为无线网络传输设备4。The network transmission module is a wireless
所述全景摄像机与特写摄像机应该架设在十字或丁字路口转角路面上的空中。The panoramic camera and close-up camera should be set up in the air on the road surface at the corner of a cross or T-junction.
本实施例中虚拟线圈为矩形。In this embodiment, the virtual coil is rectangular.
本实施例中包括两个特写摄像机及一个全景摄像机，LED有两块显示屏幕，显示信息能够方便各个道路上的车辆与行人观察，如图3所示。This embodiment includes two close-up cameras and one panoramic camera, and the LED has two display screens, and the displayed information can facilitate the observation of vehicles and pedestrians on various roads, as shown in FIG3 .
全景摄像机采集交叉路口大范围的场景信息并输出至工控机进行视频图像的处理；在交叉路口处特定区域设置虚拟线圈，特写摄像机的视场对准交叉路口处特定区域设置虚拟线圈，抓拍车辆的高清图像并输出至工控机进行图像处理；工控机上安装的软件利用视频处理和模式识别技术完成车辆、行人的检测、跟踪、分类、位置预判、识别车辆牌照号、存储车辆信息、输出预警信号等；信号的输出端有LED显示屏、数据库服务器、信息终端显示设备。LED显示屏用于显示预警信息，数据库服务器用于存储通过施工地段交叉路口的车辆信息，信息终端显示设备用于实时显示交叉路口的监控画面以及与监控画面同步的实现车辆与行人的检测、跟踪、分类、位置预判、预警等功能的效果图。The panoramic camera collects scene information of a wide range of intersections and outputs it to the industrial computer for video image processing; a virtual coil is set in a specific area at the intersection, and the field of view of the close-up camera is aimed at the specific area at the intersection to set up a virtual coil, capture high-definition images of vehicles and output them to the industrial computer for image processing; the software installed on the industrial computer uses video processing and pattern recognition technology to complete vehicle and pedestrian detection, tracking, classification, position prediction, identification of vehicle license plate numbers, storage of vehicle information, output of warning signals, etc.; the output end of the signal has an LED display, a database server, and an information terminal display device. The LED display is used to display warning information, the database server is used to store vehicle information passing through the intersection of the construction site, and the information terminal display device is used to display the monitoring screen of the intersection in real time and the effect diagram of the detection, tracking, classification, position prediction, warning and other functions of vehicles and pedestrians synchronized with the monitoring screen.
如图4所示，一种基于计算机视觉的交叉路口通行预警方法，包括如下步骤：As shown in FIG4 , a method for early warning of intersection traffic based on computer vision includes the following steps:
S1全景摄像机实时采集交叉路口的车辆与行人通行的视频图像，特写摄像机抓拍虚拟线圈内每个车辆的特写图像；The S1 panoramic camera collects real-time video images of vehicles and pedestrians at the intersection, and the close-up camera captures close-up images of each vehicle within the virtual loop;
S2根据视频图像提取运动目标，生成运动目标行进信息及根据特写图像提取出完整的车辆牌照信息，所述运动目标包括车辆及行人，所述行进信息包括车辆或行人通过交叉路口的时间段、车辆或行人的质心坐标及行进方向。S2 extracts moving targets based on video images, generates moving target travel information and extracts complete vehicle license plate information based on close-up images. The moving targets include vehicles and pedestrians. The travel information includes the time period when the vehicle or pedestrian passes through the intersection, the center of mass coordinates of the vehicle or pedestrian and the travel direction.
所述车辆牌照信息是通过对特写图像进行车牌定位、字符分割、字符识别和颜色识别而得到的。The vehicle license plate information is obtained by performing license plate positioning, character segmentation, character recognition and color recognition on the close-up image.
S2.1对交叉路口场景的视频图像建立混合高斯模型，提取出前景目标，生成二值化的运动目标前景图；S2.1 establishes a mixed Gaussian model for the video image of the intersection scene, extracts the foreground target, and generates a binary moving target foreground image;
S2.2统计各个运动目标像素点个数和像素点的图像坐标，计算运动目标质心的图像坐标
其中M、N分别表示图像中某个运动目标的最大宽度和高度，i表示图像中像素点的横坐标，j表示图像中像素点的纵坐标，xij表示运动目标中像素点的横坐标值，yij表示运动目标像素点的纵坐标值；Where M and N represent the maximum width and height of a moving target in the image, i represents the horizontal coordinate of a pixel in the image, j represents the vertical coordinate of a pixel in the image, xij represents the horizontal coordinate value of a pixel in the moving target, and yij represents the vertical coordinate value of a pixel in the moving target;
S2.3根据相邻20帧图像中运动目标的质心的图像坐标计算运动目标的运动矢量，由运动矢量确定运动目标的行进方向，行进方向可依据交叉路口的不同进行统一划分。S2.3 calculates the motion vector of the moving target according to the image coordinates of the center of mass of the moving target in 20 adjacent frames of images, and determines the moving direction of the moving target by the motion vector. The moving direction can be uniformly divided according to different intersections.
