CN112525158A - Double-shield six-degree-of-freedom measurement method and system based on monocular vision system - Google Patents

Double-shield six-degree-of-freedom measurement method and system based on monocular vision system Download PDF

Info

Publication number
CN112525158A
CN112525158A CN202011280179.2A CN202011280179A CN112525158A CN 112525158 A CN112525158 A CN 112525158A CN 202011280179 A CN202011280179 A CN 202011280179A CN 112525158 A CN112525158 A CN 112525158A
Authority
CN
China
Prior art keywords
shield
coordinate system
type
degree
coordinate
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202011280179.2A
Other languages
Chinese (zh)
Other versions
CN112525158B (en
Inventor
黄磊
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.)
Jiangsu Jicui Intelligent Photoelectric System Research Institute Co ltd
Original Assignee
Jiangsu Jicui Intelligent Photoelectric System Research Institute 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 Jiangsu Jicui Intelligent Photoelectric System Research Institute Co ltd filed Critical Jiangsu Jicui Intelligent Photoelectric System Research Institute Co ltd
Priority to CN202011280179.2A priority Critical patent/CN112525158B/en
Publication of CN112525158A publication Critical patent/CN112525158A/en
Application granted granted Critical
Publication of CN112525158B publication Critical patent/CN112525158B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C1/00—Measuring angles
    • 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
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
    • 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
    • G06T2207/30204—Marker

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Multimedia (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

The invention discloses a double-shield six-degree-of-freedom measurement method and system based on a monocular vision system, wherein the monocular vision system comprises an X-type characteristic identifier, a monocular camera, an illumination light source and an inclination angle sensor, and the method comprises the following steps: establishing a front shield coordinate system, a rear shield coordinate system and a measurement system coordinate system where a camera is located, and calibrating each coordinate system; acquiring image data of an X-type characteristic mark acquired by a camera, and determining a two-dimensional coordinate of the X-type characteristic mark in an image according to the image data; and performing PnP calculation according to the calibration coordinates of the plurality of X-type feature identifiers and the two-dimensional coordinates, determining the rotational translation relation of the anterior shield coordinate system relative to the measurement system coordinate system, determining the six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining the calibration results of the measurement system coordinate system and the posterior shield coordinate system, and correcting the six-degree-of-freedom information according to the inclination data. The invention carries the monocular camera and the tilt angle sensor to measure the pose by arranging the X-shaped optical characteristic mark, and has high measurement precision.

Description

Double-shield six-degree-of-freedom measurement method and system based on monocular vision system
Technical Field
The invention relates to the technical field of shield angle measurement, in particular to a double-shield six-degree-of-freedom measurement method and system based on a monocular vision system.
Background
The double-shield hard rock tunnel boring machine comprises a front shield and a rear shield, wherein the front shield is provided with a cutter head, a drive, a main bearing and the like, a support device is arranged on a rear shield body, the front shield and the rear shield are connected through a telescopic oil cylinder, the telescopic oil cylinder realizes the front-back stretching of the front shield relative to the rear shield through the front-back stretching, so that the movement of the front shield relative to the rear shield is realized, and the front-back distance variation range can reach 3.5m to 5 m. The double-shield hard rock tunnel boring machine can solve the problems of small-radius turning tunnels, high rock strength, complex rock properties and the like, has the characteristics of high construction speed, high efficiency and high quality, and becomes the primary choice in the boring construction of hard rock tunnels.
In the tunneling construction process, the tunnel boring machine adopting the shield method can only move forward and cannot move backward, so that the pose of the shield body needs to be accurately measured in real time, and the pose of the shield body needs to be adjusted in time by combining a set tunnel design curve, so that the deviation between an actual tunneling line of the tunnel boring machine and the set tunnel design curve is controlled.
At present, the attitude positioning for the double-shield hard rock tunnel boring machine comprises a gyroscope measuring system and a laser target measuring system, wherein the gyroscope measuring system measures the position attitude of a shield body through a gyroscope, the measuring accuracy of the gyroscope is greatly influenced by vibration and temperature in the construction process, measuring errors exist, regular calibration is needed, the stability is poor, and the measuring accuracy is low. The laser target measuring system uses a laser target to carry the total station, measures the position posture of the shield body through the light spot image, and is easy to be influenced by dust and environmental light due to the fact that the measuring site environment is severe, so that the condition that the light spot is shielded in the measuring process is caused, the data measuring accuracy is poor, and the measuring efficiency is low.
Disclosure of Invention
The invention provides a double-shield six-degree-of-freedom measuring method based on a monocular vision system, which realizes attitude positioning by using characteristic points with light reflection characteristics, has low requirement on environmental conditions, and solves the problems of low measuring precision and poor efficiency of the existing shield structure attitude positioning system of a heading machine.
In a first aspect, an embodiment of the present invention provides a double-shield six-degree-of-freedom measurement method based on a monocular vision system, which is used for measuring relative six-degree-of-freedom information between a front shield and a rear shield of a double-shield, where the monocular vision system includes a plurality of X-type feature identifiers with a light reflection characteristic, which are disposed on a rear end surface of the front shield, a monocular camera and an illumination light source, which are disposed on a front end surface of the rear shield, a first tilt sensor coaxially disposed with the front shield, and a second tilt sensor coaxially disposed with the rear shield, and the method includes the following steps: establishing a front shield coordinate system, a rear shield coordinate system and a measuring system coordinate system where the monocular camera is located, calibrating the front shield coordinate system, the rear shield coordinate system and the measuring system coordinate system, and storing a calibration result; starting measurement, acquiring image data of an X-type characteristic mark acquired by a monocular camera, and determining a two-dimensional coordinate of the X-type characteristic mark in an image according to the image data; performing PnP calculation according to a first three-dimensional calibration coordinate and a two-dimensional coordinate of a plurality of X-type feature identifiers on the rear end face of the anterior shield, acquiring a first rotational-translational relation of an anterior shield coordinate system relative to a measurement system coordinate system, and determining first six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining a second rotational-translational relation between the posterior shield coordinate system obtained through calibration and the measurement system coordinate system; acquiring first inclination angle data of the first inclination angle sensor and second inclination angle data of the second inclination angle sensor; and correcting the first six-degree-of-freedom information according to the first inclination angle data and the second inclination angle data, and determining the final real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield.
Optionally, the X-shaped feature marks include a light absorption sub mark and a light reflection sub mark, the light absorption sub mark is made of a light absorption material, the light reflection sub mark is made of a light reflection material, and the light absorption sub mark and the light reflection sub mark are alternately distributed at 90 degrees.
