CN111354035B - Automatic bridge dynamic load detection and inversion system and method - Google Patents

Automatic bridge dynamic load detection and inversion system and method Download PDF

Info

Publication number
CN111354035B
CN111354035B CN202010188342.6A CN202010188342A CN111354035B CN 111354035 B CN111354035 B CN 111354035B CN 202010188342 A CN202010188342 A CN 202010188342A CN 111354035 B CN111354035 B CN 111354035B
Authority
CN
China
Prior art keywords
bridge
vehicle
detected
vehicles
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.)
Active
Application number
CN202010188342.6A
Other languages
Chinese (zh)
Other versions
CN111354035A (en
Inventor
成高立
雷雪芹
张宏军
梁田
冉海峰
李炜
梁红涛
杨苏
刘高阳
高壮
王朵
徐娇
乔华玺
李明
贾保国
夏涛
张涛
韩晓凯
严元刚
侯群杰
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.)
Baoji Branch Of Shaanxi Baohan Expressway Construction Management Co ltd
XI'AN UNIVERSITY OF FINANCE AND ECONOMICS
SHAANXI HIGH-SPEED MECHANIZATION ENGINEERING CO LTD
Original Assignee
Baoji Branch Of Shaanxi Baohan Expressway Construction Management Co ltd
XI'AN UNIVERSITY OF FINANCE AND ECONOMICS
SHAANXI HIGH-SPEED MECHANIZATION ENGINEERING CO LTD
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 Baoji Branch Of Shaanxi Baohan Expressway Construction Management Co ltd, XI'AN UNIVERSITY OF FINANCE AND ECONOMICS, SHAANXI HIGH-SPEED MECHANIZATION ENGINEERING CO LTD filed Critical Baoji Branch Of Shaanxi Baohan Expressway Construction Management Co ltd
Priority to CN202010188342.6A priority Critical patent/CN111354035B/en
Publication of CN111354035A publication Critical patent/CN111354035A/en
Application granted granted Critical
Publication of CN111354035B publication Critical patent/CN111354035B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • 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
    • 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
    • G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
    • G01B11/0608—Height gauges
    • 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
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T15/00—3D [Three Dimensional] image rendering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30236—Traffic on road, railway or crossing
    • 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
    • Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation

Abstract

The invention discloses a bridge dynamic load automatic detection and inversion system and method, comprising a three-dimensional laser scanner (11), a computer, a plurality of cameras (12) and a timer used for being matched with the three-dimensional laser scanner (11); a plurality of road rods are uniformly distributed on the same-direction lanes on the bridge, each road rod is provided with a camera (12) for covering the field of view of the lane area between two adjacent road rods, the cameras synchronously pick up images, the synchronous precision is less than 10 microseconds, and a three-dimensional laser scanner (11) is arranged right above the same-direction double lanes at the entrance of the bridge and the measuring range covers two lanes; the bridge (21) is provided with a stress sensor at least at the maximum stress position of the bridge, and the stress sensor, the camera (12), the timer and the three-dimensional laser scanner (11) are all connected with a computer. The method realizes bridge dynamic load detection and accurate monitoring of strain of each part of the bridge.