S3对运动目标进行分类，得出分类结果；S3 classifies the moving target and obtains the classification result;
所述分类结果包括工程车、非工程车和行人，具体步骤包括：The classification results include engineering vehicles, non-engineering vehicles and pedestrians, and the specific steps include:
S3.1提取运动目标的轮廓，并依据轮廓绘制包围轮廓的外接矩形，根据运动目标外接矩形的宽高比初步区分车辆目标与行人目标；S3.1 extract the contour of the moving target, and draw a circumscribed rectangle enclosing the contour based on the contour, and preliminarily distinguish between vehicle targets and pedestrian targets based on the aspect ratio of the circumscribed rectangle of the moving target;
如果运动目标外接矩形的宽高比小于设定的经验阈值，则认为此时的运动目标为行人；If the aspect ratio of the bounding rectangle of the moving target is less than the set empirical threshold, the moving target is considered to be a pedestrian.
如果运动目标外接矩形的宽高比大于设定的经验阈值，则认为此时的运动目标为工程车、非工程车或者是多个行人粘连形成的粘连行人目标；If the aspect ratio of the bounding rectangle of the moving target is greater than the set empirical threshold, the moving target is considered to be an engineering vehicle, a non-engineering vehicle, or a contiguous pedestrian target formed by the adhesion of multiple pedestrians;
S3.2根据运动目标的方向梯度直方图特征，利用支持向量机的算法建立分类器模型来区分工程车、非工程车与粘连行人目标。S3.2 Based on the directional gradient histogram features of the moving target, a classifier model is established using the support vector machine algorithm to distinguish engineering vehicles, non-engineering vehicles and attached pedestrian targets.
S3.2.1收集工程车、非工程车与粘连行人的图像，统一图像的尺寸大小，分别建立样本集，设工程车样本集为集合A、非工程车样本集为集合B、粘连行人样本集为集合C；S3.2.1 Collect images of engineering vehicles, non-engineering vehicles and attached pedestrians, unify the image sizes, and establish sample sets respectively. Let the engineering vehicle sample set be set A, the non-engineering vehicle sample set be set B, and the attached pedestrian sample set be set C;
S3.2.2将S3.2.1中的集合A与集合B作为正集、集合C作为负集，分别提取正集与负集的方向梯度直方图特征作为支持向量机算法的输入，生成分类器模型一，所述分类器模型一用于对车辆与行人进行二分类；S3.2.2: Set A and Set B in S3.2.1 are used as positive sets, and Set C is used as a negative set. The directional gradient histogram features of the positive set and the negative set are respectively extracted as inputs of the support vector machine algorithm to generate a classifier model 1, wherein the classifier model 1 is used to perform binary classification of vehicles and pedestrians.
S3.2.3将S3.2.1中的集合A作为正集、集合B作为负集，分别提取正集与负集的方向梯度直方图特征作为支持向量机算法的输入，生成分类器模型二，所述分类器模型二用于对工程车与非工程车进行二分类；S3.2.3: Set A in S3.2.1 is used as a positive set and set B is used as a negative set. The directional gradient histogram features of the positive set and the negative set are extracted as inputs of the support vector machine algorithm to generate a second classifier model, which is used to perform binary classification of engineering vehicles and non-engineering vehicles.
S3.2.4根据提取的各个运动目标的外接矩形提取各个运动目标的对应原始RGB图像，提取各个运动目标原始RGB图像的方向梯度直方图特征；S3.2.4 extract the corresponding original RGB image of each moving target according to the extracted bounding rectangle of each moving target, and extract the directional gradient histogram feature of the original RGB image of each moving target;
S3.2.5提取的各个运动目标原始RGB图像的方向梯度直方图特征依次作为分类器模型一的输入，如果输出结果为正，则认为此时的运动目标为车辆，如果输出结果为负，则认为此时的运动目标为粘连行人；The directional gradient histogram features of the original RGB images of each moving target extracted in S3.2.5 are sequentially used as the input of the classifier model 1. If the output result is positive, the moving target at this time is considered to be a vehicle. If the output result is negative, the moving target at this time is considered to be an adhered pedestrian.
