CN110307791A - Calculation method of vehicle length and speed based on 3D vehicle bounding box - Google Patents
Calculation method of vehicle length and speed based on 3D vehicle bounding box Download PDFInfo
- Publication number
- CN110307791A CN110307791A CN201910509507.2A CN201910509507A CN110307791A CN 110307791 A CN110307791 A CN 110307791A CN 201910509507 A CN201910509507 A CN 201910509507A CN 110307791 A CN110307791 A CN 110307791A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- bounding box
- detection area
- dimensional
- length
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004364 calculation method Methods 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 claims abstract description 24
- 239000011159 matrix material Substances 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 15
- 238000010276 construction Methods 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/04—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness specially adapted for measuring length or width of objects while moving
- G01B11/043—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness specially adapted for measuring length or width of objects while moving for measuring length
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P5/00—Measuring speed of fluids, e.g. of air stream; Measuring speed of bodies relative to fluids, e.g. of ship, of aircraft
- G01P5/18—Measuring speed of fluids, e.g. of air stream; Measuring speed of bodies relative to fluids, e.g. of ship, of aircraft by measuring the time taken to traverse a fixed distance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
本发明提供了一种基于三维车辆边界框的车辆长度及速度计算方法。该方法包括：(1)基于Mask R‑CNN网络生成车辆掩膜；(2)由场景中三个消失点向生成的车辆掩膜作切线进而构建三维车辆边界框；(3)建立车辆虚拟检测区，并根据三维车辆边界框的底面前边中点是否在检测区内来判断车辆是否在检测区内；(4)利用车道虚线段及场景消失点来确定路面参考点的像素坐标，再根据已知的车道虚线段长度、车道宽度来求解道路平面世界坐标与对应像素坐标之间的单应矩阵；(5)利用单应矩阵及三维车辆边界框计算车辆实际长度；(6)利用单应矩阵、三维车辆边界框及虚拟检测区计算车辆速度。本发明计算精度高且设备成本低，可有效应用于智慧交通系统当中。
The invention provides a method for calculating vehicle length and speed based on a three-dimensional vehicle bounding box. The method includes: (1) generating a vehicle mask based on the Mask R-CNN network; (2) making tangent lines from three vanishing points in the scene to the generated vehicle mask to construct a three-dimensional vehicle bounding box; (3) establishing a virtual vehicle detection area, and judge whether the vehicle is in the detection area according to whether the midpoint of the bottom front edge of the three-dimensional vehicle bounding box is in the detection area; The homography matrix between the world coordinates of the road plane and the corresponding pixel coordinates is solved by using the known length of the dotted line segment of the lane and the width of the lane; (5) Calculate the actual length of the vehicle by using the homography matrix and the three-dimensional vehicle bounding box; (6) Use the homography matrix , 3D vehicle bounding box and virtual detection area to calculate vehicle speed. The invention has high calculation precision and low equipment cost, and can be effectively applied to intelligent traffic systems.
Description
技术领域technical field
本发明涉及一种基于三维车辆边界框的车辆长度及速度计算方法，属于计算机视觉技术及智慧交通领域。The invention relates to a method for calculating vehicle length and speed based on a three-dimensional vehicle bounding box, and belongs to the fields of computer vision technology and intelligent transportation.
背景技术Background technique
车辆长度及速度是交通车辆信息中重要的参数。车辆速度通常通过埋入路面下的传感器或通过在道路上布置雷达来获得。然而采用这种方式设备造价较高，且当车辆之间距离很近时，会带来较大检测误差甚至导致检测失败。视频技术作为一种低成本的检测技术，目前被广泛研究，但其检测精度仍不理想。而车辆长度是作为车型分类的重要指标，其有效且低成本的计算方法目前仍较欠缺。如何有效、高精度且低成本地实现车辆长度及速度的计算是当前交通领域面临的难题。Vehicle length and speed are important parameters in traffic vehicle information. Vehicle speed is usually obtained by sensors buried under the road surface or by placing radar on the road. However, the cost of equipment in this way is high, and when the distance between the vehicles is very close, it will bring a large detection error or even lead to detection failure. As a low-cost detection technology, video technology has been widely studied, but its detection accuracy is still not ideal. Vehicle length is an important indicator for vehicle classification, and its effective and low-cost calculation methods are still lacking. How to realize the calculation of vehicle length and speed effectively, with high precision and at low cost is a difficult problem in the current traffic field.
