CN113776462B - Three-dimensional shape detection method for high-speed rail ballastless track bearing platform based on digital image - Google Patents
Three-dimensional shape detection method for high-speed rail ballastless track bearing platform based on digital image Download PDFInfo
- Publication number
- CN113776462B CN113776462B CN202111073961.1A CN202111073961A CN113776462B CN 113776462 B CN113776462 B CN 113776462B CN 202111073961 A CN202111073961 A CN 202111073961A CN 113776462 B CN113776462 B CN 113776462B
- Authority
- CN
- China
- Prior art keywords
- camera
- coordinate system
- prism
- total station
- control network
- 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
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 76
- 238000005259 measurement Methods 0.000 claims abstract description 43
- 239000013256 coordination polymer Substances 0.000 claims abstract description 41
- 238000000034 method Methods 0.000 claims description 27
- 230000003287 optical effect Effects 0.000 claims description 14
- 230000009466 transformation Effects 0.000 claims description 12
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 description 9
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The utility model discloses a three-dimensional shape detection method of a high-speed railway ballastless track rail bearing platform based on digital images, which comprises a detection system formed by a binocular vision camera and a total station prism measurement unit, wherein prism groups in the binocular vision camera and the total station prism measurement unit are fixedly arranged on a measurement rack together, and camera coordinates are established; reconstructing three-dimensional point cloud contour data of the rail bearing platform by using a digital image which is obtained by shooting by a binocular vision camera and contains natural textures of the surface of the rail bearing platform; and the total station prism measurement unit obtains the position of each prism in the prism group in the CP III control network, establishes the relation between the camera coordinate system and the CP III control network, and unifies the three-dimensional point cloud contour data of the rail bearing platform under the CP III control network so as to realize the three-dimensional shape detection of the rail bearing platform. The utility model greatly improves the detection efficiency, and the measurement accuracy meets the requirement of realizing working conditions.
Description
Technical Field
The utility model relates to construction of a ballastless track, in particular to a method for detecting three-dimensional morphology of a ballastless track bearing platform.
Background
The high-speed railway has high operation speed and high density, and has high requirements on smoothness and stability of the track. In order to ensure the smoothness and stability of the high-speed railway track, the three-dimensional space morphology of the rail bearing table on the track plate is required to be measured before the track is paved in the construction of the high-speed railway track.
At present, the detection of the rail bearing table in practical application is mainly based on manual measurement, the measurement efficiency is low, the labor intensity is high, and the detection requirement of a modern high-speed railway cannot be met.
In the patent application document with the authorized bulletin number of CN 206208175U and the name of CRTSIII type track plate rail bearing table clamp mouth measuring tool, a rail bearing table clamp mouth measuring tool is disclosed, when in field measurement, the measuring tool is placed in contact with a rail bearing table slot mouth, a total station is used for measuring prisms on the measuring tool, the measuring tool is moved and detected on each rail bearing table one by one until the detection of all rail bearing tables on a CPTSIII type track plate is completed. The method is simple to operate, but has low efficiency, and is difficult to meet the requirements of high-efficiency and high-precision measurement on the construction site.
In the patent application document with publication number of CN 105906219A and named as a method for detecting the deviation of the external dimension of a CRTSIII type track plate, a detection method adopting the combination of an absolute laser tracker and a handheld laser scanner is disclosed, and the surface data of the track plate is rapidly acquired. However, the method is only aimed at factory detection of a track plate processing workshop, cannot be suitable for detection of a track bearing table on a track plate on a construction site, and meanwhile, the absolute laser tracker and the handheld laser tracker which are used lack of use conditions in civil engineering and iron construction industries.
In the chinese patent application document with publication number CN 111486831A, entitled "detection device and method for ballastless track rail-bearing table measurement tool", a method for measuring critical dimension information of a rail-bearing table by using a rail-bearing table detection mold containing a precision prism rod is disclosed, and the measurement mold is elastically connected to a lifting bracket for realizing automatic mold placement. The method reduces manual intervention to a certain extent, but the position of the measuring die needs to be continuously adjusted in each measurement, the measurement process is long, and meanwhile, the measurement information only has the elevation data of the track, so that the information quantity is small, and the requirement of actual construction on the omnibearing and multidimensional three-dimensional morphology data is difficult to meet.
Disclosure of Invention
The utility model provides a three-dimensional shape detection method of a high-speed railway unbiased rail bearing platform based on digital images, which is used for realizing high-efficiency and high-precision spatial shape measurement of a plurality of bearing platforms in an outdoor environment by fusing an optical measurement method and comprises single bearing platform shape measurement and relative position measurement of a plurality of bearing platforms; the measurement data can be used for fitting and predicting the track three-dimensional trend after the track is paved, guiding the subsequent construction and material purchase, and improving the track paving efficiency and quality.