S3.2.6将S3.2.5中分类结果为车辆的运动目标的方向梯度直方图特征作为分类器模型二的输入，如果输出结果为正，则认为此时的运动目标为工程车，如果输出结果为负，则认为此时的运动目标为非工程车。S3.2.6 uses the directional gradient histogram feature of the moving target classified as a vehicle in S3.2.5 as the input of classifier model 2. If the output result is positive, the moving target at this time is considered to be an engineering vehicle. If the output result is negative, the moving target at this time is considered to be a non-engineering vehicle.
S4根据分类结果及行进信息，计算车辆的行进速度；S4 calculates the vehicle's travel speed according to the classification result and the travel information;
S4.1根据分类结果，选取车辆目标作为计算速度的对象，不计算行人的行进速度；S4.1 Based on the classification results, the vehicle target is selected as the object for calculating the speed, and the speed of pedestrians is not calculated;
S4.2：在视频图像中，按垂直于车辆目标行进方向设置两条虚拟检测线；再测定两条虚拟检测线对应实际道路上的距离ΔD；计算实时视频图像中车辆先后到达两条虚拟检测线的帧数F；S4.2: In the video image, two virtual detection lines are set perpendicular to the direction of travel of the vehicle target; then the distance ΔD between the two virtual detection lines on the actual road is measured; and the number of frames F in which the vehicle reaches the two virtual detection lines successively in the real-time video image is calculated;
S4.3：根据视频图像的采样频率f，两条虚拟检测线对应实际道路上的距离ΔD，车辆先后到达两条虚拟检测线的帧数F，计算出车辆的行进速度V：S4.3: According to the sampling frequency f of the video image, the distance ΔD between the two virtual detection lines on the actual road, and the number of frames F at which the vehicle reaches the two virtual detection lines one after another, the vehicle's travel speed V is calculated:
S5存储通过交叉路口的车辆信息，根据车辆的行进速度及行进信息，预测所述运动目标下一时刻位置的质心坐标；S5 stores the information of vehicles passing through the intersection, and predicts the centroid coordinates of the position of the moving target at the next moment according to the moving speed and moving information of the vehicles;
所述S5具体是采用卡尔曼滤波算法预测运动目标下一时刻的位置的质心坐标。卡尔曼滤波是一种递归的估计，分为预测阶段与更新阶段，在预测阶段，卡尔曼滤波算法使用当前时刻状态的估计，做出对下一时刻状态的估计；在更新阶段，卡尔曼滤波算法利用对下一时刻状态的观测值优化在预测阶段获得的预测值，以获得一个更精确的新估计值。S5 specifically uses the Kalman filter algorithm to predict the centroid coordinates of the position of the moving target at the next moment. Kalman filtering is a recursive estimation, which is divided into a prediction phase and an update phase. In the prediction phase, the Kalman filter algorithm uses the current state estimate to make an estimate of the state at the next moment; in the update phase, the Kalman filter algorithm uses the observed value of the state at the next moment to optimize the predicted value obtained in the prediction phase to obtain a more accurate new estimate.
具体为：Specifically:
S5.1获取视频图像中上一时刻运动目标的质心的图像坐标和质心在图像上的移动速度，建立运动目标位置的预测方程，即：S5.1 obtains the image coordinates of the center of mass of the moving target in the video image at the previous moment and the moving speed of the center of mass on the image, and establishes a prediction equation for the position of the moving target, namely:
X(t+1|t)＝AX(t|t)+w(t+1)X(t+1|t)＝AX(t|t)+w(t+1)
式中：X(t+1|t)为利用当前时刻预测出的下一时刻运动目标的状态向量；X(t|t)为当前时刻最优状态估计向量；A为状态传递矩阵；w(t+1)为过程噪声，假定期望为零的白噪声，它的协方差矩阵为Q(t+1)；Where: X(t+1|t) is the state vector of the moving target at the next moment predicted by the current moment; X(t|t) is the optimal state estimation vector at the current moment; A is the state transfer matrix; w(t+1) is the process noise, assuming that the expectation is zero white noise, its covariance matrix is Q(t+1);
所述运动目标质心在图像上的移动速度v具体计算公式如下：The specific calculation formula of the moving speed v of the moving target mass center on the image is as follows:
其中t为两个时刻的时间间隔，Δd为t时间运动目标质心移动的距离。Where t is the time interval between two moments, and Δd is the distance the center of mass of the moving target moves in time t.