Optionally, the acquiring image data of the X-type feature identifier collected by the monocular camera includes the following steps: controlling the illumination light source to be lightened; controlling the monocular camera to continuously acquire image data of a plurality of X-type feature identifiers according to a preset sampling rule; and receiving the image data of the plurality of X-type characteristic identifications through a network port.
Optionally, the determining two-dimensional coordinates of the X-type feature in the image according to the image data includes: carrying out target initial positioning identification on the image data of the X-shaped feature identifier to obtain a plurality of initial positions of X-shaped corner point features; extracting sub-pixel central points of the X-type feature identification by adopting a Hessian matrix based on the initial position of the X-type corner feature; and acquiring the two-dimensional coordinates of the X-type feature identifier according to the feature extraction result.
Optionally, the performing target initial positioning recognition on the image data of the X-type feature identifier includes the following steps: carrying out gray processing on the image of the X-type feature identifier; performing threshold segmentation on the processed image according to the gray value to obtain an image segmentation result; adopting a connected domain marking algorithm to mark a connected domain of the image segmentation result; and determining the preliminary positions of the X-type characteristic marks according to the connected domain marking result.
Optionally, the establishing a anterior shield coordinate system, a posterior shield coordinate system, and a measurement system coordinate system where the monocular camera is located, and calibrating the anterior shield coordinate system, the posterior shield coordinate system, and the measurement system coordinate system includes the following steps: acquiring intrinsic parameters of a monocular camera, and establishing a camera matrix according to the intrinsic parameters of the monocular camera; measuring inherent characteristics of an anterior shield by using a total station, and establishing an anterior shield coordinate system according to a measurement result; acquiring first three-dimensional calibration coordinates of a plurality of X-type feature identifiers in the anterior shield coordinate system; measuring inherent characteristics of a rear shield by using a total station, and establishing a rear shield coordinate system according to a measurement result; acquiring a calibration image of the X-type characteristic mark by adopting the monocular camera, and acquiring two-dimensional calibration coordinates of a plurality of X-type characteristic marks based on the calibration image; measuring second three-dimensional calibration coordinates of the plurality of X-type feature identifiers in the rear shield coordinate system by using a total station; calibrating a second rotation and translation relation of the posterior shield coordinate system relative to the measurement system coordinate system according to the two-dimensional calibration coordinate and the second three-dimensional calibration coordinate; and storing the calibration result.
Optionally, the PnP calculation is performed according to the first three-dimensional calibration coordinates and the two-dimensional coordinates of the plurality of X-type feature identifiers on the rear end surface of the anterior shield, so as to obtain a first rotational-translational relationship of the anterior shield coordinate system relative to the measurement system coordinate system, and the first six-degree-of-freedom information of the anterior shield relative to the posterior shield is determined by combining a second rotational-translational relationship between the posterior shield coordinate system obtained by calibration and the measurement system coordinate system, which includes the following steps: acquiring a space geometric topological relation of the X-shaped characteristic mark of the rear end face of the anterior shield; matching the X-type characteristic identifier in the anterior shield coordinate system with the two-dimensional coordinate of the X-type characteristic identifier in the image based on the space geometric topological relation; determining a first rotation and translation relation of the anterior shield coordinate system relative to the measurement system coordinate system according to the three-dimensional coordinates of the X-type feature identifier in the anterior shield coordinate system and the two-dimensional coordinates of the X-type feature identifier in the corresponding image by adopting a PNP algorithm; obtaining a second rotation and translation relation of the calibrated trailing shield coordinate system relative to the measurement system coordinate system; and calculating first six-degree-of-freedom information of the anterior shield relative to the posterior shield through a coordinate system transformation algorithm.
Optionally, the plurality of X-type signatures with light reflection characteristics include a first X-type signature, a second X-type signature, a third X-type signature, a fourth X-type signature, a fifth X-type signature, and a sixth X-type signature, the first X-type signature and the sixth X-type signature are located on a first straight line, the second X-type signature, the third X-type signature, the fourth X-type signature, and the fifth X-type signature are located on a second straight line, the first straight line and the second straight line are parallel to the front shield rear end surface, and the first straight line intersects with the second straight line.
Optionally, the relative position relationship between the X-type feature marker and the anterior shield is kept unchanged, the relative position relationship between the monocular camera and the posterior shield is kept unchanged, the monocular camera has a first field angle, the illumination light source has a second field angle, the first field angle is smaller than or equal to the second field angle, and the motion section of the X-type feature marker falls into the first field angle.
In a second aspect, an embodiment of the present invention further provides a double-shield six-degree-of-freedom measurement system based on a monocular vision system, for measuring six degrees of freedom of a relative motion between an anterior shield and a posterior shield of a double-shield, including: the device comprises a plurality of X-shaped characteristic marks with light reflecting characteristics, a monocular camera, an illuminating light source and a control module, a first inclination angle sensor and a second inclination angle sensor, wherein the X-shaped characteristic marks are arranged on the rear end face of a front shield; the control module is used for controlling the lighting source to be turned on or off, controlling the monocular camera to acquire image data of the X-type characteristic identifier, determining a two-dimensional coordinate of the X-type characteristic identifier in an image according to the image data, performing PnP (pseudo noise) calculation according to a first three-dimensional calibration coordinate of a plurality of X-type characteristic identifiers on the rear end face of the anterior shield and the two-dimensional coordinate, acquiring a first rotational translation relation of the anterior shield coordinate system relative to a measurement system coordinate system, and determining first six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining a second rotational translation relation between the posterior shield coordinate system obtained by calibration and the measurement system coordinate system; the control module is further configured to acquire first inclination data of the first inclination sensor and second inclination data of the second inclination sensor, correct the first six-degree-of-freedom information according to the first inclination data and the second inclination data, and determine final real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield.