Description

Automatic bridge dynamic load detection and inversion system and method
Technical Field
The invention belongs to the technical field of bridge dynamic load detection, and particularly relates to an automatic bridge dynamic load detection and inversion system and method.
Background
Bridges are an important component of highways. Because of the reasons of unreasonable bridge structure, natural disasters or overload, bridge collapse accidents are frequently caused, and great property loss is brought to the country and people. Therefore, in order to avoid the occurrence of similar accidents, many newly built bridges or in-use bridges are required to be provided with bridge monitoring systems. At present, bridge monitoring is carried out at regular detection of a bridge fixed point position, or monitoring video is adopted to monitor the traffic flow of the bridge deck, or a weighing sensor is buried to monitor the load of a single section, so that the tracking or stress monitoring of the dynamic load of the whole bridge deck is not really realized.
Meanwhile, the existing system vehicle monitoring and strain measurement are two isolated systems. Meanwhile, a full-bridge vehicle tracking system based on video does not exist, and most critical is that a plurality of videos cannot be synchronized, and the existing monitoring systems are independent and have no function of distinguishing empty vehicles from heavy vehicles.
Disclosure of Invention
The invention aims to solve the technical problems, and provides an automatic bridge dynamic load detection and inversion system and method, which realize bridge dynamic load detection and accurate monitoring of strain of each part of a bridge.
The invention is realized by the following technical scheme:
The bridge dynamic load automatic detection and inversion system comprises a three-dimensional laser scanner, a computer, a plurality of cameras and a timer used for being matched with the three-dimensional laser scanner; a plurality of road rods are uniformly distributed on the same-direction lanes on the bridge, each road rod is provided with a camera which covers the field of view of the lane area between two adjacent road rods, the cameras synchronously pick up images, the synchronous precision is less than 10 microseconds, and a three-dimensional laser scanner is arranged right above the same-direction double lanes at the entrance of the bridge and the measuring range covers two lanes; the bridge is provided with a stress sensor at least at the maximum stress position of the bridge, and the stress sensor, the camera, the timer and the three-dimensional laser scanner are all connected with a computer.
The three-dimensional laser scanner is used for obtaining point cloud data of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge and sending the obtained point cloud data to the computer;
the camera is used for shooting continuous running processes of all the detected vehicles moving below the camera, and after continuous running videos of the detected vehicles moving are shot, the shot continuous running videos of the detected vehicles moving are sent to the central processing unit;
The computer is used for receiving the point cloud data sent by the three-dimensional laser scanner, obtaining three-dimensional contour maps of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge through three-dimensional reconstruction based on the point cloud data, calculating the length, width and height dimension data of each detected vehicle according to the three-dimensional contour maps of the surfaces of each detected vehicle, and judging whether the detected vehicle is empty according to the carriage height in the dimension data of the detected vehicle, so as to obtain load judgment information of the detected vehicle; meanwhile, a certain truck is selected as a sample vehicle, and the length, width and height dimension data of the truck are recorded; comparing the size data of the detected vehicle with the size data of the sample vehicle to judge the vehicle type information of the detected vehicle; the vehicle type information and the load judgment information of the detected vehicle form vehicle load calculation parameter information;
meanwhile, the computer is used for receiving the continuous running video of the detected vehicle of the motion shot by the camera and processing the continuous running video into a frame of uninterrupted image; extracting gray images of a plurality of moving detected vehicles from the images by using a background difference method, then carrying out self-adaptive binarization processing on the gray images of the plurality of moving detected vehicles by using an improved Otsu algorithm, extracting Harris feature angular points from target results of the plurality of moving detected vehicles after the binarization processing, creating a plurality of tracking frames of the sizes of the self-adaptive moving detected vehicles according to the extracted Harris feature angular points to start tracking the moving detected vehicles in a view field, obtaining the quantity of the vehicles distributed in the view field according to the quantity of the tracking frames, and simultaneously, marking the moving detected vehicles by using continuous running videos of the moving detected vehicles shot by cameras arranged at the same-direction double-lane entrance of a bridge by using the obtained tracking frames of the sizes of the moving detected vehicles;
The centroid of each moving detected vehicle is obtained through the average value of the horizontal coordinates and the vertical coordinates of each moving detected vehicle in the Harris characteristic angular points, and meanwhile, the centroids of vehicles with the same labels in each frame of video images are connected through the nearest neighbor tracking filter algorithm and the improved Kalman filtering algorithm for tracking and judging the positions of the vehicles, so that the running track of the vehicles with the same labels is obtained, the tracking of the detected vehicles with multi-target movements is realized, and the distribution information of the vehicles moving on the whole bridge deck is obtained;
obtaining the load information of all vehicles on the bridge according to the obtained vehicle load calculation parameter information and the obtained distribution information of the whole bridge deck moving vehicle, and obtaining the stress strain distribution value of each point of the bridge by adopting a finite element model algorithm according to the obtained load information and the related parameter information of the bridge during bridge design;
and correcting the obtained stress-strain distribution value of each point of the corresponding part of the bridge according to the actual stress-strain distribution value of the bridge measured by the stress sensor arranged at the maximum stress position of the bridge and the stress-strain distribution value of each point of the corresponding part of the bridge obtained in the step through polynomial regression.
A bridge dynamic load automatic detection and inversion method comprises the following steps:
(1) On the same-direction double lanes of the bridge, utilizing a three-dimensional laser scanner which is erected right above the bridge entrance and covers two lanes of the measuring range to obtain point cloud data of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge, sending the obtained point cloud data to a computer, obtaining three-dimensional contour diagrams of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge through three-dimensional reconstruction based on the point cloud data, calculating length, width and height dimension data of all detected vehicles according to the three-dimensional contour diagrams of the surfaces of all detected vehicles, and judging whether the detected vehicles are empty or not according to the carriage height in the dimension data of the detected vehicles, thereby obtaining load judgment information of the detected vehicles; meanwhile, a certain truck is selected as a sample vehicle, and the length, width and height dimension data of the truck are recorded; comparing the size data of the detected vehicle with the size data of the sample vehicle to judge the vehicle type information of the detected vehicle; the vehicle type information and the load judgment information of the detected vehicle form vehicle load calculation parameter information;
(2) Shooting continuous running processes of all the detected vehicles moving under the bridge by using cameras which are erected on all uniformly distributed road rods and have a synchronous function with other cameras for covering the field of view of the area between two adjacent road rods on the same-direction double lanes of the bridge, after continuously running videos of the detected vehicles moving under the bridge are shot, sending the shot continuous running videos of the detected vehicles moving to a central processing unit, and processing the continuous running videos into continuous images of one frame by the central processing unit; extracting gray images of a plurality of moving detected vehicles from the images by using a background difference method, then carrying out self-adaptive binarization processing on the gray images of the plurality of moving detected vehicles by using an improved Otsu algorithm, extracting Harris feature angular points from target results of the plurality of moving detected vehicles after the binarization processing, creating a plurality of tracking frames of the sizes of the self-adaptive moving detected vehicles according to the extracted Harris feature angular points to start tracking the moving detected vehicles in a view field, obtaining the quantity of the vehicles distributed in the view field according to the quantity of the tracking frames, and simultaneously, marking the moving detected vehicles by using continuous running videos of the moving detected vehicles shot by cameras arranged at the same-direction double-lane entrance of a bridge by using the obtained tracking frames of the sizes of the moving detected vehicles;
(3) The central processing unit calculates the average value of the horizontal and vertical coordinates of the Harris characteristic angular points of each detected vehicle moving in the extracted Harris characteristic angular points to obtain the mass center of each detected vehicle moving, and meanwhile, the mass centers of vehicles with the same marks in each frame of video image are connected through the nearest neighbor tracking filter algorithm and the improved Kalman filtering algorithm for tracking and judging the positions of the vehicles, so that the running track of the vehicles with the same marks is obtained, the tracking of the detected vehicles with multi-target movements is realized, and the distribution information of the vehicles moving on the whole bridge deck is obtained;
(4) The central processing unit is used for obtaining the load information of all vehicles on the bridge according to the vehicle load calculation parameter information obtained in the step 1) and the distribution information of the whole bridge deck moving vehicle obtained in the step 3), and obtaining the stress strain distribution value of each point of the bridge by adopting a finite element model algorithm according to the obtained load information and the related parameter information of the bridge during bridge design;
(5) And (3) the central processing unit corrects the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4) by polynomial regression according to the actual stress-strain distribution value of the bridge measured by the stress sensor arranged at the maximum stress of the bridge and the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4).