S5.2：更新状态X(t+1|t)的协方差矩阵：S5.2: Update the covariance matrix of the state X(t+1|t):
P(t+1|t)＝AP(t|t)AT+Q(t+1)P(t+1|t)＝AP(t|t) AT +Q(t+1)
其中：P(t+1|t)表示X(t+1|t)对应的协方差，P(t|t)表示X(t|t)对应的协方差，AT表示A的转置矩阵，Where: P(t+1|t) represents the covariance corresponding to X(t+1|t), P(t|t) represents the covariance corresponding to X(t|t), AT represents the transposed matrix of A,
S5.3：根据下一时刻的运动目标状态的测量值，结合预测出的下一时刻运动目标的状态向量，计算下一时刻运动目标状态的最优化估算值X(t+1|t+1)：S5.3: Based on the measured value of the moving target state at the next moment and the predicted state vector of the moving target at the next moment, calculate the optimal estimated value X(t+1|t+1) of the moving target state at the next moment:
X(t+1|t+1)＝X(t+1|t)+Kg(t+1)(Z(t+1)-HX(t+1|t))X(t+1|t+1)=X(t+1|t)+Kg(t+1)(Z(t+1)-HX(t+1|t))
其中Z(t+1)为运动目标下一时刻状态的测量值，H为测量矩阵，其中Kg(t+1)为卡尔曼增益：Where Z(t+1) is the measured value of the state of the moving target at the next moment, H is the measurement matrix, and Kg(t+1) is the Kalman gain:
Kg(t+1)＝P(t+1|t)HT/(HP(t+1|t)HT+R(t+1))Kg(t+1)＝P(t+1|t)H T /(HP(t+1|t)H T +R(t+1))
其中R(t+1)为测量噪声协方差矩阵，HT为H的转置矩阵；Where R(t+1) is the measurement noise covariance matrix, and H T is the transposed matrix of H;
S5.4：更新下一时刻运动目标状态X(t+1|t+1)的协方差矩阵P(t+1|t+1)：S5.4: Update the covariance matrix P(t+1|t+1) of the moving target state X(t+1|t+1) at the next moment:
P(t+1|t+1)＝(I-Kg(t+1)H)P(t+1|t)P(t+1|t+1)=(I-Kg(t+1)H)P(t+1|t)
其中I为单位矩阵。Where I is the identity matrix.
所述异常情况主要包括车辆的速度超出施工现场规定的速度，非工程车驶入施工现场的特殊区域，交叉路口通行的车辆与车辆之间、车辆与行人之间的通行方向构成视觉盲区、车辆与车辆、车辆与行人有发生碰撞的可能性；The abnormal conditions mainly include the vehicle speed exceeding the speed specified at the construction site, non-engineering vehicles entering special areas of the construction site, the directions of vehicles and pedestrians passing through the intersection forming visual blind spots, and the possibility of collision between vehicles and vehicles, and vehicles and pedestrians;
所述施工现场的特殊区域是指施工现场管理规定的非工程车辆禁止驶入的路段；The special area of the construction site refers to the road section where non-engineering vehicles are prohibited from entering according to the construction site management regulations;
预警模块根据工控机输出的预警信号生成相应的警报信息，包括以下几种情况：(1)在信息终端显示设备上显示超速车辆的监控图像并发出预警语音；(2)在信息终端显示设备上显示驶入施工现场特殊区域的非工程车辆的监控图像并发出预警语音；(3)在信息终端设备上显示车辆与行人、车辆与车辆之间有发生碰撞可能性时的场景图像并发出预警语音并在LED屏幕上显示提示标语。The warning module generates corresponding alarm information according to the warning signal output by the industrial computer, including the following situations: (1) displaying the monitoring image of the speeding vehicle on the information terminal display device and issuing a warning voice; (2) displaying the monitoring image of the non-engineering vehicle entering the special area of the construction site on the information terminal display device and issuing a warning voice; (3) displaying the scene image when there is a possibility of collision between vehicles and pedestrians or between vehicles on the information terminal device, issuing a warning voice and displaying a prompt slogan on the LED screen.