According to the double-shield six-degree-of-freedom measurement method and system based on the monocular vision system, the monocular vision system is arranged to measure the spatial position and the angular attitude of the double shields, the monocular vision system comprises a plurality of X-type characteristic marks with light reflection characteristics, which are arranged on the rear end face of a front shield, a monocular camera and an illumination light source, which are arranged on the front end face of a rear shield, a first inclination angle sensor coaxially arranged with the front shield, and a second inclination angle sensor coaxially arranged with the rear shield, before measurement, a front shield coordinate system, a rear shield coordinate system and a measurement system coordinate system are established and uniformly calibrated, and calibration results are stored; after the measurement is started, controlling a monocular camera to acquire an image of the X-type characteristic mark and acquiring a two-dimensional coordinate of the X-type characteristic mark; matching the feature identifier in the anterior shield coordinate system with the two-dimensional coordinates of the feature identifier, determining a first rotational-translational relationship between the anterior shield coordinate system and the measurement system coordinate system, determining first six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining a second rotational-translational relationship between the posterior shield coordinate system and the measurement system coordinate system obtained by the calibration result, and the first six-degree-of-freedom information is corrected according to the real-time inclination angle data acquired by the inclination angle sensor, the real-time six-degree-of-freedom of the anterior shield relative to the posterior shield is determined, the posture positioning is realized by utilizing the characteristic points with the light reflecting characteristic, the method has low requirement on environmental conditions, solves the problems of low measurement precision and poor efficiency of the existing shield structure attitude positioning system of the heading machine, has high measurement efficiency and measurement precision and high system integration level, and is favorable for realizing the light weight and low power consumption performance of the attitude positioning system.
Drawings
Fig. 1 is a flowchart of a double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a monocular vision system according to an embodiment of the present invention;
fig. 3 is a flowchart of another double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention;
fig. 4 is a flowchart of another double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention;
fig. 5 is a flowchart of another double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a double-shield six-degree-of-freedom measurement system based on a monocular vision system according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention, which is applicable to an application scenario of measuring relative six-degree-of-freedom information between a front shield and a rear shield of a double-shield of a heading machine, and the method can be executed by software and a processor configured with a double-shield six-degree-of-freedom measurement system.
Fig. 2 is a schematic structural diagram of a monocular vision system according to an embodiment of the present invention.
As shown in fig. 2, the monocular vision system includes a plurality of X-shaped feature markers with reflective characteristics, which are arranged on the rear end face of the front shield, a monocular camera and an illumination light source, which are arranged on the front end face of the rear shield, a first tilt sensor coaxially arranged with the front shield, and a second tilt sensor coaxially arranged with the rear shield, wherein the field angles of the monocular camera and the illumination light source can cover the feature motion section of the X-shaped feature markers, and the relative position relationship between the monocular camera and the rear shield is kept unchanged during the construction process of the heading machine.
The monocular camera is connected with the control module through a gigabit network port and a synchronous signal trigger line, on one hand, collected images are sent to the control module, and on the other hand, synchronous control signals sent by the control module are received; the illumination light source is connected with the control module through a control line and receives an illumination control signal sent by the control module, and the control module is also connected with a superior controller of the heading machine through a bus network to realize data transmission. Typically, the control module may include an embedded processor.
Specifically, before measurement starts, a plurality of X-type feature identifiers are pasted on the rear end face of the anterior shield according to a preset arrangement rule, the preset arrangement rule can be set according to the field angle of the monocular camera, the vertex of the preset arrangement rule falls into the field angle area of the monocular camera, and typically, the preset arrangement rule can be a cross structure or a cross-like structure.
Optionally, referring to fig. 2, the X-shaped feature marks include light absorbing sub marks and light reflecting sub marks, the light absorbing sub marks are made of light absorbing materials, the light reflecting sub marks are made of light reflecting materials, and the light absorbing sub marks and the light reflecting sub marks are alternately arranged at 90 degrees.
Specifically, black light absorption cloth can be used as a raw material of the light absorption sub identifier, bright silver chemical fiber light reflection cloth is used as a raw material of the light reflection sub identifier, the black light absorption cloth and the bright silver chemical fiber light reflection cloth are cut into fan-shaped structures with 90-degree vertex angles, the fan-shaped black light absorption cloth and the fan-shaped bright silver chemical fiber light reflection cloth are alternately distributed and arranged to manufacture X-shaped corner point characteristics to serve as the X-shaped characteristic identifier, the light absorption sub identifier of the X-shaped characteristic identifier absorbs light, and the light reflection sub identifier reflects light, so that the gray value of an image formed in the X-shaped characteristic identifier area is obviously different from the non-characteristic area.
Referring to fig. 1 in combination, the double-shield six-degree-of-freedom measurement method based on the monocular vision system includes the following steps:
step S1: establishing a front shield coordinate system, a rear shield coordinate system and a measuring system coordinate system where the monocular camera is located, calibrating the front shield coordinate system, the rear shield coordinate system and the measuring system coordinate system, and storing a calibration result.
Wherein, defining the anterior shield coordinate system as OBXBYBZBThe posterior shield coordinate system is OSXSYSZSCoordinate system of measurement system is OCXCYCZCThe coordinate system of the measuring system is established on the monocular camera.
Specifically, after all modules of the monocular vision system are fixedly installed, a front shield coordinate system O is established through the total stationBXBYBZBAnd the posterior shield coordinate system OSXSYSZSBefore the measurement is started, an anterior shield coordinate system O is obtained through calibrationBXBYBZBMeasuring system coordinate system OCXCYCZCAnd the posterior shield coordinate system OSXSYSZSA rotational-translational relationship therebetween, wherein,the rotational-translational relationship may be represented by a rotation matrix R and a translation vector T. Furthermore, the rotation and translation relationship obtained through calibration can be imported into the control module, and the control module performs coordinate transformation in the measuring process according to the rotation and translation relationship obtained through calibration.
Step S2: and starting measurement, acquiring image data of the X-type characteristic identifier acquired by the monocular camera, and determining the two-dimensional coordinate of the X-type characteristic identifier in the image according to the image data.
In this embodiment, an ARM (Advanced RISC Machines, microprocessors) + a GPU (Graphics Processing Unit, Graphics processor) may be used as a core control module to perform image Processing, so as to achieve miniaturization and low power consumption of the processor, and connect a monocular camera through a gigabit network port to obtain image data; the control signal is connected to a trigger signal receiving end of the camera to control the camera to work; the control switch is connected with the lighting source control switch through a path of control signal to control the lighting source to be turned on or turned off; and the data transmission is realized by connecting the RS422 bus to the upper-level controller.
Specifically, during the construction process of the heading machine, measurement is started, the control module controls the illumination light source to be lightened, the illumination light source irradiates a plurality of X-shaped characteristic marks arranged on the rear end face of the front shield, the monocular camera is controlled to be opened through the synchronous control signal, the images of the rear end face of the anterior shield in the construction process are continuously collected according to the preset time interval, and sends the collected image to the processor through the gigabit network port, the processor processes the image, under the irradiation of an illumination light source, the reflecting sub-mark in the X-shaped characteristic mark displays high brightness in the image, and the surrounding light-absorbing sub-markers are relatively dark in the image, the general position of the X-shaped characteristic marker in the image can be determined, and further extracting a characteristic region corresponding to the X-type characteristic identifier through a characteristic extraction algorithm, and acquiring a two-dimensional pixel coordinate corresponding to the central point of the X-type characteristic identifier under a measurement system coordinate system.