发明内容Contents of the invention
为了解决上述存在的问题，本发明公开了一种基于三维车辆边界框的车辆长度及速度计算方法，计算精度高且设备成本低。In order to solve the above existing problems, the present invention discloses a vehicle length and speed calculation method based on a three-dimensional vehicle bounding box, which has high calculation accuracy and low equipment cost.
上述的目的通过以下技术方案实现：The above-mentioned purpose is achieved through the following technical solutions:
一种基于三维车辆边界框的车辆长度及速度计算方法，该方法包括：A method for calculating vehicle length and speed based on a three-dimensional vehicle bounding box, the method comprising:
(1)基于Mask R-CNN网络生成车辆掩膜；(1) Generate a vehicle mask based on the Mask R-CNN network;
(2)由场景中三个消失点向生成的车辆掩膜作切线进而构建三维车辆边界框；(2) Make a tangent from the three vanishing points in the scene to the generated vehicle mask to construct a 3D vehicle bounding box;
(3)建立车辆虚拟检测区，并根据三维车辆边界框的底面前边中点是否在检测区内来判断车辆是否在检测区内；(3) Establish a vehicle virtual detection area, and judge whether the vehicle is in the detection area according to whether the bottom front midpoint of the three-dimensional vehicle bounding box is in the detection area;
(4)对路面标定，获得道路平面世界坐标与对应像素坐标之间的单应矩阵；(4) To calibrate the road surface, obtain the homography matrix between the world coordinates of the road plane and the corresponding pixel coordinates;
(5)利用单应矩阵及三维车辆边界框计算车辆的实际长度；(5) Calculate the actual length of the vehicle using the homography matrix and the three-dimensional vehicle bounding box;
(6)利用单应矩阵、三维车辆边界框及虚拟检测区来计算车辆速度。(6) Using the homography matrix, the 3D vehicle bounding box and the virtual detection area to calculate the vehicle speed.
所述的基于三维车辆边界框的车辆长度及速度计算方法，(2)中所述的构建三维车辆边界框，具体是根据交通场景中的车道线、车辆纹理、路灯位置分别确定场景第一、第二及第三正交消失点，并基于这三个正交消失点向由Mask R-CNN生成的车辆掩膜作切线构建三维车辆边界框。The vehicle length and speed calculation method based on the three-dimensional vehicle bounding box, and the construction of the three-dimensional vehicle bounding box described in (2), specifically determine the scene first, vehicle texture, and street light positions according to the traffic scene in the traffic scene. The second and third orthogonal vanishing points, and based on these three orthogonal vanishing points, make a tangent to the vehicle mask generated by Mask R-CNN to construct a 3D vehicle bounding box.
所述的基于三维车辆边界框的车辆长度及速度计算方法，(3)中所述的建立车辆虚拟检测区是在镜头视野范围内建立车辆虚拟检测区，并根据三维车辆边界框的底面前边中点是否在检测区内来判断车辆是否在检测区内。According to the vehicle length and speed calculation method based on the three-dimensional vehicle bounding box, the establishment of the vehicle virtual detection area described in (3) is to establish the vehicle virtual detection area within the lens field of view, and according to the bottom front edge of the three-dimensional vehicle bounding box Whether the point is in the detection area to judge whether the vehicle is in the detection area.
所述的基于三维车辆边界框的车辆长度及速度计算方法，(4)中所述的路面标定进而获得道路平面世界坐标与对应像素坐标之间的单应矩阵H，The vehicle length and speed calculation method based on the three-dimensional vehicle bounding box, the road surface calibration described in (4) and then obtain the homography matrix H between the world coordinates of the road plane and the corresponding pixel coordinates,
具体首先确定车道虚线段端点的像素坐标，再根据车道虚线段端点及场景第二消失点建立直线，然后获取这些直线与车道两侧实线交点的像素坐标；将上述这些已知像素坐标的点当作路面标定参考点；利用已知的车道虚线段实际长度与车道宽度，即可以获得这些参考点的世界坐标；将参考点的像素坐标及世界坐标带入式(2)，通过利用最小二乘法或奇异值分解即可求得单应矩阵中八个独立参数，从而为基于三维车辆边界框的车辆长度及速度的计算提供基础。Specifically, first determine the pixel coordinates of the endpoints of the dotted line segment of the lane, and then establish a straight line according to the end point of the dotted line segment of the lane and the second vanishing point of the scene, and then obtain the pixel coordinates of the intersection points of these straight lines and the solid lines on both sides of the lane; As the reference point for pavement calibration; the world coordinates of these reference points can be obtained by using the known actual length of the dotted line segment of the lane and the width of the lane; the pixel coordinates and world coordinates of the reference point are brought into formula (2), and by using the least squares Eight independent parameters in the homography matrix can be obtained by multiplication or singular value decomposition, thus providing the basis for the calculation of vehicle length and speed based on the three-dimensional vehicle bounding box.