The utility model adopts the following technical scheme for solving the technical problems:
the utility model relates to a three-dimensional shape detection method of a high-speed railway ballastless track rail bearing platform based on a digital image, which is characterized by comprising the following steps:
a detection system is arranged and comprises a binocular vision camera and a total station prism measuring unit;
the binocular vision camera consists of two cameras, namely a first camera and a second camera, which are fixedly arranged on the same height position of the detection frame and are positioned at the left side and the right side of the detection frame;
the total station prism measuring unit consists of a total station and a prism group fixedly arranged on the detection frame; the prism group comprises at least 4 prisms;
obtaining a digital image containing natural textures of the surface of the rail bearing platform by shooting through the binocular vision camera, and reconstructing three-dimensional point cloud contour data of the rail bearing platform by image processing of the digital image;
the total station prism measuring unit obtains the position of each prism in the prism group in the CP III control network through measurement, thereby obtaining the relation between the camera coordinate system of the binocular vision camera and the CP III control network;
and according to the relation between the camera coordinate system and the CP III control network, unifying the three-dimensional point cloud contour data of the rail bearing platform to the CP III control network to realize the three-dimensional shape detection of the rail bearing platform.
The method for detecting the three-dimensional shape of the high-speed rail ballastless track bearing platform based on the digital image is also characterized by comprising the following steps of:
a movable detection rack is arranged, a binocular vision camera is fixedly arranged on the detection rack, and a prism group in a total station prism measurement unit is fixedly arranged; calibrating parameters of the binocular vision camera, determining a camera coordinate system of the binocular vision camera, and calibrating three-dimensional coordinates of each prism in the prism group in the camera coordinate system;
the detection method comprises the following steps:
step 1, placing the detection rack on a track plate of a section to be detected, so that a track bearing platform to be detected is positioned in the central area of a visual field of a binocular vision camera; erecting a total station at a set distance from the detection rack, and erecting a CP III control network by adopting a method of freely setting intersection of station mark corners;
step 2, shooting by the binocular vision camera to obtain a digital image containing natural textures of the surface of the rail bearing platform to be detected, numbering and storing; meanwhile, acquiring three-dimensional coordinates of each prism on the detection rack in a CP III control network by using a total station, numbering and storing;
step 3, based on an image matching algorithm, carrying out digital image processing on the digital image containing the natural textures of the surface of the rail bearing table to be detected, and calculating to obtain pixel coordinates of the surface characteristic points of the rail bearing table in the digital image; performing three-dimensional reconstruction according to camera parameters calibrated by the binocular vision camera aiming at the pixel coordinates, thereby obtaining three-dimensional point cloud contour data of the rail bearing platform under a camera coordinate system;
step 4, establishing a coordinate transformation relation between a camera coordinate system and a CP III control network according to the three-dimensional coordinates of each prism in the CP III control network and the three-dimensional coordinates of each prism in the camera coordinate system, and converting three-dimensional point cloud contour data of the rail bearing platform under the camera coordinate system into the CP III control network according to the coordinate transformation relation;
and 5, moving the detection rack and the total station along the track plate of the section to be detected, keeping the total station and each prism in the measurement rack at a set distance, and repeating the steps 2-4 to realize three-dimensional shape detection of all the rail bearing platforms in the section to be detected.