Step S3: PnP calculation is carried out according to first three-dimensional calibration coordinates of a plurality of X-type feature identifiers on the rear end face of the anterior shield and two-dimensional coordinates in the image, a first rotational-translational relation of the anterior shield coordinate system relative to a measurement system coordinate system is obtained, and first six-degree-of-freedom information of the anterior shield relative to the posterior shield is determined by combining a second rotational-translational relation between the posterior shield coordinate system and the measurement system coordinate system obtained through calibration.
Specifically, in the process of camera perspective projection transformation, based on the principle that the space geometric topological constraint relation of the feature identifier is kept unchanged under an anterior shield coordinate system, a posterior shield coordinate system and a measurement system coordinate system, matching of pixel points of an X-type feature identifier in the anterior shield coordinate system and pixel points of the X-type feature identifier in the image is achieved by analyzing a space geometric topological structure, and two-dimensional coordinates of feature points of the X-type feature identifier in the image are obtained; solving a anterior shield coordinate system O by adopting a PnP (Passive-n-Poin) algorithm according to a first three-dimensional calibration coordinate of the X-type feature identifier in a plurality of pairs of anterior shield coordinate systems and a two-dimensional coordinate of the X-type feature identifier feature point in the monocular camera imageBXBYBZBRelative to the measurement system coordinate system OCXCYCZCIn a first roto-translational relationship. Further, a postshield coordinate system O is obtained through a calibration result stored in the processorSXSYSZSAnd the coordinate system O of the measuring systemCXCYCZCThe second rotation translation relation between the two is obtained by a coordinate system transformation algorithm to obtain an anterior shield coordinate system OBXBYBZBRelative posterior shield coordinate system OCXCYCZCThe first six-degree-of-freedom information of the anterior shield is obtained through the position and posture transformation relation of the anterior shield.
Step S4: and acquiring first inclination angle data of the first inclination angle sensor and second inclination angle data of the second inclination angle sensor.
In this embodiment, a coordinate axis of the first tilt sensor is coaxially arranged with a corresponding coordinate axis in a coordinate system of the anterior shield, and first tilt data acquired by the first tilt sensor along with the movement of the posterior shield is real-time roll axis data of the anterior shield; and the coordinate axis of the second tilt sensor is coaxial with the corresponding coordinate axis in the rear shield coordinate system, and the second tilt data acquired by the second tilt sensor along with the movement of the rear shield is the real-time rolling axis data of the rear shield.
Step S5: and correcting the first six-degree-of-freedom information according to the first inclination angle data and the second inclination angle data, and determining the final real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield.
Specifically, the first six-degree-of-freedom information comprises three position parameters and three inclination angle parameters, a real-time tilting angle of the anterior shield relative to the posterior shield can be calculated according to the first inclination angle data and the second inclination angle data, and the real-time tilting angle is adopted to replace the corresponding inclination angle parameter in the first six-degree-of-freedom information, so that the final real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield is obtained.
In this embodiment, the anterior shield motion can be controlled according to the real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield.
In this embodiment, the real-time six-degree-of-freedom information of the front shield is sent to the upper-level controller of the heading machine through a bus data transmission technology, so that the upper-level controller adjusts the motion attitude of the front shield according to the real-time six-degree-of-freedom information, and it is ensured that the deviation between the actual heading line of the heading machine and the established tunnel design curve does not exceed the threshold.
Therefore, according to the double-shield six-degree-of-freedom measurement method based on the monocular vision system provided by the embodiment of the invention, the monocular vision system is arranged to measure the spatial position and the angular posture between the double shields, the monocular vision system comprises a plurality of X-shaped characteristic marks with light reflection characteristics, which are arranged on the rear end face of a front shield, a monocular camera and an illumination light source, which are arranged on the front end face of a rear shield, a first inclination angle sensor coaxially arranged with the front shield, and a second inclination angle sensor coaxially arranged with the rear shield, before measurement, a front shield coordinate system, a rear shield coordinate system and a measurement system coordinate system are established and uniformly calibrated, and calibration results are stored; after the measurement is started, controlling a monocular camera to acquire an image of the X-type characteristic mark and acquiring a two-dimensional coordinate of the X-type characteristic mark; matching the feature identifier in the anterior shield coordinate system with the two-dimensional coordinates of the feature identifier, determining a first rotational-translational relationship between the anterior shield coordinate system and the measurement system coordinate system, determining first six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining a second rotational-translational relationship between the posterior shield coordinate system and the measurement system coordinate system obtained by the calibration result, and the first six-degree-of-freedom information is corrected according to the real-time inclination angle data acquired by the inclination angle sensor, the real-time six-degree-of-freedom of the anterior shield relative to the posterior shield is determined, the posture positioning is realized by utilizing the characteristic points with the light reflecting characteristic, the method has low requirement on environmental conditions, solves the problems of low measurement precision and poor efficiency of the existing shield structure attitude positioning system of the heading machine, has high measurement efficiency and measurement precision and high system integration level, and is favorable for realizing the light weight and low power consumption performance of the attitude positioning system.
Optionally, the acquiring image data of the X-type feature identifier collected by the monocular camera includes the following steps: controlling the illumination light source to be lightened; controlling a monocular camera to continuously acquire image data of a plurality of X-type feature identifiers according to a preset sampling rule; image data of a plurality of X-type signatures is received through a network port.
In this embodiment, the preset sampling rule refers to that the monocular camera is controlled to perform image acquisition once every preset interval time in the sampling period.
Specifically, in the construction process of the heading machine, the control module directly controls the illumination light source to be normally on, and simultaneously transmits a synchronous control signal to the monocular camera according to a preset sampling rule, the monocular camera is controlled to collect images of the feature identification area in the rear end face of the front shield, the monocular camera transmits the images to the processor through the gigabit network port, and the processor receives a plurality of image data containing the X-type feature identification and performs image processing on the received image data.
Fig. 3 is a flowchart of another double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention.
Optionally, as shown in fig. 3, determining two-dimensional coordinates of the X-type feature in the coordinate system of the measurement system according to the image data includes the following steps:
step S201: and carrying out target initial positioning identification on the image data of the X-type feature identifier to obtain the initial positions of the X-type corner point features.
In this embodiment, under the irradiation of the illumination light source, the brightness of the reflection sub-mark in the X-shaped feature mark is greater than the brightness of the absorption sub-mark, and the preliminary positions of the multiple X-shaped corner point features in the image can be obtained through adaptive binarization processing and morphological operations.