式中(xi,yi)和(Xi,Yi)分别代表参考点i的世界坐标与像素坐标,m为单应矩阵H中的独立参数。where ( xi ,y i ) and (X i ,Y i ) represent the world coordinates and pixel coordinates of the reference point i respectively, and m is an independent parameter in the homography matrix H.
所述的基于三维车辆边界框的车辆长度及速度计算方法，(5)中所述的车辆长度计算中，根据路面标定得到的单应矩阵计算三维车辆边界框底面前后两边的中点世界坐标，该两点世界坐标的差值即为车辆的实际长度。In the vehicle length and speed calculation method based on the three-dimensional vehicle bounding box, in the vehicle length calculation described in (5), the world coordinates of the midpoint of the three-dimensional vehicle bounding box bottom front and rear sides are calculated according to the homography matrix obtained by road surface calibration, The difference between the two world coordinates is the actual length of the vehicle.
所述的基于三维车辆边界框的车辆长度及速度计算方法，(6)中所述的车辆速度计算中，利用车辆在虚拟检测区中行驶的距离与时间来计算车辆速度，其中行驶距离根据车辆在进入及离开检测区时所对应的三维车辆边界框底面前边中点间的世界坐标差来计算，而行驶时间则根据视频帧率来确定，车速计算公式如下：In the vehicle length and speed calculation method based on the three-dimensional vehicle bounding box, in the vehicle speed calculation described in (6), the vehicle speed is calculated by using the distance and time the vehicle travels in the virtual detection area, wherein the travel distance is calculated according to the vehicle speed. When entering and leaving the detection area, the corresponding world coordinate difference between the bottom and front midpoints of the three-dimensional vehicle bounding box is calculated, and the driving time is determined according to the video frame rate. The formula for calculating the vehicle speed is as follows:
式中：V为车速,L及T分别是车辆在虚拟检测区内的行驶距离与时间,Nf是车辆在检测区内行驶过程中所对应的总帧数,Fr是视频的帧率,和分别是车辆进入与离开检测区所对应的三维边界框的底面前边中点世界坐标。In the formula: V is the speed of the vehicle, L and T are the driving distance and time of the vehicle in the virtual detection area respectively, N f is the total number of frames corresponding to the vehicle driving in the detection area, F r is the frame rate of the video, and They are the world coordinates of the midpoint of the bottom front of the three-dimensional bounding box corresponding to the vehicle entering and leaving the detection area.
与现有技术相比，本发明的有益效果在于：Compared with prior art, the beneficial effect of the present invention is:
(1)本发明提出的道路平面标定方法较利用三个正交消失点及镜头高度的计算方法精度更高。(1) The road plane calibration method proposed by the present invention has higher accuracy than the calculation method using three orthogonal vanishing points and lens height.
(2)本发明与传统基于埋入式传感器或雷达的测速方案相比，所需设备成本大幅降低。(2) Compared with the traditional speed measurement scheme based on embedded sensors or radars, the required equipment cost of the present invention is greatly reduced.
(3)本发明提出了基于单目相机的车辆长度计算方法，为交通车辆分类提供了新的方式。(3) The present invention proposes a vehicle length calculation method based on a monocular camera, which provides a new way for traffic vehicle classification.
附图说明Description of drawings
图1是三维车辆边界框的构建；Figure 1 is the construction of the 3D vehicle bounding box;
图2是路面标定参考点位置及验证目标；Figure 2 is the location of reference points for pavement calibration and verification targets;
图3是车长计算示意图；Figure 3 is a schematic diagram of vehicle length calculation;
图4是车速计算示意图。Figure 4 is a schematic diagram of vehicle speed calculation.
具体实施方式Detailed ways
下面结合具体实施方式对本发明作更进一步的说明。The present invention will be further described below in combination with specific embodiments.