The method for detecting the three-dimensional shape of the high-speed rail ballastless track bearing platform based on the digital image is also characterized by comprising the following steps of:
the parameter calibration is carried out for the binocular vision camera, a camera coordinate system of the binocular vision camera is determined, and three-dimensional coordinates of each prism in the calibration prism group in the camera coordinate system are carried out in the following mode:
step 3.1, adjusting a first camera and a second camera in the binocular vision camera to enable the first camera and the second camera to be capable of clearly imaging a rail bearing platform to be detected at the same time; respectively obtaining calibration images of a calibration plate through image acquisition by using the first camera and the second camera, calibrating the first camera and the second camera by using the calibration images of the first camera and the second camera and the information of the calibration plate, calculating to obtain camera parameters of the first camera and the second camera, and establishing a camera coordinate system on the first camera;
the step of calibrating the board information includes the distance between the non-collinear 3 characteristic points on the calibrating board;
the camera parameters include: equivalent focal length, principal point coordinates and distortion parameters of the first camera and the second camera respectively; the positional relationship of the first camera to the second camera comprises an angular relationship between the optical axis of the first camera and the optical axis of the second camera, and a distance relationship between the optical center of the first camera and the optical center of the second camera;
step 3.2, erecting a total station at a set distance from the measuring rack, establishing a total station coordinate system, carrying out three-dimensional coordinate measurement on each prism on the measuring rack one by the total station, repeating the measurement for a plurality of times, and taking an average value as a prism three-dimensional coordinate of each prism in the total station coordinate system;
step 3.3, placing a circular marking point with a cross center in the area to be measured, and shooting the circular marking point by the first camera and the second camera to obtain the three-dimensional coordinates of the marking point of the circular marking point under a camera coordinate system; measuring the central cross of the circular marking point by using the total station and the prism centering rod to obtain the three-dimensional coordinate of the marking point of the circular marking point under the total station coordinate system;
step 3.4, randomly moving the circular marking points in the region to be detected, and repeating the step 3.3 to obtain three-dimensional coordinates of the circular marking points in the camera coordinate system and the total station coordinate system when different positions are obtained, so that a coordinate transformation relation between the camera coordinate system and the total station coordinate system is obtained through calculation; the region to be detected is a space region defined according to the surface contour of the rail bearing table;
and 3.5, calculating to obtain the three-dimensional coordinates of the prism group in the camera coordinate system according to the three-dimensional coordinates of the prisms in the total station coordinate system obtained in the step 3.2 and the transformation relation between the camera coordinate system and the total station coordinate system in the step 2.4, and completing calibration.
The method for detecting the three-dimensional shape of the high-speed rail ballastless track bearing platform based on the digital image is also characterized by comprising the following steps of: the CP III control network refers to a third-level foundation pile control network in a ballastless track passenger dedicated line railway engineering measurement plane control network; the CP III control network is a GPS control network and a Beidou control network, and is a three-dimensional rectangular coordinate system which is arranged on a fixed point and does not move along with the detection rack.
Compared with the prior art, the utility model has the beneficial effects that:
1. the method of the utility model shoots pictures of the rail bearing platform in the binocular camera based on binocular stereoscopic vision, uses a digital image correlation method to carry out correlation calculation by utilizing specific concrete textures on the surface of the rail bearing platform, and does not need a projector and a laser emitter to realize measurement of the three-dimensional surface profile of the rail bearing platform.
2. Compared with the method for measuring by adopting a laser scanner in the prior art, the method has the advantages of short measuring time, high measuring efficiency and capability of completing measurement by collecting pictures at one time.
3. The measuring precision of the utility model is kept below 0.1mm, and the requirements of actual working conditions are met.
4. The method of the utility model realizes the conversion of the relation between the camera coordinate system and the CP III control network by measuring the position of the prism in the camera coordinate system in advance, and the total station and the prism used are all measuring instruments mastered by construction units, so that independent training is not required for measuring personnel, and the method is easy to accurately implement.
Drawings
FIG. 1 is a schematic diagram of a detection method according to the present utility model;
FIG. 2 is a schematic diagram of a measuring rack in the detection method of the present utility model;
FIG. 3 is a schematic diagram of the calibration process of the measuring rack and the total station in the detection method of the present utility model;
FIGS. 4a and 4b are front and top views, respectively, of a rail bearing platform;
reference numerals in the drawings: 1 detection frame, 2 first prism, 3 second prism, 4 first camera, 5 second camera, 6 total powerstation, 7 track board, 8 hold rail platform, 9 prism centering rod, 10 circular mark points.
Detailed Description
Referring to fig. 1, 2 and 3, a detection system is provided for realizing three-dimensional shape detection of a high-speed railway unbiased rail bearing platform based on digital images, and comprises a binocular vision camera and a total station prism measurement unit; the binocular vision camera consists of two cameras, namely a first camera 4 and a second camera 5, which are fixedly arranged at the same height position of the detection frame 1 and are respectively arranged at the left side and the right side of the detection frame 1; in specific implementation, the binocular vision camera consists of two industrial cameras and a matched lens, wherein the model of the camera is sea-Kangwei vision MV-CH089-10UM, the resolution is 4096 multiplied by 2160 pixels, and the length-width ratio is 16:9, approaching the shape of a single rail bearing platform; the lens matched with the industrial camera is a sea-Kangwei MVL-KF2528M-12MP, a 25mm fixed focus lens, and the selection of the lens with the focal length of 25mm can realize that a rail bearing table shot at the distance of 1.5 meters occupies the main part of an image; the two cameras are fixedly installed on the cross beam of the detection frame, the height from the ground is 1.5 m, the distance between the two cameras is 410 mm, and the installation included angle is 16.3 degrees.