Optionally, the method for performing initial target location recognition on the image data of the X-type feature identifier includes the following steps: carrying out gray processing on the image of the X-type characteristic mark; performing threshold segmentation on the processed image according to the gray value to obtain an image segmentation result; adopting a connected domain marking algorithm to mark a connected domain of the image segmentation result; and determining the preliminary positions of the X-type characteristic marks according to the connected domain marking result.
Specifically, a processor is adopted to carry out graying processing on an image of the X-type characteristic identifier to obtain a grayed image, and the grayscale image is subjected to threshold segmentation according to a preset segmentation threshold and a formula to obtain a binary image; and then, adopting a connected domain marking algorithm to mark a connected domain in the binary image, and determining the initial positions of the X-shaped corner point features.
Step S202: and extracting the sub-pixel central point of the X-type feature identification by adopting a Hessian matrix based on the initial position of the X-type corner feature.
Specifically, a hessian matrix algorithm is adopted to perform sub-pixel extraction on the image after the preliminary screening of the X-type angular point features, and since the X-type angular point features are almost not present in the rear end face of the anterior shield and the environment background except the X-type feature identifier, the hessian matrix algorithm is adopted to perform sub-pixel extraction on the preliminary position of the X-type angular point features determined in the step S201, and the sub-pixel central point of the X-type angular point features in the image is obtained. Typically, the image processing process in this embodiment may be accelerated by the GPU, so as to achieve the effect of fast image processing.
Step S203: and acquiring the two-dimensional coordinates of the central point of the X-type characteristic mark in the coordinate system of the measuring system according to the characteristic extraction result.
Specifically, the image collected by the single-camera may be cut into a plurality of sub-pixel units with the pixel size of m × n according to a specific step length, and coordinate labels of the sub-pixels in the original image are recorded, where the coordinate of the corner point of the X-type feature identifier is defined as (X)0，y0) Angular point (x)0，y0) Hessian matrix ofThe formula I is as follows:
Figure BDA0002780512740000141
wherein r isxxRepresenting the angular point (x) of the image captured by the binocular camera0，y0) At a second order gradient in the x-direction, rxyRepresenting the angular point (x) of the image captured by the binocular camera0，y0) First order gradient in y direction of first order gradient in x direction, ryyRepresenting the angular point (x) of the image captured by the binocular camera0，y0) A second order gradient in the y-direction.
Along the corner point (x) as shown in equation one0，y0) The second order gradient in the normal direction of (a) is the maximum absolute eigenvalue of the hessian matrix of that point, the corner point (x)0，y0) Normal direction (nx, ny) of the point.
In this embodiment, a corner point (x) is defined0，y0) The first-order zero-crossing point of the reference point has a sub-pixel coordinate of (x)0+s，y0+ t), if (s, t) ∈ [ -0.5, 0.5]*[-0.5，0.5]That is, the first-order zero crossing point of the edge is located in the current pixel, and the second derivative of the second-order gradient value and the normal direction is greater than the preset threshold, then the point is the edge center point, and the diagonal point (x)0，y0) And performing second-order Taylor expansion on the gray value of the inner sub pixel point to obtain a formula II shown as follows:
Figure BDA0002780512740000142
wherein r isxRepresenting the image captured by a monocular camera at a corner (x)0，y0) At a first gradient in the x direction, ryRepresenting the image captured by a monocular camera at a corner (x)0，y0) With a first order gradient in the y-direction.
By calculation, the following three and four formulas are shown:
Figure BDA0002780512740000143
Figure BDA0002780512740000144
and combining the formula, and obtaining the two-dimensional coordinates of the central point of the X-type characteristic mark by a sub-pixel edge extraction method.
Fig. 4 is a flowchart of another double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention.
Optionally, as shown in fig. 4, the establishing of the anterior shield coordinate system, the posterior shield coordinate system, and the measurement system coordinate system where the monocular camera is located, and the calibrating of the anterior shield coordinate system, the posterior shield coordinate system, and the measurement system coordinate system include the following steps:
step S101: and acquiring intrinsic parameters of the monocular camera, and establishing a camera matrix according to the intrinsic parameters of the monocular camera.
Step S102: measuring inherent characteristics of the anterior shield by using a total station, and establishing an anterior shield coordinate system according to a measurement result;
step S103: and acquiring first three-dimensional calibration coordinates of a plurality of X-type feature identifiers in a anterior shield coordinate system.
Step S104: and measuring inherent characteristics of the rear shield by using a total station, and establishing a rear shield coordinate system according to a measurement result.
Step S105: and acquiring a calibration image of the X-type characteristic mark by adopting a monocular camera, and acquiring two-dimensional calibration coordinates of a plurality of X-type characteristic marks based on the calibration image.
Step S106: and measuring a second three-dimensional calibration coordinate of a plurality of X-type characteristic marks in a rear shield coordinate system by using a total station.
Step S107: and calibrating a second rotation and translation relation of the shield coordinate system relative to the measurement system coordinate system according to the two-dimensional calibration coordinate and the second three-dimensional calibration coordinate.
Step S108: and storing the calibration result.
Specifically, before the monocular camera is mounted on the anterior face of the posterior shield, it is necessary toThe intrinsic parameters of the monocular camera are well defined under a standard laboratory environment, an intrinsic parameter matrix of the monocular camera is established according to the intrinsic parameters of the monocular camera, and the three-dimensional coordinates of the two-dimensional coordinates in the image acquired by the camera under a coordinate system of a measuring system can be calculated according to the intrinsic parameter matrix of the monocular camera. Measuring inherent characteristics of the anterior shield by using a total station, and establishing an anterior shield coordinate system O by coordinate value fitting of the inherent characteristics of the anterior shieldBXBYBZBThe anterior shield coordinate system OBXBYBZBAs a built-in measurement coordinate system of the total station, measuring each X-type characteristic mark in an anterior shield coordinate system OBXBYBZBAnd marking each coordinate according to a preset space geometric topological structure.