以某桥面交通场景为例，对镜头中通过车辆的长度及速度进行检测。具体包含如下内容：Taking a traffic scene on a bridge deck as an example, the length and speed of passing vehicles in the camera are detected. Specifically include the following:
1.根据交通场景中的车道线、车辆纹理、路灯位置分别确定场景第一、第二及第三消失点，根据最小二乘法来确定场景中同一方向不同直线的交点即消失点。并基于这三个正交消失点向由Mask R-CNN生成的车辆掩膜作切线进而构建三维车辆边界框，如图1所示，首先根据切线间交点可直接确定三维边界框的1,2,3,4角点，然后利用这四个角点可以确定三维边界框另外四个角点5,6,7,8。八个角点全部确定后即可构建出三维车辆边界框。1. Determine the first, second, and third vanishing points of the scene according to the lane line, vehicle texture, and street light position in the traffic scene, and determine the intersection point of different straight lines in the same direction in the scene, that is, the vanishing point, according to the least square method. And based on these three orthogonal vanishing points, make tangents to the vehicle mask generated by Mask R-CNN to construct a 3D vehicle bounding box, as shown in Figure 1, firstly, the 1,2 of the 3D bounding box can be directly determined according to the intersection points between tangents ,3,4 corner points, and then use these four corner points to determine the other four corner points 5,6,7,8 of the 3D bounding box. After all the eight corner points are determined, the 3D vehicle bounding box can be constructed.
2.为了计算车速，在镜头视野范围内建立车辆虚拟检测区，并利用三维车辆边界框的底面前边中点是否在检测区内来判断车辆是否在检测区内。2. In order to calculate the vehicle speed, a vehicle virtual detection area is established within the field of view of the lens, and whether the midpoint of the bottom face of the 3D vehicle bounding box is within the detection area is used to determine whether the vehicle is in the detection area.
3.因为式(1)中单应矩阵有8个独立的未知参数，因此需要至少4个不在一条直线上的参考点来求解单应矩阵H。首先确定车道虚线段端点的像素坐标，再根据车道虚线段端点及第二消失点建立直线，然后获取这些直线与车道两侧实线交点的像素坐标。将这些已知像素坐标的点当作标定参考点，如图2所示。利用已知的车道虚线段实际长度与车道宽度，可以获得这些参考点的世界坐标。将参考点的像素坐标及世界坐标带入式(2)，通过最小二乘法或奇异值分解即可求得单应矩阵H。为了验证本发明所提标定方法的可靠性，15段已知长度的车道虚线段被当作验证目标来检验本发明的标定结果，验证目标如图2中所示。根据三个正交消失点及相机高度计算得出的验证目标长度及根据本专利方法计算得到的验证目标长度均列在表1中。从表中可以看出本专利提出方法的计算误差都小于5％，大幅超过传统方法。3. Because the homography matrix in formula (1) has 8 independent unknown parameters, at least 4 reference points that are not on a straight line are needed to solve the homography matrix H. First determine the pixel coordinates of the endpoints of the dashed lane segments, then establish straight lines based on the endpoints of the dashed lane segments and the second vanishing point, and then obtain the pixel coordinates of the intersections of these straight lines with the solid lines on both sides of the lane. These points with known pixel coordinates are taken as calibration reference points, as shown in Figure 2. The world coordinates of these reference points can be obtained by using the known actual length of the dashed lane segment and the width of the lane. Bring the pixel coordinates and world coordinates of the reference point into formula (2), and the homography matrix H can be obtained by the least square method or singular value decomposition. In order to verify the reliability of the calibration method proposed in the present invention, 15 lane dotted line segments with known lengths are used as verification targets to verify the calibration results of the present invention, and the verification targets are shown in FIG. 2 . The length of the verification target calculated according to the three orthogonal vanishing points and the height of the camera and the length of the verification target calculated according to the method of this patent are listed in Table 1. It can be seen from the table that the calculation errors of the methods proposed in this patent are all less than 5%, greatly exceeding the traditional methods.