The total station prism measuring unit consists of a total station and a prism group fixedly arranged on the detection frame 1; the prism group comprises at least 4 prisms, and 5 prisms are arranged in the embodiment, wherein two prisms face one side of the rail bearing platform to be tested, and the other three prisms are arranged on one side of the rail bearing platform facing away from the rail bearing platform to be tested; the first prism 2 and the second prism 3 are respectively shown by reference numerals in fig. 2, and have other prisms at different positions; in fig. 1, binocular vision cameras are respectively arranged on two sides of the detection frame 1, and are used for simultaneously detecting two rail bearing tables positioned on two sides of the rail plate, so that detection efficiency is improved.
The detection method comprises the following steps: obtaining a digital image containing natural textures of the surface of the rail bearing platform 8 through shooting by a binocular vision camera, and reconstructing three-dimensional point cloud contour data of the rail bearing platform through image processing of the digital image; the total station prism measuring unit obtains the position of each prism in the prism group in the CP III control network through measurement, thereby obtaining the relation between the camera coordinate system of the binocular vision camera and the CP III control network; and according to the relation between the camera coordinate system and the CP III control network, unifying the three-dimensional point cloud contour data of the rail bearing table 8 to the CP III control network to realize the three-dimensional shape detection of the rail bearing table.
As shown in fig. 1, 2 and 3, a movable detection frame 1 is provided in the present embodiment, a binocular vision camera is fixedly provided on the detection frame 1, and a prism group in a prism measurement unit of a total station is fixedly provided; parameter calibration is carried out on the binocular vision camera, a camera coordinate system M of the binocular vision camera is determined, and three-dimensional coordinates of each prism in the prism group in the camera coordinate system M are calibrated; the detection method comprises the following steps:
step 1, placing a detection rack 1 on a track plate of a section to be detected, so that a track bearing platform to be detected is positioned in the central area of a visual field of a binocular vision camera; and erecting a total station at a position 10 meters away from the detection frame 1, and erecting a CP III control network by adopting a method of freely setting intersection of station standard corners.
Step 2, shooting by a binocular vision camera, obtaining a digital image containing natural textures of the surface of the rail bearing platform to be tested, numbering and storing; meanwhile, the total station is used for collecting and obtaining the three-dimensional coordinates of each prism on the detection frame 1 in the CP III control network, and the three-dimensional coordinates are numbered and stored asWherein: o is from 1 to 5.
Step 3, based on the image matching algorithm, carrying out digital image processing on the digital image containing the natural textures of the surface of the rail bearing table to be detected, and calculating to obtain the bearingPixel coordinates of the rail platform surface feature points in the digital image; for pixel coordinates, performing three-dimensional reconstruction according to camera parameters calibrated by a binocular vision camera, thereby obtaining three-dimensional point cloud contour data of the rail bearing platform under a camera coordinate system M as followsWherein: t is from 1 to 4000.
Step 4, according to the three-dimensional coordinates of each prism in the CP III control networkAnd the three-dimensional coordinates of the prisms in the camera coordinate system M>Establishing a coordinate transformation relation between a camera coordinate system M and a CP III control network, and enabling three-dimensional point cloud contour data of the rail bearing platform under the camera coordinate system M to be ∈>Switching to CP III control network is +.>
And 5, moving the detection frame 1 and the total station along the track plate 7 of the section to be detected, keeping the total station and each prism in the measurement frame at a set distance, and repeating the steps 2-4 to realize three-dimensional shape detection of all the rail bearing platforms 8 in the section to be detected.
In this embodiment, parameter calibration is performed for the binocular vision camera, a camera coordinate system M of the binocular vision camera is determined, and three-dimensional coordinates of each prism in the calibration prism set in the camera coordinate system M are performed as follows:
placing a single rail bearing table in a laboratory, moving the detection rack to the position above the rail bearing table, and adjusting the first camera 4 and the second camera 5 in the binocular vision camera, so that the first camera 4 and the second camera 5 can simultaneously and clearly image the rail bearing table to be detected. Considering that the illumination is sufficient in the outdoor environment, additional lamps are required to be arranged for light supplement in the laboratory calibration process. After the binocular vision camera is regulated to image clearly, the rail bearing table is removed, and the calibration work of the detection rack is started.