图2为本发明实施例中预警模块的LED信息屏安装结构示意图，如图1的道路场景示意图所示，车辆与行人的通行方向由于施工路段的建筑区域导致视觉盲区的出现，此时预警模块输出预警信息至LED显示屏，提醒车辆和行人注意安全。Figure 2 is a schematic diagram of the installation structure of the LED information screen of the early warning module in an embodiment of the present invention. As shown in the road scene schematic diagram of Figure 1, the direction of travel of vehicles and pedestrians causes visual blind spots due to the construction area of the construction section. At this time, the early warning module outputs early warning information to the LED display screen to remind vehicles and pedestrians to pay attention to safety.
上述实施例为本发明较佳的实施方式，但本发明的实施方式并不受所述实施例的限制，其他的任何未背离本发明的精神实质与原理下所作的改变、修饰、替代、组合、简化，均应为等效的置换方式，都包含在本发明的保护范围之内。The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principles of the present invention shall be equivalent replacement modes and shall be included in the protection scope of the present invention.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610735587.XA CN106373430B (en) | 2016-08-26 | 2016-08-26 | Intersection traffic early warning method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610735587.XA CN106373430B (en) | 2016-08-26 | 2016-08-26 | Intersection traffic early warning method based on computer vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106373430A CN106373430A (en) | 2017-02-01 |
| CN106373430B true CN106373430B (en) | 2023-03-31 |
Family
ID=57904183
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610735587.XA Expired - Fee Related CN106373430B (en) | 2016-08-26 | 2016-08-26 | Intersection traffic early warning method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106373430B (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108200552B (en) * | 2017-12-14 | 2020-08-25 | 华为技术有限公司 | A V2X communication method and device |
| US11749111B2 (en) | 2018-03-19 | 2023-09-05 | Derq Inc. | Early warning and collision avoidance |
| CN108492567A (en) * | 2018-04-24 | 2018-09-04 | 汪宇明 | Monitor terminal, road traffic method for early warning and system |
| CN110610118A (en) * | 2018-06-15 | 2019-12-24 | 杭州海康威视数字技术股份有限公司 | Traffic parameter acquisition method and device |
| CN108877269B (en) * | 2018-08-20 | 2020-10-27 | 清华大学 | A method for vehicle status detection and V2X broadcasting at intersections |
| CN109147326A (en) * | 2018-09-06 | 2019-01-04 | 北京理工大学 | A kind of Campus transport safety warning system |
| CN109191852B (en) * | 2018-10-25 | 2021-07-06 | 西北工业大学 | Vehicle-Road-Cloud Collaborative Traffic Flow Situation Prediction Method |
| CN111260928B (en) * | 2018-11-30 | 2021-07-20 | 浙江宇视科技有限公司 | Vehicle disobedience pedestrian detection method and device |
| CN109830123B (en) * | 2019-03-22 | 2022-01-14 | 大陆投资(中国)有限公司 | Crossing collision early warning method and system |
| CN112216097A (en) * | 2019-07-09 | 2021-01-12 | 华为技术有限公司 | Method and device for detecting blind area of vehicle |
| CN110443161B (en) * | 2019-07-19 | 2023-08-29 | 宁波工程学院 | Monitoring method based on artificial intelligence in banking scene |
| EP4022591A4 (en) | 2019-08-29 | 2022-10-19 | Derq Inc. | Enhanced onboard equipment |
| CN111462501B (en) * | 2020-05-21 | 2021-08-17 | 山东师范大学 | A 5G network-based over-the-horizon traffic system and its implementation method |
| CN113793514A (en) * | 2021-08-30 | 2021-12-14 | 中冶南方城市建设工程技术有限公司 | Traffic safety warning system for entrances and exits of surrounding plots of construction roads |
| CN114936335A (en) * | 2022-03-28 | 2022-08-23 | 北京理工大学 | Real-time monitoring dynamic display method and device for large-scale movable motion matrix |
| CN115240471B (en) * | 2022-08-09 | 2024-03-01 | 东揽(南京)智能科技有限公司 | Intelligent factory collision avoidance early warning method and system based on image acquisition |
| CN115273479B (en) * | 2022-09-19 | 2023-03-31 | 深圳市博科思智能股份有限公司 | Operation and maintenance management method, device and equipment based on image processing and storage medium |
| CN115880888B (en) * | 2022-11-28 | 2025-01-24 | 复旦大学 | A method, device and medium for guiding intersection safety based on digital twin |
| CN116778665B (en) * | 2023-06-26 | 2025-02-11 | 中建八局第二建设有限公司 | An alarm warning device used at the gate of a construction site in an urban area |
| TWI852799B (en) * | 2023-10-16 | 2024-08-11 | 中國鋼鐵股份有限公司 | Early warning system for large vehicles at intersection |