4.将获得的单应矩阵求逆得H-1，然后根据三维车辆边界框底面前后两边的中点像素坐标的齐次形式乘以H-1得到相应的齐次世界坐标，再将齐次世界坐标用第三个分量进行归一化得到对应的二维世界坐标，车辆边界框底面前后两边的中点二维世界坐标的距离即为车辆的实际长度，车长计算示意图如图3所示。4. Invert the obtained homography matrix to obtain H -1 , and then multiply the homogeneous form of the pixel coordinates of the midpoint on both sides of the front and rear sides of the 3D vehicle bounding box by H -1 to obtain the corresponding homogeneous world coordinates, and then divide the homogeneous The world coordinates are normalized by the third component to obtain the corresponding two-dimensional world coordinates. The distance between the two-dimensional world coordinates of the midpoints of the front and rear sides of the bottom of the vehicle bounding box is the actual length of the vehicle. The schematic diagram of the calculation of the vehicle length is shown in Figure 3 .
5.利用车辆在虚拟检测区中行驶的距离与时间来计算车辆速度，如图4所示，其中行驶距离利用车辆在进入及离开检测区时所对应的三维车辆边界框底面前边中点间的世界坐标差来计算，而行驶时间则根据视频帧率来确定，车速按公式(3)计算如下：5. Use the distance and time the vehicle travels in the virtual detection area to calculate the vehicle speed, as shown in Figure 4, where the travel distance is the distance between the midpoints of the bottom front and the front edge of the three-dimensional vehicle bounding box corresponding to the vehicle entering and leaving the detection area The world coordinate difference is calculated, while the driving time is determined according to the video frame rate, and the vehicle speed is calculated according to formula (3) as follows:
综上，根据本发明提出的基于三维车辆边界框的车辆长度及速度计算方法可有效应用于智慧交通系统当中。In conclusion, the vehicle length and speed calculation method based on the three-dimensional vehicle bounding box proposed by the present invention can be effectively applied in intelligent transportation systems.
表1路面上验证目标长度计算结果Table 1 Calculation results of verification target length on the road
Claims (6)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910509507.2A CN110307791B (en) | 2019-06-13 | 2019-06-13 | Calculation method of vehicle length and speed based on 3D vehicle bounding box |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910509507.2A CN110307791B (en) | 2019-06-13 | 2019-06-13 | Calculation method of vehicle length and speed based on 3D vehicle bounding box |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110307791A true CN110307791A (en) | 2019-10-08 |
| CN110307791B CN110307791B (en) | 2020-12-29 |
Family
ID=68076628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910509507.2A Active CN110307791B (en) | 2019-06-13 | 2019-06-13 | Calculation method of vehicle length and speed based on 3D vehicle bounding box |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110307791B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909620A (en) * | 2019-10-30 | 2020-03-24 | 北京迈格威科技有限公司 | Vehicle detection method and device, electronic equipment and storage medium |
| CN112489106A (en) * | 2020-12-08 | 2021-03-12 | 深圳市哈工交通电子有限公司 | Video-based vehicle size measuring method and device, terminal and storage medium |
| CN112798811A (en) * | 2020-12-30 | 2021-05-14 | 杭州海康威视数字技术股份有限公司 | Speed measurement method, device and equipment |
| CN113011388A (en) * | 2021-04-23 | 2021-06-22 | 吉林大学 | Vehicle outer contour size detection method based on license plate and lane line |
| CN113689713A (en) * | 2020-05-19 | 2021-11-23 | 昆山研达电脑科技有限公司 | Vehicle speed monitoring method based on automobile data recorder |
| WO2021237750A1 (en) * | 2020-05-29 | 2021-12-02 | Siemens Ltd., China | Method and apparatus for vehicle length estimation |
| CN114419547A (en) * | 2021-12-30 | 2022-04-29 | 山东奥邦交通设施工程有限公司 | Vehicle detection method and system based on monocular vision and deep learning |
| CN114863025A (en) * | 2022-05-18 | 2022-08-05 | 禾多科技（北京）有限公司 | Three-dimensional lane line generation method and device, electronic device and computer readable medium |
| CN115019557A (en) * | 2022-06-09 | 2022-09-06 | 杭州电子科技大学 | TUIO protocol-based lane virtual boundary construction and boundary crossing detection method |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010214258A (en) * | 2009-03-13 | 2010-09-30 | Toyota Motor Corp | Masking jig |
| CN102410836A (en) * | 2011-07-26 | 2012-04-11 | 清华大学 | Space six-freedom degrees article locating system based on two-dimensional sensitive sensor |