Step a1, placing a single rail bearing table in a laboratory, moving a detection rack to a position above the rail bearing table, and adjusting a first camera 4 and a second camera 5 in a binocular vision camera to enable the first camera 4 and the second camera 5 to be capable of clearly imaging the rail bearing table to be detected at the same time; and then removing the rail bearing platform, placing a calibration plate at the position of the rail bearing platform, respectively obtaining calibration images of the calibration plate through image acquisition by utilizing the first camera 4 and the second camera 5, randomly moving the calibration plate while keeping the calibration plate within the range of the area to be measured, carrying out image acquisition again, repeating the operation for at least 14 times, carrying out calibration calculation on the first camera 4 and the second camera 5 by utilizing the calibration images obtained by shooting the first camera 4 and the second camera 5 and the calibration plate information, obtaining camera parameters of the first camera 4 and the second camera 5, establishing a camera coordinate system M on the first camera 4, wherein the coordinate origin of the camera coordinate system M is at the optical center of the first camera 4, the xy plane coincides with the target surface of the first camera 4, and the z axis is perpendicular to the target surface of the first camera 4 and points to the area to be measured.
The calibration plate information comprises distances among 3 non-collinear characteristic points on the calibration plate;
the camera parameters include: equivalent focal length, principal point coordinates and distortion parameters of each of the first camera 4 and the second camera 5; the positional relationship of the first camera 4 to the second camera 5 includes an angular relationship between the optical axis of the first camera 4 and the optical axis of the second camera 5, a distance relationship between the optical center of the first camera 4 and the optical center of the second camera 5;
step a2, erecting the total station 6 at a set distance from the measuring frame 1 and adopting a free standing methodEstablishing a total station coordinate system N, carrying out three-dimensional coordinate measurement on each prism on the measuring frame 1 one by the total station 6, repeating the measurement for a plurality of times, and taking an average value as a prism three-dimensional coordinate of each prism in the total station coordinate system NWherein: o is from 1 to 5.
Step s3, the rail bearing table 8 is replaced in the area to be detected, a central cross circular marking point 10 is placed in the area to be detected, the circular marking point 10 is made of reflective materials and is stuck on a 2mm thick aluminum plate, the circular marking point 10 is shot by the first camera 4 and the second camera 5, the point location is carried out on the circular marking point in the shot images of the first camera 4 and the second camera 5 by using the findcircle function in opencv, and the three-dimensional space coordinate value of the circular marking plate is reconstructed and calculated by using the camera parameters obtained in the step a1, so that the three-dimensional coordinate of the circular marking point 10 under the camera coordinate system M is obtained;
the central cross of the circular marking point 10 is measured by the total station 6 and the prism centering rod 9, namely the prism centering rod 9 is propped against the central cross of the circular marking point, the position of the prism centering rod 9 is adjusted, so that the horizontal bubble on the prism centering rod is centered, the total station 6 is used for measuring the three-dimensional coordinates of the prism of the centering rod, and the three-dimensional coordinates of the circular marking point 10 under the total station coordinate system N are obtained.
Step a4, randomly moving the circular marking points 10 in the region to be detected, and repeating the step 3.3 to obtain three-dimensional coordinates of the circular marking points 10 in the camera coordinate system M when 20 groups of different positions are obtainedWherein: i is from 1 to 20, and the three-dimensional coordinates in the total station coordinate system N +.>Wherein: i is from 1 to 20, thereby calculating and obtaining a coordinate transformation relation R between the camera coordinate system M and the total station coordinate system N N-M And T N-M ；
The region to be measured is a space region defined according to the surface contour of the rail bearing table;
step a5, obtaining prism three-dimensional coordinates of each prism in the total station coordinate system N according to the step a2o from 1 to 5) and the transformation relationship R between the camera coordinate system M and the total station coordinate system N in step a4 N-M And T N -M Three-dimensional coordinates (+_in) of the prism group in the camera coordinate system M are calculated>o from 1 to 5):
and (5) finishing calibration.
The CP III control network refers to a third-level foundation pile control network in a ballastless track passenger dedicated line railway engineering measurement plane control network; the CP III control network is a GPS control network and a Beidou control network, and is a three-dimensional rectangular coordinate system which is arranged on a fixed point and does not move along with the detection rack.
The precision of the utility model can reach below 0.1mm, and the requirements of actual working conditions are met.
And (3) verifying measurement accuracy:
in an outdoor environment, the method is adopted for detecting the rail bearing platform A and the rail bearing platform B on a single rail plate, and simultaneously, a three-dimensional laser scanner is adopted for detecting, so that relevant size information is obtained, wherein the size information is shown in a table 1 and comprises the inclination angles alpha 1 and alpha 2 of the inner side of the rail bearing platform and the width L of the bottom edge of the rail bearing platform shown in fig. 4a and 4B.