The method for measuring the length of the passing detected vehicle by using the three-dimensional laser scanner comprises the following steps:
when the length of the passing detected vehicle is measured, if the head of the detected vehicle just enters a detection surface formed by the three-dimensional laser scanner, a timer used together with the three-dimensional laser scanner starts to count, when the tail of the detected vehicle leaves a detection plane formed by the three-dimensional laser scanner, the timer stops counting, the speed of the detected vehicle is v in the process of passing through the three-dimensional laser scanner, the passing time of the detected vehicle recorded by the timer is t, and the length L of the detected vehicle can be expressed as L=vt.
The method for measuring the passing detected vehicle width by using the three-dimensional laser scanner comprises the following steps:
when the width of the passing detected vehicle is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, in the polar coordinate system taking the three-dimensional laser scanner as the pole, the polar axis is along the vertical direction, when laser irradiates the leftmost position of the vehicle body of the detected vehicle, the distance measurement result is recorded as r1, the included angle between the laser line and the polar axis is a1, and then the distance from the left side of the detected vehicle to the polar axis direction is r 1 sin a 1 The method comprises the steps of carrying out a first treatment on the surface of the When the laser irradiates the rightmost position of the detected vehicle body, the distance measurement result is r2, the included angle between the laser line and the polar axis is a2, and the distance between the right side of the detected vehicle and the polar axis direction is r 2 sin a 2 The width w of the detected vehicle can be expressed as w=r 2 sin a 2 -r 1 sin a 1 。
The method for measuring the height of the passing detected vehicle by using the three-dimensional laser scanner comprises the following steps:
when the height of the detected vehicle body is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, in the polar coordinate system taking the three-dimensional laser scanner as the pole, the polar axis is along the horizontal direction, when laser irradiates the highest position of the detected vehicle, the distance measurement result is recorded as r, at the moment, the included angle between the laser line and the polar axis is a, and then the distance from the highest position of the detected vehicle to the polar axis is r sin a; the three-dimensional laser scanner is spaced from the ground by a distance H, and the detected vehicle height H can be expressed as h=h—r sin a.
The step 1) is specifically as follows:
the three-dimensional laser scanner scans the top of the detected vehicle body and one side surface of the vehicle by utilizing the scanning light curtain constructed by the three-dimensional laser scanner, so that a plurality of continuous scanning cross sections of the detected vehicle are obtained, a three-dimensional contour map of the detected vehicle is drawn according to the plurality of continuous scanning cross sections, and different types of detected vehicles are distinguished from the three-dimensional contour map of the detected vehicle;
for flatbed trucks and tipping trucks, whether the truck is empty is determined by confirming whether the three-dimensional laser scanner is the shortest distance to the left or right edge of the truck.
The principle of the improved Otsu algorithm is as follows:
1) According to the formulaObtaining the distribution probability F of the image gray level 1 ，F 2 ，F 3 …F i . Wherein N is the total number of pixels, N i Is the number of pixels with a gray value i.
2) The average gray value μ of the image is calculated.
3) Let the gray level probability of the background image be P, and the average gray level value of the background image be mu 1 . The image variance is Var and the variance at the maximum value is maxVar. Let p=0, μ 1 ＝0，maxVar＝0。
4) i is iterated from 1 to 256, and the following values are calculated in turn:
P(i)＝P(i-1)+F i
μ 1 (i)＝(P(i-1)×μ 1 (i-1)+i×F i )/P(i)
the inter-class variance can be expressed as:
Var＝P·(μ 1 -μ) 2 +P 2 ·(μ 2 -μ) 2 (3.16)
wherein P is 2 Is the gray level probability of the foreground image, μ 2 Is the gray average of the foreground image. And because of
μ＝P·μ 1 +P 2 ·μ 2 (3.17)
Bringing formula 3.17 into formula 3.16 yields:
Var(i)＝P(i)×(μ-μ 1 (i)) 2 /(1-P(i))
5) Judging whether Var (i) is larger than maxVar, if so, making maxVar=Var (i), and T=i; if not, let i=i+1, return to step 4.
6) And finishing the iterative search, wherein T is the requirement.
The camera has a day-night conversion monitoring function, the effective pixels of the camera reach 300 ten thousand, the horizontal field angle is 80 degrees, the maximum image size reaches 1920 multiplied by 1080, the resolution ratio of video is 1296 (horizontal) multiplied by 732 (vertical), and the video frame rate is 50Hz:25fps, 7W power supply and 12VDC power supply. The size of the material is 194.04 multiplied by 93.85 multiplied by 89.52mm, and the weight of the material is 1000g; the synchronization time precision of the camera is not less than 10 microseconds;
The cameras are composed of a plurality of high-resolution cameras, each camera covers 50 meters of the bridge deck, and the multiple cameras need to be precisely synchronized in time, so that dynamic changes of loads of each bridge span of the bridge deck are obtained.
The specific algorithm for obtaining the load information of all vehicles to the bridge through the vehicle load calculation parameter information obtained in the step 1) and the distribution information of the whole bridge deck moving vehicle obtained in the step 3) is as follows:
the load F applied to the road surface by the whole vehicle can be divided into a vehicle static load F j And dynamic load F d I.e.
F＝F j +F d
Wherein the vehicle is under a static load F j I.e. gravity load.
In the process of solving the dynamic load of the vehicle, according to the D' Alemmbert principle, the motion equation of the vibration system of the vehicle is firstly obtained:
wherein: quality matrix
Damping matrix
Stiffness matrix
Displacement array y= [ y ] t ,y s ] T ；
Excitation arrayIn the above matrix, m s Is a sprung mass; m is m t Is a non-sprung mass; c s Damping for the suspension; c t Damping the tire. k (k) 1 The rigidity coefficient of the rear tire of the automobile; k (k) 2 The rigidity coefficient of the rear suspension of the automobile; k (k) t Is the tire stiffness; y is s ,y t Vertical absolute displacement for sprung and unsprung masses; y is r Exciting vertical displacement of the road surface unevenness; Δh is the amount of vertical deformation of the road surface.
The random dynamic load of the vehicle to the road surface is as follows:
After the load of a single vehicle to the road surface is obtained, the total load of all vehicles running on the bridge to the bridge surface can be calculated by calculating the number of the vehicles on the bridge surface;
the dynamic load on the bridge deck can be obtained through the step, and the stress strain distribution value of the whole bridge is calculated by adopting a finite element method according to the bridge structure.
Compared with the prior art, the invention has the following beneficial technical effects:
according to the bridge dynamic load automatic detection and inversion system and method provided by the invention, the vehicle monitoring and the strain measurement are unified, the bridge dynamic load automatic detection is realized, and a plurality of videos are synchronized and the full-bridge vehicle is tracked.
The bridge dynamic load automatic detection and inversion system and method are characterized in that a three-dimensional laser scanner is utilized to obtain a point cloud image of each detected vehicle entering the movement of a bridge, the three-dimensional reconstruction is carried out on each section, the model of the detected vehicle and the load judgment information of the vehicle are judged, a group of high-resolution camera combinations which are accurately synchronous and are distributed at equal intervals and cover the whole bridge deck are utilized, the vehicle dynamic distribution on each span structure of the bridge deck is obtained through a vehicle segmentation and track tracking algorithm, the model information of each vehicle and the load judgment information of the vehicle are combined, the static load of the bridge deck is obtained through calculation, and the dynamic stress distribution of each point of the whole bridge deck is obtained through a bridge finite element solving method; the stress sensor is installed at the key position of the bridge and is used as a correction reference, the calculation result of the finite element is inverted, and the strain of each position of the bridge is accurately monitored finally.
Drawings
FIG. 1 is a schematic diagram of a system for measuring vehicle size data and tracking vehicles provided by a bridge dynamic load automatic detection and inversion system according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a bridge load from a moving vehicle provided by an automatic bridge dynamic load detection and inversion system according to an embodiment of the invention;
FIG. 3 is a schematic illustration of measuring a width of a vehicle body according to one embodiment of the invention;
FIG. 4 is a schematic view of measuring a height of a vehicle body according to one embodiment of the invention;
FIG. 5 is a schematic diagram of determining whether a flatbed truck is empty in accordance with one embodiment of the present invention;
FIG. 6 is a schematic illustration of determining whether a dump truck is empty according to one embodiment of the invention;
FIG. 7 is a schematic illustration of inversion according to the present invention;
fig. 8 is a flow chart of the present invention.
10: the bridge dynamic load automatic detection and inversion system;
11: a three-dimensional laser scanner; 12: a camera;
20: bridge load;
21: a bridge; 22: a vehicle;
30: measuring the width of the vehicle body;
r1: the distance from the three-dimensional laser scanner to the leftmost side of the vehicle; a1: the included angle between the laser line and the polar axis; r2: the distance from the three-dimensional laser scanner to the rightmost side of the vehicle; a2: the included angle between the laser line and the polar axis;
40: measuring the height of the vehicle body;
r: the distance from the three-dimensional laser scanner to the highest position of the vehicle; a: the included angle between the laser line and the polar axis; h: the three-dimensional laser scanner reaches the ground level;
50: judging whether the flat wagon is empty;
d: the distance from the three-dimensional laser scanner to the leftmost side of the flat plate; d: three-dimensional laser scanner to cargo distance;
60: judging whether the skip wagon is empty;
d: the distance from the three-dimensional laser scanner to the leftmost side of the tipping bucket; d: three-dimensional laser scanner to cargo distance.
Detailed Description
The invention will now be described in further detail with reference to specific examples, which are intended to illustrate, but not to limit, the invention.