Further, measuring inherent characteristics of the posterior shield by using a total station, and establishing a posterior shield coordinate system O by fitting coordinate values of the inherent characteristics of the posterior shieldSXSYSZSThe postshield coordinate system OSXSYSZSAs a built-in measurement coordinate system of the total station, measuring each X-type characteristic mark in a posterior shield coordinate system OSXSYSZSAnd then, shooting an image of the X-type characteristic identifier in the calibration process by using a fixed monocular camera, acquiring two-dimensional calibration coordinates of a plurality of X-type characteristic identifiers based on the image in the calibration process, and adopting a PnP algorithm according to the principle that the spatial geometrical topological structure is kept unchanged in a posterior shield coordinate system O by using a PnP algorithmSXSYSZSThe second three-dimensional calibration coordinate and the corresponding two-dimensional calibration coordinate in the image, and a calibrated shield coordinate system OSXSYSZSRelative to the measurement system coordinate system OCXCYCZCAnd the second rotation translation relation is used for introducing coordinate transformation relation parameters of each coordinate system in the calibration parameters into the embedded processor to be used as reference parameters for subsequent measurement and calculation.
Fig. 5 is a flowchart of another double-shield six-degree-of-freedom measurement method based on a monocular vision system according to an embodiment of the present invention.
Alternatively, as shown in fig. 5, the step S3 includes the following steps:
step S301: and acquiring the space geometric topological relation of the X-shaped characteristic mark of the rear end face of the anterior shield.
Optionally, referring to fig. 2 in combination, according to a change of a field of view of the monocular camera, 6X-type feature identifiers are disposed on a rear end surface of the front shield, and the plurality of X-type feature identifiers having a light reflection characteristic include a first X-type feature identifier, a second X-type feature identifier, a third X-type feature identifier, a fourth X-type feature identifier, a fifth X-type feature identifier, and a sixth X-type feature identifier, where the first X-type feature identifier and the sixth X-type feature identifier are located on a first straight line k, the second X-type feature identifier, the third X-type feature identifier, the fourth X-type feature identifier, and the fifth X-type feature identifier are located on a second straight line j, and the first straight line k and the second straight line j are both parallel to the rear end surface of the front shield, and the first straight line k intersects with the second straight line j.
Specifically, defining a first X-type feature identifier as P1, a second X-type feature identifier as P2, a third X-type feature identifier as P3, a fourth X-type feature identifier as P4, a fifth X-type feature identifier as P5, and a sixth X-type feature identifier as P6, the identifier number can be determined by the distance between the X-type feature identifiers.
Step S302: and matching the X-type characteristic identifier in the anterior shield coordinate system with the two-dimensional coordinates of the X-type characteristic identifier in the image based on the space geometric topological relation.
In the projection transformation process of the monocular camera, the characteristics that the collinear relation is kept unchanged and the hour order of the coplanar points is kept unchanged are provided. Based on the above spatial characteristics, taking six X-type signatures shown in fig. 2 as an example, the following description is made on the matching process between pixel points in an image and the X-type signatures:
firstly, the processor obtains two-dimensional coordinates of the central point of each X-type characteristic mark through the steps S201 to S203, and performs straight line fitting on the extracted two-dimensional coordinates by using a least square method, wherein straight lines obtained through fitting are second straight lines j where second X-type characteristic marks P2 to fifth X-type characteristic marks P5 are located; and the processor stores a preset precision threshold epsilon, calculates the distance D from each two-dimensional coordinate to the second straight line j, and if the distance D is smaller than the preset precision threshold epsilon, the processor judges that the feature identifier corresponding to the two-dimensional coordinate is one of a second X-type feature identifier P2 to a fifth X-type feature identifier P5, and the feature identifiers corresponding to the remaining two-dimensional coordinates are a first X-type feature identifier P1 and a sixth X-type feature identifier P6.
Further, the processor performs straight line fitting on the two-dimensional coordinates corresponding to the first X-type feature identifier P1 and the sixth X-type feature identifier P6 by using a least square method, the obtained straight line is the first straight line k, coordinate values corresponding to the intersection point of the first straight line k and the second straight line j are obtained, the distance between the two-dimensional coordinates corresponding to each feature point and the coordinate values corresponding to the intersection point is respectively calculated, the feature point with the minimum distance from the intersection point is the third X-type feature identifier P2, the feature point with the maximum distance from the intersection point is the fifth X-type feature identifier P5, the feature point with the minimum distance from the fifth X-type feature identifier P5 is the fourth X-type feature identifier P4, the feature point with the farthest distance from the fifth X-type feature identifier P5 is the second X-type feature identifier P2, the feature point above the second straight line j is the first X-type feature identifier P1, and the feature point below the second straight line j is the sixth X-type feature identifier P6, thus, the processor completes the matching of all the X-type characteristic marks and the two-dimensional coordinates of the image.
Step S303: and determining a first rotation and translation relation of the anterior shield coordinate system relative to the measurement system coordinate system by adopting a PNP algorithm according to the three-dimensional coordinates of the X-type characteristic mark in the anterior shield coordinate system and the two-dimensional coordinates of the X-type characteristic mark in the corresponding image.
Step S304: and acquiring a second rotation and translation relation of the rear shield coordinate system obtained by calibration relative to the measurement system coordinate system.
Specifically, the three-dimensional coordinate of the i X th type characteristic mark Pi in the anterior shield coordinate is defined as [ X, Y, Z, 1 ]]TThe homogeneous coordinate thereof in the image coordinate system in units of pixels is [ u, v, 1 ]]TThe monocular camera perspective projection model is as follows:
Figure BDA0002780512740000181
Figure BDA0002780512740000182
wherein A represents a camera intrinsic parameter matrix obtained by calibrating a monocular camera, lambda represents a scale factor, R represents a rotation matrix, and T represents a translation vector.
In this embodiment, the formula for defining the rotation matrix R and the translation vector T is as follows:
Figure BDA0002780512740000191
Figure BDA0002780512740000192
after the monocular camera is calibrated, the parameters in the monocular camera are known, and the coordinates of the two-dimensional image point in the coordinate system of the monocular camera measuring system can be calculated. The rotation matrix R and the translation vector T between the measurement system coordinate system and the world coordinate system can be calculated from the three-dimensional coordinates of the feature identifier in the measurement system coordinate system and the three-dimensional coordinates of the feature identifier in the world coordinate system. The rotation matrix R is a 3 x 3 unit orthogonal matrix with three degrees of freedom, so the transformation relationship between two coordinate systems has six degrees of freedom. And substituting the world coordinates of the feature identifier in the anterior shield coordinate system into a fifth formula, and sorting to obtain a linear equation shown as the following.
Figure BDA0002780512740000193
The constraint of two degrees of freedom can be provided due to a pair of matched three-dimensional feature points and two-dimensional pixel point pairs. And calculating a rotation matrix R and a translation vector T between a coordinate system of the measurement system and a world coordinate system through a calibrated parameter matrix A in the camera, coordinates of more than 3 non-collinear feature points in the world coordinate system and corresponding image coordinates.