| CN117789486B (en) * | 2024-02-28 | 2024-05-10 | 南京莱斯信息技术股份有限公司 | Monitoring system and method for right turn stop of intersection of large-sized vehicle |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5590217A (en) * | 1991-04-08 | 1996-12-31 | Matsushita Electric Industrial Co., Ltd. | Vehicle activity measuring apparatus |
| CN101179710A (en) * | 2007-11-30 | 2008-05-14 | 浙江工业大学 | Intelligent video monitoring device for railway crossing |
| CN101388145A (en) * | 2008-11-06 | 2009-03-18 | 北京汇大通业科技有限公司 | Auto alarming method and device for traffic safety |
| CN103345840A (en) * | 2013-05-28 | 2013-10-09 | 南京正保通信网络技术有限公司 | Video detection method of road crossing event at cross road |
| CN103971521A (en) * | 2014-05-19 | 2014-08-06 | 清华大学 | Method and device for detecting road traffic abnormal events in real time |
| CN104050684A (en) * | 2014-05-27 | 2014-09-17 | 华中科技大学 | Video moving object classification method and system based on on-line training |
-
2016
- 2016-08-26 CN CN201610735587.XA patent/CN106373430B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5590217A (en) * | 1991-04-08 | 1996-12-31 | Matsushita Electric Industrial Co., Ltd. | Vehicle activity measuring apparatus |
| CN101179710A (en) * | 2007-11-30 | 2008-05-14 | 浙江工业大学 | Intelligent video monitoring device for railway crossing |
| CN101388145A (en) * | 2008-11-06 | 2009-03-18 | 北京汇大通业科技有限公司 | Auto alarming method and device for traffic safety |
| CN103345840A (en) * | 2013-05-28 | 2013-10-09 | 南京正保通信网络技术有限公司 | Video detection method of road crossing event at cross road |
| CN103971521A (en) * | 2014-05-19 | 2014-08-06 | 清华大学 | Method and device for detecting road traffic abnormal events in real time |
| CN104050684A (en) * | 2014-05-27 | 2014-09-17 | 华中科技大学 | Video moving object classification method and system based on on-line training |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106373430A (en) | 2017-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106373430B (en) | Intersection traffic early warning method based on computer vision | |
| Zhang et al. | Video-based vehicle detection and classification system for real-time traffic data collection using uncalibrated video cameras | |
| CN104008645B (en) | A lane prediction and early warning method suitable for urban roads | |
| TWI409718B (en) | Method of locating license plate of moving vehicle | |
| CN103366571B (en) | The traffic incidents detection method at night of intelligence | |
| CN111800507A (en) | Traffic monitoring method and traffic monitoring system | |
| KR101104625B1 (en) | Apparatus and method for calculating traffic information of intelligent transportation system | |
| KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN101364347A (en) | A video-based detection method for vehicle control delays at intersections | |
| CN117292318B (en) | A computer vision-based transmission line safety monitoring and early warning method and system | |
| CN115762230A (en) | Parking lot intelligent guiding method and device based on remaining parking space amount prediction | |
| CN111951576A (en) | Traffic light control system based on vehicle identification and method thereof | |
| CN112686956A (en) | Method for detecting tilt fault of urban road signal lamp post | |
| CN119132101B (en) | Intelligent parking space state recognition system and method based on video detection | |
| CN113538968B (en) | Method and apparatus for outputting information | |
| CN115239093A (en) | Comprehensive risk assessment scheme based on airport environment target detection | |
| CN114581863A (en) | Method and system for identifying dangerous state of vehicle | |
| CN111383248A (en) | A method, device and electronic device for judging pedestrians running a red light | |
| KR100532058B1 (en) | Traffic information acquisition method and apparatus using camera calibration | |
| Ardestani et al. | Signal timing detection based on spatial–temporal map generated from CCTV surveillance video | |
| CN205983875U (en) | Early warning system passes in intersection based on computer vision | |
| Yu et al. | A traffic light detection method | |
| CN112215109B (en) | Vehicle detection method and system based on scene analysis | |
| CN115100903A (en) | Highway curve bidirectional early warning system based on YOLOV3 target detection algorithm | |
| KR20120124953A (en) | Auto Transportation Information Extraction System and Thereby Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20230331 |