TABLE 1
Compared with a three-dimensional laser scanner, the method has short measurement time (only one picture needs to be acquired) and high measurement efficiency; when the precision of the method is compared with that of the laser scanner, compared with the nominal precision of 0.02mm of the laser scanner, the precision of the method is slightly reduced, but the precision is still below 0.1mm, and the actual working condition requirement is met.
In addition, the method realizes the conversion of the relation between the camera coordinate system and the CP III control network by measuring the position of the prism in the camera coordinate system in advance, and the total station and the prism are measuring instruments which are already mastered by a construction unit, so that independent training of measuring staff is not needed.
Claims (4)
1. A three-dimensional shape detection method of a high-speed railway ballastless track bearing platform based on digital images is characterized by comprising the following steps:
a detection system is arranged and comprises a binocular vision camera and a total station prism measuring unit;
the binocular vision camera consists of two cameras, namely a first camera (4) and a second camera (5), which are fixedly arranged at the same height position of the detection frame (1) and are respectively arranged at the left side and the right side of the detection frame (1);
the total station prism measuring unit consists of a total station and a prism group fixedly arranged on the detection frame (1); the prism group comprises at least 4 prisms;
obtaining a digital image containing natural textures of the surface of the rail bearing table (8) through shooting by the binocular vision camera, and reconstructing three-dimensional point cloud contour data of the rail bearing table through image processing of the digital image;
the total station prism measurement unit obtains the position of each prism in the prism group in the CP III control network through measurement, thereby obtaining the relation between the camera coordinate system of the binocular vision camera and the CP III control network, which means that: based on three-dimensional coordinates of prisms in CP III control networkAnd the three-dimensional coordinates of the prisms in the camera coordinate system M>Establishing a coordinate transformation relation between a camera coordinate system M and a CP III control network, and enabling three-dimensional point cloud contour data of the rail bearing platform under the camera coordinate system M to be ∈>Switching to CP III control network is +.>
And according to the relation between the camera coordinate system and the CP III control network, unifying the three-dimensional point cloud contour data of the rail bearing platform (8) to the CP III control network to realize the three-dimensional shape detection of the rail bearing platform.
2. The three-dimensional shape detection method of the high-speed rail ballastless track bearing platform based on the digital image according to claim 1, which is characterized by comprising the following steps:
a movable detection frame (1) is arranged, a binocular vision camera is fixedly arranged on the detection frame (1), and a prism group in a total station prism measurement unit is fixedly arranged; calibrating parameters of the binocular vision camera, determining a camera coordinate system (M) of the binocular vision camera, and calibrating three-dimensional coordinates of each prism in the prism group in the camera coordinate system (M);
the detection method comprises the following steps:
step 1, placing the detection rack (1) on a track plate of a section to be detected, so that a track bearing platform to be detected is positioned in the central area of a visual field of a binocular vision camera; erecting a total station at a set distance from the detection frame (1), and erecting a CP III control network by adopting a method of freely setting intersection of station mark corners;
step 2, shooting by the binocular vision camera to obtain a digital image containing natural textures of the surface of the rail bearing platform to be detected, numbering and storing; meanwhile, acquiring three-dimensional coordinates of each prism on the detection frame (1) in a CP III control network by using a total station, numbering and storing;
step 3, based on an image matching algorithm, carrying out digital image processing on the digital image containing the natural textures of the surface of the rail bearing platform to be detected, and calculating to obtain pixel coordinates of the surface characteristic points of the rail bearing platform in the digital image; performing three-dimensional reconstruction according to camera parameters calibrated by the binocular vision camera aiming at the pixel coordinates, thereby obtaining three-dimensional point cloud contour data of the rail bearing platform under a camera coordinate system (M);
step 4, establishing a coordinate transformation relation between a camera coordinate system (M) and a CP III control network according to the three-dimensional coordinates of each prism in the CP III control network and the three-dimensional coordinates of each prism in the camera coordinate system (M), and converting three-dimensional point cloud contour data of the rail bearing platform under the camera coordinate system (M) into the CP III control network according to the coordinate transformation relation;
and 5, moving the detection frame (1) and the total station along the track plate of the section to be detected, keeping the total station and each prism in the measurement frame at a set distance, and repeating the steps 2-4 to realize three-dimensional shape detection of all the rail bearing platforms in the section to be detected.