In this embodiment, six X-type signatures can be set, and the least square solution of the rotation matrix R and the translation vector T is solved by the least square method, so as to obtain the anterior shield coordinate system OBXBYBZBRelative to the measurement system coordinate system OCXCYCZCRelative pose of.
Step S305: and calculating first six-degree-of-freedom information of the anterior shield relative to the posterior shield through a coordinate system transformation algorithm.
The coordinate system transformation refers to a process of transforming from one space coordinate system to another space coordinate system.
Specifically, coordinate system transformation can be achieved by establishing a one-to-one correspondence between two coordinate systems, and the transformation formula is as follows:
Figure BDA0002780512740000201
where R denotes a rotation matrix transformed from coordinate system 2 to coordinate system 1, T denotes a translation vector transformed from coordinate system 2 to coordinate system 1, and O denotes a row vector in which elements of one row and three columns are all 0.
In this embodiment, the anterior shield coordinate system O needs to be first obtainedBXBYBZBTransformation to the measurement System coordinate System OCXCYCZCAfter transformation to the posterior shield coordinate system OSXSYSZSDefining the three-dimensional coordinate of the X-type characteristic mark under the anterior shield coordinate as [ X ]3，y3，z3，1]TIts three-dimensional coordinate in the coordinate system of the measuring system is [ x ]2，y2，z2，1]TIts three-dimensional coordinate under the trailing shield coordinate is [ x ]1，y1，z1，1]TThen, the specific calculation formula of the coordinate transformation is as follows:
Figure BDA0002780512740000202
wherein R is3＝R2R1，T3＝R2T1+T2，R1Representing a rotation matrix, T, of the anterior shield coordinate system relative to the measurement system coordinate system1Representing the translation vector, R, of the anterior shield coordinate system relative to the measurement system coordinate system2Representing the rotation matrix, T, of the trailing shield coordinate system relative to the measurement system coordinate system2And expressing a translation vector of the rear shield coordinate system relative to the measurement system coordinate system, so that a six-degree-of-freedom transformation relation of the front shield coordinate system relative to the rear shield coordinate system can be obtained, namely first six-degree-of-freedom information of the front shield relative to the rear shield, and correcting an inclination angle parameter in the first six-degree-of-freedom information by adopting an inclination angle obtained by the inclination angle sensor to obtain final real-time six-degree-of-freedom information.
Further, the real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield can be transmitted to the upper-level controller by adopting an RS-422 bus, so that motion feedback is realized, and the action of the anterior shield is controlled according to the fed-back real-time six-degree-of-freedom information.
Optionally, when the spatial position and the angular posture of the anterior shield change, the relative position relationship between the X-type feature marker and the anterior shield remains unchanged, the relative position relationship between the monocular camera and the posterior shield remains unchanged, the monocular camera has a first angle of view, the illumination light source has a second angle of view, the first angle of view is smaller than or equal to the second angle of view, and the motion section of the X-type feature marker falls into the first angle of view.
In this embodiment, the field angle of the monocular camera is set to be smaller than the field angle of the illumination light source, and the motion section of the X-type feature identifier falls into the field angle of the monocular camera, so that effective acquisition of the monocular camera can be ensured, and the measurement accuracy is improved.
Example two
The embodiment of the invention provides a double-shield six-degree-of-freedom measuring system based on a monocular vision system. Fig. 6 is a schematic structural diagram of a double-shield six-degree-of-freedom measurement system based on a monocular vision system according to an embodiment of the present invention. The embodiment is suitable for an application scene of measuring the relative six-degree-of-freedom information between the anterior shield and the posterior shield of the double shields.
As shown in fig. 6, the double-shield six-degree-of-freedom measurement system 100 based on the monocular vision system includes: the system comprises a plurality of X-shaped characteristic marks 10 with light reflecting characteristics, a monocular camera 20, an illuminating light source 30 and a control module 40, a first inclination angle sensor 50, a second inclination angle sensor 60, a monocular camera 20 and an illuminating light source 30, wherein the X-shaped characteristic marks 10 are arranged on the rear end face of a front shield, the monocular camera 20, the illuminating light source 30 and the control module 40 are arranged on the front end face of a rear shield, the first inclination angle sensor 50 and the rear shield are coaxially arranged, and the monocular camera 20 and the illuminating light source 30 are; the control module 40 is configured to control the illumination light source 30 to be turned on or off, control the monocular camera 20 to acquire image data of the X-type feature identifier 10, determine a two-dimensional coordinate of the X-type feature identifier 10 in a measurement system coordinate system according to the image data, perform PnP calculation according to a first three-dimensional calibration coordinate and a two-dimensional coordinate of a plurality of X-type feature identifiers 10 on a rear end surface of the anterior shield, acquire a first rotational translation relationship of the anterior shield coordinate system with respect to the measurement system coordinate system, determine first six-degree-of-freedom information of the anterior shield by combining a second rotational translation relationship between the posterior shield coordinate system and the measurement system coordinate system, which is obtained by calibration, acquire first tilt angle data of the first tilt angle sensor and second tilt angle data of the second tilt angle sensor, and correct the first six-degree-of-freedom information according to the first tilt angle data and the second tilt angle data, and determining the final real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield, and performing motion control on the anterior shield according to the real-time six-degree-of-freedom information.
Optionally, the X-shaped feature identifier 10 includes a light absorption sub identifier and a light reflection sub identifier, the light absorption sub identifier is made of a light absorption material, the light reflection sub identifier is made of a light reflection material, and the light absorption sub identifier and the light reflection sub identifier are alternately distributed at 90 degrees.
Optionally, the monocular camera 20 is connected to the control module 40 through a gigabit network port and a synchronization signal trigger line, the illumination light source 30 is connected to the control module 40 through a control line, the control module 40 is configured to control the illumination light source to be turned on, and simultaneously control the monocular camera 20 to continuously acquire image data of the plurality of X-type feature identifiers 10 according to a preset sampling rule, and the control module 40 further receives the image data of the plurality of X-type feature identifiers through the gigabit network port.
Optionally, the control module 40 includes an initial positioning unit and a calculating unit, where the initial positioning unit is configured to perform initial target positioning identification on the image data of the X-type feature identifier to obtain initial positions of the plurality of X-type corner point features; and the calculation unit is used for extracting the sub-pixel central point of the X-type characteristic mark by adopting a Hessian matrix according to the initial position of the X-type angular point characteristic and acquiring the two-dimensional coordinate of the X-type characteristic mark according to the characteristic extraction result.