| US20130128064A1 (en) * | 2011-04-08 | 2013-05-23 | Hailin Jin | Methods and Apparatus for Robust Video Stabilization |
| CN104200483A (en) * | 2014-06-16 | 2014-12-10 | 南京邮电大学 | Human body central line based target detection method under multi-camera environment |
| CN105573047A (en) * | 2014-10-10 | 2016-05-11 | 中芯国际集成电路制造(上海)有限公司 | System and method for detecting mask figure fidelity |
| CN105718870A (en) * | 2016-01-15 | 2016-06-29 | 武汉光庭科技有限公司 | Road marking line extracting method based on forward camera head in automatic driving |
| CN106408589A (en) * | 2016-07-14 | 2017-02-15 | 浙江零跑科技有限公司 | Vehicle-mounted overlooking camera based vehicle movement measurement method |
| CN107122792A (en) * | 2017-03-15 | 2017-09-01 | 山东大学 | Indoor arrangement method of estimation and system based on study prediction |
| US10055853B1 (en) * | 2017-08-07 | 2018-08-21 | Standard Cognition, Corp | Subject identification and tracking using image recognition |
| CN108550143A (en) * | 2018-04-03 | 2018-09-18 | 长安大学 | A kind of measurement method of the vehicle length, width and height size based on RGB-D cameras |
| CN108759667A (en) * | 2018-05-29 | 2018-11-06 | 福州大学 | Front truck distance measuring method based on monocular vision and image segmentation under vehicle-mounted camera |
| US20190103026A1 (en) * | 2017-09-29 | 2019-04-04 | Uber Technologies, Inc. | Image Processing for Vehicle Collision Avoidance System |
| WO2019097456A1 (en) * | 2017-11-17 | 2019-05-23 | C 3 Limited | Object measurement system |
-
2019
- 2019-06-13 CN CN201910509507.2A patent/CN110307791B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010214258A (en) * | 2009-03-13 | 2010-09-30 | Toyota Motor Corp | Masking jig |
| US20130128064A1 (en) * | 2011-04-08 | 2013-05-23 | Hailin Jin | Methods and Apparatus for Robust Video Stabilization |
| CN102410836A (en) * | 2011-07-26 | 2012-04-11 | 清华大学 | Space six-freedom degrees article locating system based on two-dimensional sensitive sensor |
| CN104200483A (en) * | 2014-06-16 | 2014-12-10 | 南京邮电大学 | Human body central line based target detection method under multi-camera environment |
| CN105573047A (en) * | 2014-10-10 | 2016-05-11 | 中芯国际集成电路制造(上海)有限公司 | System and method for detecting mask figure fidelity |
| CN105718870A (en) * | 2016-01-15 | 2016-06-29 | 武汉光庭科技有限公司 | Road marking line extracting method based on forward camera head in automatic driving |
| CN106408589A (en) * | 2016-07-14 | 2017-02-15 | 浙江零跑科技有限公司 | Vehicle-mounted overlooking camera based vehicle movement measurement method |
| CN107122792A (en) * | 2017-03-15 | 2017-09-01 | 山东大学 | Indoor arrangement method of estimation and system based on study prediction |
| US10055853B1 (en) * | 2017-08-07 | 2018-08-21 | Standard Cognition, Corp | Subject identification and tracking using image recognition |
| US20190103026A1 (en) * | 2017-09-29 | 2019-04-04 | Uber Technologies, Inc. | Image Processing for Vehicle Collision Avoidance System |
| WO2019097456A1 (en) * | 2017-11-17 | 2019-05-23 | C 3 Limited | Object measurement system |
| CN108550143A (en) * | 2018-04-03 | 2018-09-18 | 长安大学 | A kind of measurement method of the vehicle length, width and height size based on RGB-D cameras |
| CN108759667A (en) * | 2018-05-29 | 2018-11-06 | 福州大学 | Front truck distance measuring method based on monocular vision and image segmentation under vehicle-mounted camera |
Non-Patent Citations (3)
| Title |
|---|
| BENJAMIN COIFMAN等: "Speed estimation and length based vehicle classification from freeway single-loop detectors", 《TRANSPORTATION RESEARCH PART C》 * |
| 张利平等: "基于光流旳运动车辆检测和跟踪技术的研究", 《车辆与动力技术》 * |
| 赵俊梅等: "交通视频中运动车辆检测和跟踪技术的研究", 《车辆与动力技术》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909620A (en) * | 2019-10-30 | 2020-03-24 | 北京迈格威科技有限公司 | Vehicle detection method and device, electronic equipment and storage medium |
| CN113689713A (en) * | 2020-05-19 | 2021-11-23 | 昆山研达电脑科技有限公司 | Vehicle speed monitoring method based on automobile data recorder |
| CN115668297A (en) * | 2020-05-29 | 2023-01-31 | 西门子（中国）有限公司 | Method and apparatus for vehicle length estimation |