3. The three-dimensional shape detection method of the high-speed rail ballastless track bearing platform based on the digital image according to claim 2, which is characterized by comprising the following steps:
the parameter calibration is carried out for the binocular vision camera, a camera coordinate system (M) of the binocular vision camera is determined, and three-dimensional coordinates of each prism in the calibration prism group in the camera coordinate system (M) are carried out in the following mode:
step 3.1, adjusting a first camera (4) and a second camera (5) in the binocular vision camera, so that the first camera (4) and the second camera (5) can simultaneously and clearly image a rail bearing platform to be detected; respectively obtaining calibration images of a calibration plate through image acquisition by using the first camera (4) and the second camera (5), calibrating the first camera (4) and the second camera (5) by using the calibration images of the first camera (4) and the second camera (5) and calibration plate information, calculating camera parameters of the first camera (4) and the second camera (5), and establishing a camera coordinate system (M) on the first camera (4);
the step of calibrating the board information includes the distance between the non-collinear 3 characteristic points on the calibrating board;
the camera parameters include: equivalent focal length, principal point coordinates and distortion parameters of the first camera (4) and the second camera (5) respectively; a positional relationship of the first camera (4) to the second camera (5), including an angular relationship between an optical axis of the first camera (4) and an optical axis of the second camera (5), a distance relationship between an optical center of the first camera (4) and an optical center of the second camera (5);
step 3.2, erecting a total station (6) at a set distance from the detection frame (1) and establishing a total station coordinate system (N), carrying out three-dimensional coordinate measurement on each prism on the detection frame (1) one by the total station (6), and taking an average value as a prism three-dimensional coordinate of each prism in the total station coordinate system (N) after repeated measurement for a plurality of times;
step 3.3, placing a circular marking point (10) with a cross center in the area to be measured, and shooting the circular marking point (10) by the first camera (4) and the second camera (5) to obtain a marking point three-dimensional coordinate of the circular marking point (10) under a camera coordinate system (M); measuring the central cross of the circular marking point (10) by the total station (6) and the prism centering rod (9) to obtain the three-dimensional coordinates of the marking point of the circular marking point (10) under the total station coordinate system (N);
step 3.4, randomly moving the circular marking point (10) in the region to be detected, and repeating the step 3.3 to obtain three-dimensional coordinates of the circular marking point (10) in the camera coordinate system (M) and the total station coordinate system (N) at different positions, so as to calculate and obtain a coordinate transformation relation between the camera coordinate system (M) and the total station coordinate system (N); the region to be detected is a space region defined according to the surface contour of the rail bearing table;
and 3.5, calculating to obtain the three-dimensional coordinates of the prism group in the camera coordinate system (M) according to the three-dimensional coordinates of the prisms in the total station coordinate system (N) obtained in the step 3.2 and the transformation relation between the camera coordinate system (M) and the total station coordinate system (N) in the step 2.4, and completing calibration.
4. The three-dimensional shape detection method of the high-speed rail ballastless track bearing platform based on the digital image according to claim 1, which is characterized by comprising the following steps: the CP III control network refers to a third-level foundation pile control network in a ballastless track passenger dedicated line railway engineering measurement plane control network; the CP III control network is a GPS control network and a Beidou control network, and is a three-dimensional rectangular coordinate system which is arranged on a fixed point and does not move along with the detection rack.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111073961.1A CN113776462B (en) | 2021-09-14 | 2021-09-14 | Three-dimensional shape detection method for high-speed rail ballastless track bearing platform based on digital image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111073961.1A CN113776462B (en) | 2021-09-14 | 2021-09-14 | Three-dimensional shape detection method for high-speed rail ballastless track bearing platform based on digital image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113776462A CN113776462A (en) | 2021-12-10 |
| CN113776462B true CN113776462B (en) | 2024-03-01 |
Family
ID=78843584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111073961.1A Active CN113776462B (en) | 2021-09-14 | 2021-09-14 | Three-dimensional shape detection method for high-speed rail ballastless track bearing platform based on digital image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113776462B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114719745B (en) * | 2022-03-04 | 2024-01-30 | 中铁第四勘察设计院集团有限公司 | Construction method and system of high-speed magnetic levitation track cross wire network |