Referring to fig. 1 to 8, an automatic bridge dynamic load detection and inversion system includes a three-dimensional laser scanner 11, a computer, a plurality of cameras 12, and a timer for use with the three-dimensional laser scanner 11; a plurality of road rods are uniformly distributed on the same-direction lanes on the bridge, each road rod is provided with a camera 12 for covering the field of view of the lane area between two adjacent road rods, the cameras synchronously pick up images, the synchronous precision is less than 10 microseconds, and a three-dimensional laser scanner 11 is arranged right above the same-direction double lanes at the entrance of the bridge and the measuring range covers two lanes; the bridge 21 is provided with a stress sensor at least at the maximum stress position of the bridge, and the stress sensor, the camera 12, the timer and the three-dimensional laser scanner 11 are all connected with a computer.
Specifically, the three-dimensional laser scanner 11 is configured to obtain point cloud data of respective surfaces of all detected vehicles entering the same-direction double lanes of the bridge, and send the obtained point cloud data to the computer;
the camera 12 is configured to capture continuous running processes of all the vehicles under the camera, and send the captured continuous running videos of the moving vehicles to the central processing unit after capturing the continuous running videos of the moving vehicles;
the computer is used for receiving the point cloud data sent by the three-dimensional laser scanner, obtaining three-dimensional contour maps of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge through three-dimensional reconstruction based on the point cloud data, calculating the length, width and height dimension data of each detected vehicle according to the three-dimensional contour maps of the surfaces of each detected vehicle, and judging whether the detected vehicle is empty according to the carriage height in the dimension data of the detected vehicle, so as to obtain load judgment information of the detected vehicle; meanwhile, a certain truck is selected as a sample vehicle, and the length, width and height dimension data of the truck are recorded; comparing the size data of the detected vehicle with the size data of the sample vehicle to judge the vehicle type information of the detected vehicle; the vehicle type information and the load judgment information of the detected vehicle form vehicle load calculation parameter information;
Meanwhile, the computer is used for receiving the continuous running video of the detected vehicle of the motion shot by the camera and processing the continuous running video into a frame of uninterrupted image; extracting gray images of a plurality of moving detected vehicles from the images by using a background difference method, then carrying out self-adaptive binarization processing on the gray images of the plurality of moving detected vehicles by using an improved Otsu algorithm, extracting Harris feature angular points from target results of the plurality of moving detected vehicles after the binarization processing, creating a plurality of tracking frames of the sizes of the self-adaptive moving detected vehicles according to the extracted Harris feature angular points to start tracking the moving detected vehicles in a view field, obtaining the quantity of the vehicles distributed in the view field according to the quantity of the tracking frames, and simultaneously, marking the moving detected vehicles by using continuous running videos of the moving detected vehicles shot by cameras arranged at the same-direction double-lane entrance of a bridge by using the obtained tracking frames of the sizes of the moving detected vehicles;
the centroid of each moving detected vehicle is obtained through the average value of the horizontal coordinates and the vertical coordinates of each moving detected vehicle in the Harris characteristic angular points, and meanwhile, the centroids of vehicles with the same labels in each frame of video images are connected through the nearest neighbor tracking filter algorithm and the improved Kalman filtering algorithm for tracking and judging the positions of the vehicles, so that the running track of the vehicles with the same labels is obtained, the tracking of the detected vehicles with multi-target movements is realized, and the distribution information of the vehicles moving on the whole bridge deck is obtained;
Obtaining the load information of all vehicles on the bridge according to the obtained vehicle load calculation parameter information and the obtained distribution information of the whole bridge deck moving vehicle, and obtaining the stress strain distribution value of each point of the bridge by adopting a finite element model algorithm according to the obtained load information and the related parameter information of the bridge during bridge design;
and correcting the obtained stress-strain distribution value of each point of the corresponding part of the bridge according to the actual stress-strain distribution value of the bridge measured by the stress sensor arranged at the maximum stress position of the bridge and the stress-strain distribution value of each point of the corresponding part of the bridge obtained in the step through polynomial regression.
Referring to fig. 1 to 8, a bridge dynamic load automatic detection and inversion method comprises the following steps:
1) On the same-direction double lanes of the bridge, utilizing a three-dimensional laser scanner which is erected right above the bridge entrance and covers two lanes of the measuring range to obtain point cloud data of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge, sending the obtained point cloud data to a computer, obtaining three-dimensional contour diagrams of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge through three-dimensional reconstruction based on the point cloud data, calculating length, width and height dimension data of all detected vehicles according to the three-dimensional contour diagrams of the surfaces of all detected vehicles, and judging whether the detected vehicles are empty or not according to the carriage height in the dimension data of the detected vehicles, thereby obtaining load judgment information of the detected vehicles; meanwhile, a certain truck is selected as a sample vehicle, and the length, width and height dimension data of the truck are recorded; comparing the size data of the detected vehicle with the size data of the sample vehicle to judge the vehicle type information of the detected vehicle; the vehicle type information and the load judgment information of the detected vehicle constitute vehicle load calculation parameter information.
The step 1) is specifically as follows:
the three-dimensional laser scanner scans the top of the detected vehicle body and one side surface of the vehicle by utilizing the scanning light curtain constructed by the three-dimensional laser scanner, so that a plurality of continuous scanning cross sections of the detected vehicle are obtained, a three-dimensional contour map of the detected vehicle is drawn according to the plurality of continuous scanning cross sections, and different types of detected vehicles are distinguished from the three-dimensional contour map of the detected vehicle;
for flatbed trucks and tipping trucks, whether the truck is empty is determined by confirming whether the three-dimensional laser scanner is the shortest distance to the left or right edge of the truck.
2) Shooting continuous running processes of all the detected vehicles moving under the bridge by using cameras which are erected on all uniformly distributed road rods and have a synchronous function with other cameras for covering the field of view of the area between two adjacent road rods on the same-direction double lanes of the bridge, after continuously running videos of the detected vehicles moving under the bridge are shot, sending the shot continuous running videos of the detected vehicles moving to a central processing unit, and processing the continuous running videos into continuous images of one frame by the central processing unit; extracting gray images of a plurality of moving detected vehicles from the images by using a background difference method, then carrying out self-adaptive binarization processing on the gray images of the plurality of moving detected vehicles by using an improved Otsu algorithm, extracting Harris feature angular points from target results of the plurality of moving detected vehicles after the binarization processing, creating a plurality of tracking frames of the sizes of the self-adaptive moving detected vehicles according to the extracted Harris feature angular points to start tracking the moving detected vehicles in a view field, obtaining the quantity of the vehicles distributed in the view field according to the quantity of the tracking frames, and simultaneously, marking the moving detected vehicles by using continuous running videos of the moving detected vehicles shot by cameras arranged at the same-direction double-lane entrance of a bridge by using the obtained tracking frames of the sizes of the moving detected vehicles;
3) The central processing unit calculates the average value of the horizontal and vertical coordinates of the Harris characteristic angular points of each detected vehicle moving in the extracted Harris characteristic angular points to obtain the mass center of each detected vehicle moving, and meanwhile, the mass centers of vehicles with the same marks in each frame of video image are connected through the nearest neighbor tracking filter algorithm and the improved Kalman filtering algorithm for tracking and judging the positions of the vehicles, so that the running track of the vehicles with the same marks is obtained, the tracking of the detected vehicles with multi-target movements is realized, and the distribution information of the vehicles moving on the whole bridge deck is obtained;
4) The central processing unit is used for obtaining the load information of all vehicles on the bridge according to the vehicle load calculation parameter information obtained in the step 1) and the distribution information of the whole bridge deck moving vehicle obtained in the step 3), and obtaining the stress strain distribution value of each point of the bridge by adopting a finite element model algorithm according to the obtained load information and the related parameter information of the bridge during bridge design;
5) And (3) the central processing unit corrects the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4) by polynomial regression according to the actual stress-strain distribution value of the bridge measured by the stress sensor arranged at the maximum stress of the bridge and the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4).
Specifically, the method for measuring the length of the passing detected vehicle by using the three-dimensional laser scanner comprises the following steps:
when the length of the passing detected vehicle is measured, if the head of the detected vehicle just enters the detection surface formed by the three-dimensional laser scanner, the timer used together with the three-dimensional laser scanner 11 starts to count, when the tail of the detected vehicle leaves the detection plane formed by the three-dimensional laser scanner, the timer stops counting, the speed of the detected vehicle is v in the process of passing through the three-dimensional laser scanner, the passing time of the detected vehicle recorded by the timer is t, and the length L of the detected vehicle can be expressed as l=vt.
It should be noted that the timer is set in the computer acquisition card; alternatively, the timer is juxtaposed with the three-dimensional laser scanner.
The method for measuring the passing detected vehicle width by using the three-dimensional laser scanner comprises the following steps:
when the width of the passing detected vehicle is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, in the polar coordinate system taking the three-dimensional laser scanner as the pole, the polar axis is along the vertical direction, when laser irradiates the leftmost position of the vehicle body of the detected vehicle, the distance measurement result is recorded as r1, the included angle between the laser line and the polar axis is a1, and then the distance from the left side of the detected vehicle to the polar axis direction is r 1 sin a 1 The method comprises the steps of carrying out a first treatment on the surface of the When the laser irradiates the rightmost position of the detected vehicle body, the distance measurement result is r2, the included angle between the laser line and the polar axis is a2, and the distance between the right side of the detected vehicle and the polar axis direction is r 2 sin a 2 The width w of the detected vehicle can be expressed as w=r 2 sin a 2 -r 1 sin a 1 。
The method for measuring the height of the passing detected vehicle by using the three-dimensional laser scanner comprises the following steps:
when the height of the detected vehicle body is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, in the polar coordinate system taking the three-dimensional laser scanner as the pole, the polar axis is along the horizontal direction, when laser irradiates the highest position of the detected vehicle, the distance measurement result is recorded as r, at the moment, the included angle between the laser line and the polar axis is a, and then the distance from the highest position of the detected vehicle to the polar axis is r sin a; the three-dimensional laser scanner is spaced from the ground by a distance H, and the detected vehicle height H can be expressed as h=h—r sin a.
The principle of the improved Otsu algorithm is as follows:
1) According to the formulaObtaining the distribution probability F of the image gray level 1 ，F 2 ，F 3 …F i . Wherein N is the total number of pixels, N i Is the number of pixels with a gray value i.
2) The average gray value μ of the image is calculated.
3) Let the gray level probability of the background image be P, and the average gray level value of the background image be mu 1 . The image variance is Var and the variance at the maximum value is maxVar. Let p=0, μ 1 ＝0，maxVar＝0。
4) i is iterated from 1 to 256, and the following values are calculated in turn:
P(i)＝P(i-1)+F i
μ 1 (i)＝(P(i-1)×μ 1 (i-1)+i×F i )/P(i)
the inter-class variance can be expressed as:
Var＝P·(μ 1 -μ) 2 +P 2 ·(μ 2 -μ) 2 (3.16)
wherein P is 2 Is the gray level probability of the foreground image, μ 2 Is the gray average of the foreground image. And because of
μ＝P·μ 1 +P 2 ·μ 2 (3.17)
Bringing formula 3.17 into formula 3.16 yields:
Var(i)＝P(i)×(μ-μ 1 (i)) 2 /(1-P(i))
5) Judging whether Var (i) is larger than maxVar, if so, making maxVar=Var (i), and T=i; if not, let i=i+1, return to step 4.
6) And finishing the iterative search, wherein T is the requirement.
The camera has a day-night conversion monitoring function, the effective pixels of the camera reach 300 ten thousand, the horizontal field angle is 80 degrees, the maximum image size reaches 1920 multiplied by 1080, the resolution ratio of video is 1296 horizontal multiplied by 732 vertical, and the video frame rate is 50Hz:25fps, 7W power supply and 12VDC power supply. The size of the material is 194.04 multiplied by 93.85 multiplied by 89.52mm, and the weight of the material is 1000g; the synchronization time precision of the camera is not less than 10 microseconds;
the cameras 12 are composed of a plurality of high resolution cameras, each covering 50 meters of the bridge deck, and the multiple cameras need to be precisely synchronized in time, so as to obtain dynamic changes of loads of different time spans of the bridge deck.
The specific algorithm for obtaining the load information of all vehicles to the bridge through the vehicle load calculation parameter information obtained in the step 1) and the distribution information of the whole bridge deck moving vehicle obtained in the step 3) is as follows:
The load F applied to the road surface by the whole vehicle can be divided into a vehicle static load F j And dynamic load F d I.e.
F＝F j +F d
Wherein the vehicle is under a static load F j I.e. gravity load.
In the process of solving the dynamic load of the vehicle, according to the D' Alemmbert principle, the motion equation of the vibration system of the vehicle is firstly obtained:
wherein: quality matrix
Damping matrix
Stiffness matrix
Displacement array y= [ y ] t ,y s ] T ；
Excitation arrayIn the above matrix, m s Is a sprung mass; m is m t Is a non-sprung mass; c s Damping for the suspension; c t Damping the tire. k (k) 1 The rigidity coefficient of the rear tire of the automobile; k (k) 2 The rigidity coefficient of the rear suspension of the automobile; k (k) t Is the tire stiffness; y is s ,y t Vertical absolute displacement for sprung and unsprung masses; y is r Is the vertical displacement of the road surface unevennessExcitation; Δh is the amount of vertical deformation of the road surface.
The random dynamic load of the vehicle to the road surface is as follows:
after the load of a single vehicle to the road surface is obtained, the total load of all vehicles running on the bridge to the bridge surface can be calculated by calculating the number of the vehicles on the bridge surface;
the dynamic load on the bridge deck can be obtained through the step, and the stress strain distribution value of the whole bridge is calculated by adopting a finite element method according to the bridge structure.
Referring to fig. 1 to 6, an automatic bridge dynamic load detection and inversion method includes the following steps:
1) On the same-direction double lanes of the bridge, using a three-dimensional laser scanner which is erected right above the bridge entrance and covers two lanes of the measuring range to obtain point cloud data of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge, sending the obtained point cloud data to a computer, obtaining three-dimensional contour diagrams of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge through three-dimensional reconstruction based on the point cloud data, calculating the length, width and height dimension data of all detected vehicles according to the three-dimensional contour diagrams of the surfaces of all detected vehicles, judging whether the detected vehicles are empty or not according to the carriage height in the dimension data of the detected vehicles, and obtaining load judgment information of the detected vehicles; meanwhile, a certain truck is selected as a sample vehicle, and the length, width and height dimension data of the truck are recorded; comparing the size data of the detected vehicle with the size data of the sample vehicle to judge the vehicle type information of the detected vehicle; the vehicle type information and the load judgment information of the detected vehicle constitute vehicle load calculation parameter information.
The step 1) is specifically as follows:
The three-dimensional laser scanner scans the top of the detected vehicle body and one side surface of the vehicle by utilizing the scanning light curtain constructed by the three-dimensional laser scanner, so that a plurality of continuous scanning cross sections of the detected vehicle are obtained, a three-dimensional contour map of the detected vehicle is drawn according to the plurality of continuous scanning cross sections, and different types of detected vehicles are distinguished from the three-dimensional contour map of the detected vehicle;
for flatbed trucks and tipping trucks, whether the truck is empty is determined by confirming whether the three-dimensional laser scanner is the shortest distance to the left or right edge of the truck.
Since trucks with larger loads are generally used for roads and bridges, the system focuses on truck detection. The vehicle type information specifically comprises: whether the vehicle is a cart or a trolley.
2) Shooting continuous running processes of all the detected vehicles moving under the bridge by using cameras which are erected on all uniformly distributed road rods and have a synchronous function with other cameras for covering the field of view of the area between two adjacent road rods on the same-direction double lanes of the bridge, after continuously running videos of the detected vehicles moving under the bridge are shot, sending the shot continuous running videos of the detected vehicles moving to a central processing unit, and processing the continuous running videos into continuous images of one frame by the central processing unit; extracting gray images of a plurality of moving detected vehicles from the images by using a background difference method, then carrying out self-adaptive binarization processing on the gray images of the plurality of moving detected vehicles by using an improved Otsu algorithm, extracting Harris characteristic corner points from target results of the plurality of moving detected vehicles after the binarization processing, creating a plurality of tracking frames of the sizes of the self-adaptive moving detected vehicles according to the extracted Harris characteristic corner points to start tracking the moving detected vehicles in a view field, obtaining the quantity of the vehicles distributed in the view field according to the quantity of the tracking frames, and simultaneously, marking the moving detected vehicles by using the tracking frames of the sizes of the moving detected vehicles which are obtained by utilizing continuous driving videos of the moving detected vehicles which are shot by cameras arranged at the entrance of the bridge in the same direction and double lanes. Specifically, the road rod interval 50m is set, the camera is a sea-Kangwei video camera, and the camera is erected at the road rod height of 10 meters, so that the coverage of the camera to the field of view of the 50m area is ensured.
3) The central processing unit calculates the average value of the horizontal and vertical coordinates of the Harris characteristic angular points of each detected vehicle moving in the extracted Harris characteristic angular points to obtain the mass center of each detected vehicle moving, and meanwhile, the mass centers of vehicles with the same marks in each frame of video image are connected through the nearest neighbor tracking filter algorithm and the improved Kalman filtering algorithm for tracking and judging the positions of the vehicles, so that the running track of the vehicles with the same marks is obtained, the tracking of the detected vehicles with multi-target movements is realized, and the distribution information of the vehicles moving on the whole bridge deck is obtained;
4) And the central processing unit is used for obtaining the load information of all vehicles on the bridge according to the vehicle load calculation parameter information obtained in the step 1) and the distribution information of the whole bridge deck moving vehicle obtained in the step 3), and obtaining the stress strain distribution value of each point of the bridge by adopting a finite element model algorithm through the obtained load information and the related parameter information of the bridge during bridge design.
5) And (3) correcting the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4) by the central processing unit through polynomial regression according to the actual stress-strain distribution value of the bridge measured by the stress sensor arranged at the maximum stress of the bridge and the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4), thereby realizing inversion.
Specifically, the method for measuring the length of the passing detected vehicle by using the three-dimensional laser scanner comprises the following steps:
when the length of the passing detected vehicle is measured, if the head of the detected vehicle just enters the detection surface formed by the three-dimensional laser scanner, the timer used together with the three-dimensional laser scanner 11 starts to count, when the tail of the detected vehicle leaves the detection plane formed by the three-dimensional laser scanner, the timer stops counting, the speed of the detected vehicle is v in the process of passing through the three-dimensional laser scanner, the passing time of the detected vehicle recorded by the timer is t, and the length L of the detected vehicle can be expressed as l=vt.
It should be noted that the timer is set in the computer acquisition card; alternatively, the timer is juxtaposed with the three-dimensional laser scanner.
The method for measuring the passing detected vehicle width by using the three-dimensional laser scanner comprises the following steps:
when the width of the passing detected vehicle is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, in the polar coordinate system taking the three-dimensional laser scanner as the pole, the polar axis is along the vertical direction, when laser irradiates the leftmost position of the vehicle body of the detected vehicle, the distance measurement result is recorded as r1, the included angle between the laser line and the polar axis is a1, and then the distance from the left side of the detected vehicle to the polar axis direction is r 1 sin a 1 The method comprises the steps of carrying out a first treatment on the surface of the When the laser irradiates the rightmost position of the detected vehicle body, the distance measurement result is r2, the included angle between the laser line and the polar axis is a2, and the distance between the right side of the detected vehicle and the polar axis direction is r 2 sin a 2 The width w of the detected vehicle can be expressed as w=r 2 sin a 2 -r 1 sin a 1 。
The method for measuring the height of the passing detected vehicle by using the three-dimensional laser scanner comprises the following steps:
when the height of the detected vehicle body is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, in the polar coordinate system taking the three-dimensional laser scanner as the pole, the polar axis is along the horizontal direction, when laser irradiates the highest position of the detected vehicle, the distance measurement result is recorded as r, at the moment, the included angle between the laser line and the polar axis is a, and then the distance from the highest position of the detected vehicle to the polar axis is r sin a; the three-dimensional laser scanner is spaced from the ground by a distance H, and the detected vehicle height H can be expressed as h=h—r sin a.
The bridge dynamic load automatic detection and inversion system carries out self-adaptive threshold binarization on a moving vehicle before the vehicle is tracked, and an improved Otsu algorithm is used, so that the advantages of the Otsu algorithm in terms of real-time performance are enhanced, and certain detected separated communication areas belonging to the same moving vehicle can be connected. The principle of the improved Otsu algorithm is as follows:
1)According to the formulaObtaining the distribution probability F of the image gray level 1 ，F 2 ，F 3 …F i . Wherein N is the total number of pixels, N i Is the number of pixels with a gray value i.
2) The average gray value μ of the image is calculated.
3) Let the gray level probability of the background image be P, and the average gray level value of the background image be mu 1 . The image variance is Var and the variance at the maximum value is maxVar. Let p=0, μ 1 ＝0，maxVar＝0。
4) i is iterated from 1 to 256, and the following values are calculated in turn:
P(i)＝P(i-1)+F i
μ 1 (i)＝(P(i-1)×μ 1 (i-1)+i×F i )/P(i)
the inter-class variance can be expressed as:
Var＝P·(μ 1 -μ) 2 +P 2 ·(μ 2 -μ) 2 (3.16)
wherein P is 2 Is the gray level probability of the foreground image, μ 2 Is the gray average of the foreground image. And because of
μ＝P·μ 1 +P 2 ·μ 2 (3.17)
Bringing formula 3.17 into formula 3.16 yields:
Var(i)＝P(i)×(μ-μ 1 (i)) 2 /(1-P(i))
5) Judging whether Var (i) is larger than maxVar, if so, making maxVar=Var (i), and T=i; if not, let i=i+1, return to step 4.
6) And finishing the iterative search, wherein T is the requirement.
The novel ocean health vision camera has a day-night conversion monitoring function, the effective pixels of the novel ocean health vision camera reach 300 ten thousand, the horizontal field angle is 80 degrees, the maximum image size reaches 1920 multiplied by 1080, the resolution ratio of video is 1296 horizontal multiplied by 732 vertical, and the video frame rate is 50Hz:25fps, 7W power supply and 12VDC power supply. The size of the material is 194.04 multiplied by 93.85 multiplied by 89.52mm, and the weight of the material is 1000g; the synchronization time precision of the camera is not less than 10 microseconds;
The cameras 12 are composed of a plurality of high resolution cameras, each covering 50 meters of the bridge deck, and the multiple cameras need to be precisely synchronized in time, so as to obtain dynamic changes of loads of different time spans of the bridge deck.
When the vehicle passes through the bridge, the vehicle can generate pressure on the bridge deck to deform the bridge deck due to the self weight of the vehicle and the condition of loading heavy objects; the specific algorithm for obtaining the load information of all vehicles to the bridge through the vehicle load calculation parameter information obtained in the step 1) and the distribution information of the whole bridge deck moving vehicle obtained in the step 3) is as follows:
the load F applied to the road surface by the whole vehicle can be divided into a vehicle static load F j And dynamic load F d I.e.
F＝F j +F d
Wherein the vehicle is under a static load F j I.e. gravity load.
In the process of solving the dynamic load of the vehicle, according to the D' Alemmbert principle, the motion equation of the vibration system of the vehicle is firstly obtained:
wherein: quality matrix
Damping matrix
Stiffness matrix
Displacement array y= [ y ] t ,y s ] T ；
Excitation arrayIn the above matrix, m s Is a sprung mass; m is m t Is a non-sprung mass; c s Damping for the suspension; c t Damping the tire. k (k) 1 The rigidity coefficient of the rear tire of the automobile; k (k) 2 The rigidity coefficient of the rear suspension of the automobile; k (k) t Is the tire stiffness; y is s ,y t Vertical absolute displacement for sprung and unsprung masses; y is r Exciting vertical displacement of the road surface unevenness; Δh is the amount of vertical deformation of the road surface. />
The random dynamic load of the vehicle to the road surface is as follows:
after the load of a single vehicle to the road surface is obtained, the total load of all vehicles running on the bridge to the bridge surface can be calculated by calculating the number of the vehicles on the bridge surface;
the dynamic load on the bridge deck can be obtained through the step, and the stress strain distribution value of the whole bridge is calculated by adopting a finite element method according to the bridge structure.
When the vehicle runs on the bridge, the pressure of the vehicle on the bridge deck changes due to the weight of the vehicle, especially for trucks with overlarge load, the load of the vehicle on the bridge can be obtained by acquiring the relevant parameters of the vehicle or the bridge, and a finite element model algorithm is adopted.
Because of the many bridge forms, the difference of the vehicle types is also larger, and the stress strain value of each point of the bridge beam calculated by the method and the actual detection have errors. Therefore, stress sensors are arranged at key parts of the bridge in the system, and structural parameters of the bridge are continuously corrected according to the difference value calculated by the actual detection and the model until the errors of the bridge and the structural parameters meet a certain range.
The scanning light curtain constructed by the three-dimensional laser scanner can scan the top of the vehicle body and one side surface of the vehicle, so that a plurality of continuous scanning cross sections are obtained, and the cross section data establishes a good foundation for drawing the outline. After drawing the three-dimensional profile of the vehicle, different types of vehicles, such as cars, SUVs, trucks, and the like, can be identified.
For flatbed trucks and dumped trucks, it is also possible to determine if the truck is empty by confirming if the three-dimensional laser scanner is the shortest distance to the left or right edge of the vehicle.
Typically, flatbed trucks are loaded with fixed-shape objects such as containers, other vehicles, and the like. When the three-dimensional laser scanner is at the upper left of the vehicle, if the flat wagon is in an empty state, the distance from the range finder to the left edge of the flat wagon of the vehicle is the shortest from the range finder to any point of the flat wagon of the vehicle; if the flat wagon is loaded with goods, the distance from the distance meter to a certain point of the goods is smaller than the distance from the distance meter to the left edge of the flat wagon. And trucks with skips are generally loaded with bulk goods such as silt, stones and bricks, etc. that vary in overall shape. The method of determining whether the dump truck is empty is similar to a flatbed truck.
A stress-strain sensor is provided at the maximum strain of each bridge span. When the strain exceeds the threshold value, inversion is carried out on the obtained stress strain value of the whole bridge, and the reliability of early warning can be increased.
The invention organically integrates vehicle type detection, vehicle distribution, vehicle tracking and strain stress monitoring into a monitoring platform with feedback effect by adopting a multi-sensor fusion technology.
According to the bridge dynamic load automatic detection and inversion system and method provided by the invention, the vehicle monitoring and the strain measurement are unified, the bridge dynamic load automatic detection is realized, and a plurality of videos are synchronized and the full-bridge vehicle is tracked.
The bridge dynamic load automatic detection and inversion system and method are characterized in that a three-dimensional laser scanner is utilized to obtain a point cloud image of each detected vehicle entering the movement of a bridge, the three-dimensional reconstruction is carried out on each section, the model of the detected vehicle and the load judgment information of the vehicle are judged, a group of high-resolution camera combinations which are accurately synchronous and are distributed at equal intervals and cover the whole bridge deck are utilized, the vehicle dynamic distribution on each span structure of the bridge deck is obtained through a vehicle segmentation and track tracking algorithm, the model information of each vehicle and the load judgment information of the vehicle are combined, the static load of the bridge deck is obtained through calculation, and the dynamic stress distribution of each point of the whole bridge deck is obtained through a bridge finite element solving method; the stress sensor is installed at the key position of the bridge and is used as a correction reference, the calculation result of the finite element is inverted, and the strain of each position of the bridge is accurately monitored finally.
The foregoing has shown and described the basic principles and main features of the present invention and the advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (9)