| WO2021237750A1 (en) * | 2020-05-29 | 2021-12-02 | Siemens Ltd., China | Method and apparatus for vehicle length estimation |
| CN112489106A (en) * | 2020-12-08 | 2021-03-12 | 深圳市哈工交通电子有限公司 | Video-based vehicle size measuring method and device, terminal and storage medium |
| CN112798811A (en) * | 2020-12-30 | 2021-05-14 | 杭州海康威视数字技术股份有限公司 | Speed measurement method, device and equipment |
| CN113011388A (en) * | 2021-04-23 | 2021-06-22 | 吉林大学 | Vehicle outer contour size detection method based on license plate and lane line |
| CN113011388B (en) * | 2021-04-23 | 2022-05-06 | 吉林大学 | A detection method of vehicle outer contour size based on license plate and lane line |
| CN114419547A (en) * | 2021-12-30 | 2022-04-29 | 山东奥邦交通设施工程有限公司 | Vehicle detection method and system based on monocular vision and deep learning |
| CN114863025A (en) * | 2022-05-18 | 2022-08-05 | 禾多科技（北京）有限公司 | Three-dimensional lane line generation method and device, electronic device and computer readable medium |
| CN114863025B (en) * | 2022-05-18 | 2023-03-10 | 禾多科技（北京）有限公司 | Three-dimensional lane line generation method, device, electronic device and computer readable medium |
| CN115019557A (en) * | 2022-06-09 | 2022-09-06 | 杭州电子科技大学 | TUIO protocol-based lane virtual boundary construction and boundary crossing detection method |
| CN115019557B (en) * | 2022-06-09 | 2024-05-14 | 杭州电子科技大学 | Lane virtual boundary construction and boundary crossing detection method based on TUIO protocol |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110307791B (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110307791B (en) | Calculation method of vehicle length and speed based on 3D vehicle bounding box | |
| CN103760569B (en) | A kind of drivable region detection method based on laser radar | |
| CN110031829B (en) | An accurate ranging method for targets based on monocular vision | |
| CN107705331A (en) | A kind of automobile video frequency speed-measuring method based on multiple views video camera | |
| CN110349221A (en) | A kind of three-dimensional laser radar merges scaling method with binocular visible light sensor | |
| CN103065323B (en) | Subsection space aligning method based on homography transformational matrix | |
| CN106228110A (en) | A kind of barrier based on vehicle-mounted binocular camera and drivable region detection method | |
| CN107284455B (en) | An ADAS system based on image processing | |
| CN107590438A (en) | A kind of intelligent auxiliary driving method and system | |
| CN107577996A (en) | A method and system for identifying vehicle driving path deviation | |
| CN107133985A (en) | A kind of vehicle-mounted vidicon automatic calibration method for the point that disappeared based on lane line | |
| CN103487034A (en) | Method for measuring distance and height by vehicle-mounted monocular camera based on vertical type target | |
| CN107705563B (en) | Continuous vehicle speed detection method based on lidar | |
| CN105203034A (en) | Height and area measurement method based on monocular camera three-dimensional distance measurement model | |
| CN106156723A (en) | A kind of crossing fine positioning method of view-based access control model | |
| CN106651953A (en) | Vehicle position and gesture estimation method based on traffic sign | |
| CN114758504B (en) | Online vehicle overspeed early warning method and system based on filtering correction | |
| CN102592454A (en) | Intersection vehicle movement parameter measuring method based on detection of vehicle side face and road intersection line | |
| WO2014129064A1 (en) | Obstacle detection device and obstacle detection method | |
| CN104729485A (en) | Visual positioning method based on vehicle-mounted panorama image and streetscape matching | |
| WO2020156923A2 (en) | Map and method for creating a map | |
| CN112489106A (en) | Video-based vehicle size measuring method and device, terminal and storage medium | |
| Hayakawa et al. | Ego-motion and surrounding vehicle state estimation using a monocular camera | |
| CN115265493B (en) | Lane-level positioning method and device based on non-calibrated camera | |
| CN111380503A (en) | Monocular camera ranging method adopting laser-assisted calibration |
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 |