| CN115265487B (en) | 2022-07-28 | 2023-03-31 | 宁波市特种设备检验研究院 | Platform clearance instrument based on photogrammetry principle and measurement method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003185589A (en) * | 2001-12-20 | 2003-07-03 | Nishimatsu Constr Co Ltd | Investigation system and investigation method for change in surface state of concrete |
| CN101962925A (en) * | 2010-08-12 | 2011-02-02 | 上海铁路局科学技术研究所 | Method for efficiently measuring three-dimensional coordinates of track based on track precise control net |
| CN102692185A (en) * | 2012-06-13 | 2012-09-26 | 东南大学 | Rapid detection method for ballastless track boards of high-speed railway |
| CN106092057A (en) * | 2016-07-28 | 2016-11-09 | 南昌航空大学 | A kind of helicopter rotor blade dynamic trajectory measuring method based on four item stereo visions |
| CN109253706A (en) * | 2018-08-24 | 2019-01-22 | 中国科学技术大学 | A kind of tunnel 3 D measuring method based on digital picture |
| CN111238454A (en) * | 2020-01-17 | 2020-06-05 | 中国人民解放军战略支援部队信息工程大学 | Space three-dimensional coordinate measuring instrument system and three-dimensional control network measuring method thereof |
| CN111809466A (en) * | 2020-09-14 | 2020-10-23 | 深圳大学 | Retest method and device for CPIII control network of high-speed railway |
-
2021
- 2021-09-14 CN CN202111073961.1A patent/CN113776462B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003185589A (en) * | 2001-12-20 | 2003-07-03 | Nishimatsu Constr Co Ltd | Investigation system and investigation method for change in surface state of concrete |
| CN101962925A (en) * | 2010-08-12 | 2011-02-02 | 上海铁路局科学技术研究所 | Method for efficiently measuring three-dimensional coordinates of track based on track precise control net |
| CN102692185A (en) * | 2012-06-13 | 2012-09-26 | 东南大学 | Rapid detection method for ballastless track boards of high-speed railway |
| CN106092057A (en) * | 2016-07-28 | 2016-11-09 | 南昌航空大学 | A kind of helicopter rotor blade dynamic trajectory measuring method based on four item stereo visions |
| CN109253706A (en) * | 2018-08-24 | 2019-01-22 | 中国科学技术大学 | A kind of tunnel 3 D measuring method based on digital picture |
| CN111238454A (en) * | 2020-01-17 | 2020-06-05 | 中国人民解放军战略支援部队信息工程大学 | Space three-dimensional coordinate measuring instrument system and three-dimensional control network measuring method thereof |
| CN111809466A (en) * | 2020-09-14 | 2020-10-23 | 深圳大学 | Retest method and device for CPIII control network of high-speed railway |
Non-Patent Citations (2)
| Title |
|---|
| 城市轨道精密三维测控新技术的研究与应用;韩三琪;;都市快轨交通(第01期);65-68 * |
| 基于CPⅣ轨道基准网轨道三维检测系统的研究;植立才;杨雪荣;成思源;杨世峰;;铁道标准设计(第04期);16-20 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113776462A (en) | 2021-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113776462B (en) | Three-dimensional shape detection method for high-speed rail ballastless track bearing platform based on digital image | |
| CN101526336B (en) | Calibration method of linear structured light three-dimensional visual sensor based on measuring blocks | |
| CN107121093A (en) | A kind of gear measurement device and measuring method based on active vision | |
| CN113029009B (en) | Double-visual-angle vision displacement measurement system and method | |
| CN105258710A (en) | High-precision camera principal point calibration method | |
| CN110375648A (en) | The spatial point three-dimensional coordinate measurement method that the single camera of gridiron pattern target auxiliary is realized | |
| CN106403900B (en) | Flying object tracking location system and method | |
| CN110966932B (en) | Structured light three-dimensional scanning method based on known mark points | |
| CN106289086B (en) | A kind of double camera measurement method for apart from Accurate Calibration between optical indicia point | |
| CN108805940B (en) | Method for tracking and positioning zoom camera in zooming process | |
| CN115682937A (en) | Calibration method of automatic three-dimensional laser scanner | |
| CN109541626B (en) | Target plane normal vector detection device and detection method | |
| CN103323216B (en) | A kind of device and method detecting parallax error of collimator | |
| CN112665523B (en) | Combined measurement method for complex profile | |
| CN113409406A (en) | Large-range monocular tracking scanning device | |
| CN110657750B (en) | Detection system and method for passivation of cutting edge of cutter | |
| CN108548506A (en) | A method of the measurement of planeness being carried out to high precision plane using optical markers | |
| CN106225773A (en) | A kind of photographic camera focal length and the measuring method of vanishing point | |
| CN114111578B (en) | Automatic pose determining method for large-caliber element | |
| CN109373901A (en) | Method for calculating center position of hole on plane | |
| Wu et al. | A novel precise guiding method for visual guiding theodolite measurement in volume space | |
| US11640680B2 (en) | Imaging system and a method of calibrating an image system | |
| CN210378040U (en) | Device for measuring amplification ratio of self-organized telescope by reticle method | |
| CN114663486A (en) | Building height measurement method and system based on binocular vision | |
| CN203053432U (en) | Digital level instrument |
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 |