1. The automatic bridge dynamic load detection and inversion system is characterized by comprising a three-dimensional laser scanner (11), a computer, a plurality of cameras (12) and a timer used for being matched with the three-dimensional laser scanner (11); a plurality of road rods are uniformly distributed on the same-direction lanes on the bridge, each road rod is provided with a camera (12) for covering the field of view of the lane area between two adjacent road rods, the cameras synchronously pick up images, the synchronous precision is less than 10 microseconds, and a three-dimensional laser scanner (11) is arranged right above the same-direction double lanes at the entrance of the bridge and the measuring range covers two lanes; the bridge (21) is provided with a stress sensor at least at the position of maximum stress of the bridge, and the stress sensor, the camera (12), the timer and the three-dimensional laser scanner (11) are all connected with a computer;
The three-dimensional laser scanner (11) is used for obtaining point cloud data of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge and sending the obtained point cloud data to the computer;
the camera (12) is used for shooting the continuous running process of all the detected vehicles moving below the camera, and after the continuous running video of the detected vehicles moving is shot, the shot continuous running video of the detected vehicles moving is sent to the central processing unit;
the computer is used for receiving the point cloud data sent by the three-dimensional laser scanner, obtaining three-dimensional contour maps of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge through three-dimensional reconstruction based on the point cloud data, calculating the length, width and height dimension data of each detected vehicle according to the three-dimensional contour maps of the surfaces of each detected vehicle, and judging whether the detected vehicle is empty according to the carriage height in the dimension data of the detected vehicle, so as to obtain load judgment information of the detected vehicle; meanwhile, a certain truck is selected as a sample vehicle, and the length, width and height dimension data of the truck are recorded; comparing the size data of the detected vehicle with the size data of the sample vehicle to judge the vehicle type information of the detected vehicle; the vehicle type information and the load judgment information of the detected vehicle form vehicle load calculation parameter information;
Meanwhile, the computer is used for receiving the continuous running video of the detected vehicle of the motion shot by the camera and processing the continuous running video into a frame of uninterrupted image; extracting gray images of a plurality of moving detected vehicles from the images by using a background difference method, then carrying out self-adaptive binarization processing on the gray images of the plurality of moving detected vehicles by using an improved Otsu algorithm, extracting Harris feature angular points from target results of the plurality of moving detected vehicles after the binarization processing, creating a plurality of tracking frames of the sizes of the self-adaptive moving detected vehicles according to the extracted Harris feature angular points to start tracking the moving detected vehicles in a view field, obtaining the quantity of the vehicles distributed in the view field according to the quantity of the tracking frames, and simultaneously, marking the moving detected vehicles by using continuous running videos of the moving detected vehicles shot by cameras arranged at the same-direction double-lane entrance of a bridge by using the obtained tracking frames of the sizes of the moving detected vehicles;
the centroid of each moving detected vehicle is obtained through the average value of the horizontal coordinates and the vertical coordinates of each moving detected vehicle in the Harris characteristic angular points, and meanwhile, the centroids of vehicles with the same labels in each frame of video images are connected through the nearest neighbor tracking filter algorithm and the improved Kalman filtering algorithm for tracking and judging the positions of the vehicles, so that the running track of the vehicles with the same labels is obtained, the tracking of the detected vehicles with multi-target movements is realized, and the distribution information of the vehicles moving on the whole bridge deck is obtained;
Obtaining the load information of all vehicles on the bridge according to the obtained vehicle load calculation parameter information and the obtained distribution information of the whole bridge deck moving vehicle, and obtaining the stress strain distribution value of each point of the bridge by adopting a finite element model algorithm according to the obtained load information and the related parameter information of the bridge during bridge design;
and correcting the obtained stress-strain distribution value of each point of the corresponding part of the bridge according to the actual stress-strain distribution value of the bridge measured by the stress sensor arranged at the maximum stress position of the bridge and the stress-strain distribution value of each point of the corresponding part of the bridge obtained in the step through polynomial regression.
2. The automatic bridge dynamic load detection and inversion method is characterized by comprising the following steps of:
(1) On the same-direction double lanes of the bridge, utilizing a three-dimensional laser scanner which is erected right above the bridge entrance and covers two lanes of the measuring range to obtain point cloud data of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge, sending the obtained point cloud data to a computer, obtaining three-dimensional contour diagrams of the surfaces of all detected vehicles entering the same-direction double lanes of the bridge through three-dimensional reconstruction based on the point cloud data, calculating length, width and height dimension data of all detected vehicles according to the three-dimensional contour diagrams of the surfaces of all detected vehicles, and judging whether the detected vehicles are empty or not according to the carriage height in the dimension data of the detected vehicles, thereby obtaining load judgment information of the detected vehicles; meanwhile, a certain truck is selected as a sample vehicle, and the length, width and height dimension data of the truck are recorded; comparing the size data of the detected vehicle with the size data of the sample vehicle to judge the vehicle type information of the detected vehicle; the vehicle type information and the load judgment information of the detected vehicle form vehicle load calculation parameter information;
(2) Shooting continuous running processes of all the detected vehicles moving under the bridge by using cameras which are erected on all uniformly distributed road rods and have a synchronous function with other cameras for covering the field of view of the area between two adjacent road rods on the same-direction double lanes of the bridge, after continuously running videos of the detected vehicles moving under the bridge are shot, sending the shot continuous running videos of the detected vehicles moving to a central processing unit, and processing the continuous running videos into continuous images of one frame by the central processing unit; extracting gray images of a plurality of moving detected vehicles from the images by using a background difference method, then carrying out self-adaptive binarization processing on the gray images of the plurality of moving detected vehicles by using an improved Otsu algorithm, extracting Harris feature angular points from target results of the plurality of moving detected vehicles after the binarization processing, creating a plurality of tracking frames of the sizes of the self-adaptive moving detected vehicles according to the extracted Harris feature angular points to start tracking the moving detected vehicles in a view field, obtaining the quantity of the vehicles distributed in the view field according to the quantity of the tracking frames, and simultaneously, marking the moving detected vehicles by using continuous running videos of the moving detected vehicles shot by cameras arranged at the same-direction double-lane entrance of a bridge by using the obtained tracking frames of the sizes of the moving detected vehicles;
(3) The central processing unit calculates the average value of the horizontal and vertical coordinates of the Harris characteristic angular points of each detected vehicle moving in the extracted Harris characteristic angular points to obtain the mass center of each detected vehicle moving, and meanwhile, the mass centers of vehicles with the same marks in each frame of video image are connected through the nearest neighbor tracking filter algorithm and the improved Kalman filtering algorithm for tracking and judging the positions of the vehicles, so that the running track of the vehicles with the same marks is obtained, the tracking of the detected vehicles with multi-target movements is realized, and the distribution information of the vehicles moving on the whole bridge deck is obtained;
(4) The central processing unit is used for obtaining the load information of all vehicles on the bridge according to the vehicle load calculation parameter information obtained in the step 1) and the distribution information of the whole bridge deck moving vehicle obtained in the step 3), and obtaining the stress strain distribution value of each point of the bridge by adopting a finite element model algorithm according to the obtained load information and the related parameter information of the bridge during bridge design;
(5) And (3) the central processing unit corrects the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4) by polynomial regression according to the actual stress-strain distribution value of the bridge measured by the stress sensor arranged at the maximum stress of the bridge and the stress-strain distribution value of each point of the corresponding part of the bridge in the step (4).
3. The method for automatically detecting and inverting the bridge dynamic load according to claim 2, wherein the method for measuring the length of the passing detected vehicle by using a three-dimensional laser scanner is as follows:
when the length of the passing detected vehicle is measured, if the head of the detected vehicle just enters a detection surface formed by the three-dimensional laser scanner, a timer used together with the three-dimensional laser scanner (11) starts to count, when the tail of the detected vehicle leaves the detection surface formed by the three-dimensional laser scanner, the timer stops counting, the speed of the detected vehicle in the process of passing through the three-dimensional laser scanner is v, the passing time of the detected vehicle recorded by the timer is t, and the length L of the detected vehicle can be expressed as L=vt.
4. The method for automatically detecting and inverting the bridge dynamic load according to claim 2, wherein the method for measuring the width of the passing detected vehicle by using a three-dimensional laser scanner is as follows:
when the width of the passing detected vehicle is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, and the three-dimensional laser scanner is taken as the poleIn the polar coordinate system, the polar axis is along the vertical direction, when laser emitted by the three-dimensional laser scanner irradiates the leftmost position of the vehicle body of the detected vehicle, the distance measurement result is recorded as r1, and the included angle between the laser emitted by the three-dimensional laser scanner and the polar axis is a1, then the distance between the left side of the detected vehicle and the polar axis direction is r 1 sina 1 The method comprises the steps of carrying out a first treatment on the surface of the When the laser irradiates the rightmost position of the detected vehicle body, the distance measurement result is r2, the included angle between the laser line and the polar axis is a2, and the distance between the right side of the detected vehicle and the polar axis direction is r 2 sina 2 The width w of the detected vehicle can be expressed as w=r 2 sina 2 -r 1 sina 1 。
5. The automatic bridge dynamic load detection and inversion method according to claim 2, wherein the method for measuring the height of the passing detected vehicle by using a three-dimensional laser scanner comprises the following steps:
when the height of the detected vehicle body is measured, a polar coordinate system is established by taking the three-dimensional laser scanner as a pole, in the polar coordinate system taking the three-dimensional laser scanner as the pole, the polar axis is along the horizontal direction, when laser emitted by the three-dimensional laser scanner irradiates the highest position of the detected vehicle, a distance measurement result is recorded as r, at the moment, the included angle between the laser emitted by the three-dimensional laser scanner and the polar axis is a, and the distance from the highest position of the detected vehicle to the polar axis is rsina; the three-dimensional laser scanner is spaced from the ground by a distance H, the detected vehicle height H may be expressed as h=h-rsina.
6. The method for automatically detecting and inverting the bridge dynamic load according to claim 2, wherein the step 1) is specifically:
The three-dimensional laser scanner scans the top of the detected vehicle body and one side surface of the vehicle by utilizing the scanning light curtain constructed by the three-dimensional laser scanner, so that a plurality of continuous scanning cross sections of the detected vehicle are obtained, a three-dimensional contour map of the detected vehicle is drawn according to the plurality of continuous scanning cross sections, and different types of detected vehicles are distinguished from the three-dimensional contour map of the detected vehicle;
for flatbed trucks and tipping trucks, whether the truck is empty is determined by confirming whether the three-dimensional laser scanner is the shortest distance to the left or right edge of the truck.
7. The automatic bridge dynamic load detection and inversion method according to claim 2, wherein the principle of the improved Otsu algorithm is as follows:
1) According to the formulaObtaining the distribution probability F of the image gray level 1 ,F 2 ,F 3 …F i The method comprises the steps of carrying out a first treatment on the surface of the Wherein N is the total number of pixels, N i The number of pixels with gray value i;
2) Calculating an average gray value mu of the image;
3) Let the gray level probability of the background image be P, and the average gray level value of the background image be mu 1 The method comprises the steps of carrying out a first treatment on the surface of the The image variance is Var, and the variance when the numerical value is maximum is maxVar; let p=0, μ 1 ＝0，maxVar＝0；
4) i is iterated from 1 to 256, and the following values are calculated in turn:
P(i)＝P(i-1)+F i
μ 1 (i)＝(P(i-1)×μ 1 (i-1)+i×F i )/P(i)
The inter-class variance can be expressed as:
Var＝P·(μ 1 -μ) 2 +P 2 ·(μ 2 -μ) 2 (3.16)
wherein P is 2 Is the gray level probability of the foreground image, μ 2 Is the gray average value of the foreground image; and because of
v＝P·μ 1 +P 2 ·μ 2 (3.17)
Bringing formula 3.17 into formula 3.16 yields:
Var(i)＝P(i)×(μ-μ 1 (i)) 2 /(1-P(i))
5) Judging whether Var (i) is larger than maxVar, if so, making maxVar=Var (i), and T=i; if not, let i=i+1, return to step 4;
6) And finishing the iterative search, wherein T is the requirement.
8. The automatic bridge dynamic load detection and inversion method according to claim 7, wherein said camera has a day-to-night conversion monitoring function, effective pixels of which reach 300 ten thousand, a horizontal angle of view of 80 °, a maximum image size of 1920×1080, resolution of 1296 (horizontal) ×732 (vertical) for video, a video frame rate of 50hz:25fps, a power supply of 7W, and a power supply voltage of 12VDC; the size of the material is 194.04 multiplied by 93.85 multiplied by 89.52mm, and the weight of the material is 1000g; the synchronization time precision of the camera is not less than 10 microseconds;
the cameras (12) are composed of a plurality of high resolution, each camera covers 50 meters of the bridge deck, and the multiple cameras need to be precisely synchronized in time, so that dynamic changes of loads of each bridge span of the bridge deck are obtained.
9. The automatic bridge dynamic load detection and inversion method according to claim 2, wherein the specific algorithm for obtaining the load information of all vehicles to the bridge by the vehicle load calculation parameter information obtained in step 1) and the distribution information of the whole bridge deck moving vehicle obtained in step 3) is as follows:
The load F applied to the road surface by the whole vehicle can be divided into a vehicle static load F j And dynamic load F d I.e.
F＝F j +F d
Wherein the vehicle is under a static load F j I.e. gravity load;
in the process of solving the dynamic load of the vehicle, according to the D' Alemmbert principle, the motion equation of the vibration system of the vehicle is firstly obtained:
wherein: quality matrix
Damping matrix
Stiffness matrix
Displacement array y= [ y ] t ,y s ] T ；
Excitation arrayIn the above matrix, m s Is a sprung mass; m is m t Is a non-sprung mass; c s Damping for the suspension; c t Damping the tire; k (k) 1 The rigidity coefficient of the rear tire of the automobile; k (k) 2 The rigidity coefficient of the rear suspension of the automobile; k (k) t Is the tire stiffness; y is s ,y t Vertical absolute displacement for sprung and unsprung masses; y is r Exciting vertical displacement of the road surface unevenness; Δh is the vertical deformation of the road surface;
the random dynamic load of the vehicle to the road surface is as follows:
after the load of a single vehicle to the road surface is obtained, the total load of all vehicles running on the bridge to the bridge surface can be calculated by calculating the number of the vehicles on the bridge surface;
the dynamic load on the bridge deck can be obtained through the step, and the stress strain distribution value of the whole bridge is calculated by adopting a finite element method according to the bridge structure.
CN202010188342.6A 2020-03-17 2020-03-17 Automatic bridge dynamic load detection and inversion system and method Active CN111354035B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010188342.6A CN111354035B (en) 2020-03-17 2020-03-17 Automatic bridge dynamic load detection and inversion system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010188342.6A CN111354035B (en) 2020-03-17 2020-03-17 Automatic bridge dynamic load detection and inversion system and method

Publications (2)

Publication Number Publication Date
CN111354035A CN111354035A (en) 2020-06-30
CN111354035B true CN111354035B (en) 2023-09-05

Family

ID=71197510

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010188342.6A Active CN111354035B (en) 2020-03-17 2020-03-17 Automatic bridge dynamic load detection and inversion system and method

Country Status (1)

Country Link
CN (1) CN111354035B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112362149B (en) * 2020-09-21 2022-01-18 中铁第四勘察设计院集团有限公司 Method and system for dynamically identifying vehicle axle load based on vertical displacement influence surface loading
CN112387982B (en) * 2020-10-21 2021-10-12 上海交通大学 Laser additive process power combined regulation and control method
CN113077632A (en) * 2021-06-07 2021-07-06 四川紫荆花开智能网联汽车科技有限公司 V2X intelligent network connection side system and realizing method
CN114993544A (en) * 2022-07-18 2022-09-02 安徽建筑大学 Bridge structure stress detection device
CN115506431A (en) * 2022-09-16 2022-12-23 福建晋工新能源科技有限公司 Automatic material shoveling system and automatic material shoveling control method of automatic material shoveling system
CN115265378A (en) * 2022-09-23 2022-11-01 北京东方国信科技股份有限公司 Detection system for three-dimensional size of carriage

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002101328A1 (en) * 2001-06-12 2002-12-19 Hans Heister Method for recording the surface of a roadway
JP2011257256A (en) * 2010-06-09 2011-12-22 Fukken Co Ltd Method of measuring altitude of bridge in no-live-load state
US8306747B1 (en) * 2007-01-19 2012-11-06 Starodub, Inc. Travel way measurement system
CN103593678A (en) * 2013-10-16 2014-02-19 长安大学 Long-span bridge vehicle dynamic load distribution detection method
CN104598753A (en) * 2015-02-06 2015-05-06 华北水利水电大学 Bridge moving vehicle load recognition method based on Brakhage V method
CN106123797A (en) * 2016-08-29 2016-11-16 北京交通大学 Bridge floor deflection metrology system based on image procossing and method
CN108460231A (en) * 2018-03-23 2018-08-28 中交公路长大桥建设国家工程研究中心有限公司 A kind of bridge builds foster overall process intellectual monitoring assessment early warning decision system and method
CN109167956A (en) * 2018-05-21 2019-01-08 同济大学 The full-bridge face traveling load spatial distribution merged based on dynamic weighing and more video informations monitors system
CN109635386A (en) * 2018-11-27 2019-04-16 中电建冀交高速公路投资发展有限公司 A kind of bridge move vehicle Load Identification Methods

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002101328A1 (en) * 2001-06-12 2002-12-19 Hans Heister Method for recording the surface of a roadway
US8306747B1 (en) * 2007-01-19 2012-11-06 Starodub, Inc. Travel way measurement system
JP2011257256A (en) * 2010-06-09 2011-12-22 Fukken Co Ltd Method of measuring altitude of bridge in no-live-load state
CN103593678A (en) * 2013-10-16 2014-02-19 长安大学 Long-span bridge vehicle dynamic load distribution detection method
CN104598753A (en) * 2015-02-06 2015-05-06 华北水利水电大学 Bridge moving vehicle load recognition method based on Brakhage V method
CN106123797A (en) * 2016-08-29 2016-11-16 北京交通大学 Bridge floor deflection metrology system based on image procossing and method
CN108460231A (en) * 2018-03-23 2018-08-28 中交公路长大桥建设国家工程研究中心有限公司 A kind of bridge builds foster overall process intellectual monitoring assessment early warning decision system and method
CN109167956A (en) * 2018-05-21 2019-01-08 同济大学 The full-bridge face traveling load spatial distribution merged based on dynamic weighing and more video informations monitors system
CN109635386A (en) * 2018-11-27 2019-04-16 中电建冀交高速公路投资发展有限公司 A kind of bridge move vehicle Load Identification Methods

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
周志祥 ; 姜腾蛟 ; 唐亮 ; 楚玺 ; 杨晶 ; .移动式三维激光扫描系统在桥面全息变形监测的应用研究.应用基础与工程科学学报.2018,(05),全文. *

Also Published As

Publication number Publication date
CN111354035A (en) 2020-06-30

Similar Documents

Publication Publication Date Title
CN111354035B (en) Automatic bridge dynamic load detection and inversion system and method
CN109064495B (en) Bridge deck vehicle space-time information acquisition method based on fast R-CNN and video technology
CA2644006C (en) Systems and methods for obtaining improved accuracy measurements of moving rolling stock components
CN107314741A (en) Measurement of cargo measuring method
CN104599249B (en) Cableway platform bridge floor car load is distributed real-time detection method
CN101223416A (en) Object detection device
CN103718224A (en) 3d object detector and 3d object detection method
CN103886757A (en) Method For Automatically Classifying Moving Vehicles
CN103196418A (en) Measuring method of vehicle distance at curves
CN107843186A (en) A kind of compartment loading condition detecting system and method
CN111694011A (en) Road edge detection method based on data fusion of camera and three-dimensional laser radar
CN104132818A (en) Three-dimensional four-wheel locator and measuring method thereof
US9098774B2 (en) Method for detection of targets in stereoscopic images
CN112161685B (en) Vehicle load measuring method based on surface characteristics
CN107621229B (en) Real-time railway track width measurement system and method based on area array black-and-white camera
CN111551122A (en) Train wagon number and length measuring system and method based on laser radar
CN105780631A (en) Push-type cement concrete pavement joint seam damage detection device
CN112278891B (en) Carriage internal attitude detection method
CN111023975B (en) Device and method for detecting ore volume of dump truck for strip mine
CN206905735U (en) One kind bridge joint slope sedimentation automatic detection device and vehicle
CN114022537A (en) Vehicle loading rate and unbalance loading rate analysis method for dynamic weighing area
CN108507529B (en) Bridge slope subsides automatic checkout device and vehicle
CN207751450U (en) Road detection apparatus and system
RU2486597C1 (en) Method of automatic classification of vehicles
CN104197851B (en) Vehicle-mounted bridge floor morphosis safety detecting system and method

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