Optionally, the primary positioning unit includes a grayscale processing unit, an image segmentation unit, a connected domain marking unit, and an identification unit, and the grayscale processing unit is configured to perform graying processing on the image of the X-type feature identifier; the image segmentation unit is used for carrying out threshold segmentation on the processed image according to the gray value to obtain an image segmentation result; the connected domain marking unit is used for marking the connected domain of the image segmentation result according to a preset connected domain marking algorithm; the identification unit is used for determining the preliminary positions of the X-type characteristic marks according to the connected domain marking result.
Optionally, the double-shield six-degree-of-freedom measurement system 100 based on the monocular vision system further includes a calibration module, where the calibration module is configured to obtain intrinsic parameters of the monocular camera, and establish a camera matrix according to the intrinsic parameters of the monocular camera; measuring inherent characteristics of the anterior shield by using a total station, and establishing an anterior shield coordinate system according to a measurement result; acquiring first three-dimensional calibration coordinates of a plurality of X-type feature identifiers in a anterior shield coordinate system; measuring inherent characteristics of a rear shield by using a total station, and establishing a rear shield coordinate system according to a measurement result; acquiring a calibration image of the X-type characteristic mark by adopting a monocular camera, and acquiring two-dimensional calibration coordinates of a plurality of X-type characteristic marks based on the calibration image; measuring second three-dimensional calibration coordinates of a plurality of X-type characteristic marks in a rear shield coordinate system by using a total station; calibrating a second rotation and translation relation of the shield coordinate system relative to the measurement system coordinate system according to the two-dimensional calibration coordinate and the second three-dimensional calibration coordinate; and storing the calibration result.
Optionally, the control module 40 further includes a coordinate system transformation processing unit, where the coordinate system transformation processing unit is configured to obtain a spatial geometric topological relation of the X-type feature identifier of the anterior shield rear end face; matching the X-type characteristic identifier in the anterior shield coordinate system with the two-dimensional coordinate of the X-type characteristic identifier in the image based on the space geometric topological relation; determining a first rotation and translation relation of the anterior shield coordinate system relative to the measurement system coordinate system according to the three-dimensional coordinate of the X-type characteristic identifier in the anterior shield coordinate system and the two-dimensional coordinate of the X-type characteristic identifier in the corresponding image by adopting a PNP algorithm; obtaining a second rotation and translation relation of the rear shield coordinate system obtained through calibration relative to the measurement system coordinate system; and calculating six-degree-of-freedom information of the anterior shield relative to the posterior shield by a coordinate system transformation algorithm.
Optionally, the plurality of X-type signatures 10 with light reflection characteristics include a first X-type signature, a second X-type signature, a third X-type signature, a fourth X-type signature, a fifth X-type signature, and a sixth X-type signature, where the first X-type signature and the sixth X-type signature are located on a first straight line, the second X-type signature, the third X-type signature, the fourth X-type signature, and the fifth X-type signature are located on a second straight line, the first straight line and the second straight line are parallel to the rear end surface of the front shield, and the first straight line intersects the second straight line.
Optionally, when the spatial position and the angular posture of the anterior shield change, the relative position relationship between the X-type feature marker and the anterior shield remains unchanged, the relative position relationship between the monocular camera and the posterior shield remains unchanged, the monocular camera has a first angle of view, the illumination light source has a second angle of view, the first angle of view is smaller than or equal to the second angle of view, and the motion section of the X-type feature marker falls into the first angle of view.
To sum up, the embodiment of the invention carries out measurement of six-degree-of-freedom information between double shields by arranging a monocular vision system, wherein the monocular vision system comprises a plurality of X-shaped characteristic marks with light reflection characteristics arranged on the rear end surface of a front shield, a monocular camera and an illumination light source arranged on the front end surface of a rear shield, a first inclination angle sensor coaxially arranged with the front shield, and a second inclination angle sensor coaxially arranged with the rear shield; after the measurement is started, controlling a monocular camera to acquire an image of the X-type characteristic mark and acquiring a two-dimensional coordinate of the X-type characteristic mark; matching the feature identifier in the anterior shield coordinate system with the two-dimensional coordinates of the feature identifier, determining a first rotational-translational relationship between the anterior shield coordinate system and the measurement system coordinate system, determining first six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining a second rotational-translational relationship between the posterior shield coordinate system and the measurement system coordinate system obtained by the calibration result, and the first six-degree-of-freedom information is corrected according to the real-time inclination angle data acquired by the inclination angle sensor, the real-time six-degree-of-freedom of the anterior shield relative to the posterior shield is determined, the posture positioning is realized by utilizing the characteristic points with the light reflecting characteristic, the method has low requirement on environmental conditions, solves the problems of low measurement precision and poor efficiency of the existing shield structure attitude positioning system of the heading machine, has high measurement efficiency and measurement precision and high system integration level, and is favorable for realizing the light weight and low power consumption performance of the attitude positioning system.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims (10)

1. A double-shield six-degree-of-freedom measurement method based on a monocular vision system is used for measuring relative six-degree-of-freedom information between a front shield and a rear shield of the double shields, and is characterized in that the monocular vision system comprises a plurality of X-shaped feature marks with light reflecting characteristics, which are arranged on the rear end face of the front shield, a monocular camera and an illuminating light source, which are arranged on the front end face of the rear shield, a first tilt sensor coaxially arranged with the front shield, and a second tilt sensor coaxially arranged with the rear shield, and the method comprises the following steps:
establishing a front shield coordinate system, a rear shield coordinate system and a measuring system coordinate system where the monocular camera is located, calibrating the front shield coordinate system, the rear shield coordinate system and the measuring system coordinate system, and storing a calibration result;
starting measurement, acquiring image data of an X-type characteristic mark acquired by a monocular camera, and determining a two-dimensional coordinate of the X-type characteristic mark in an image according to the image data;
performing PnP calculation according to a first three-dimensional calibration coordinate and a two-dimensional coordinate of a plurality of X-type feature identifiers on the rear end face of the anterior shield, acquiring a first rotational-translational relation of an anterior shield coordinate system relative to a measurement system coordinate system, and determining first six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining a second rotational-translational relation between the posterior shield coordinate system obtained through calibration and the measurement system coordinate system;
acquiring first inclination angle data of the first inclination angle sensor and second inclination angle data of the second inclination angle sensor;
and correcting the first six-degree-of-freedom information according to the first inclination angle data and the second inclination angle data, and determining the final real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield.
2. The binocular six-degree-of-freedom measurement method based on the monocular vision system as recited in claim 1, wherein the X-shaped feature markers include light absorbing sub markers and light reflecting sub markers, the light absorbing sub markers are made of light absorbing materials, the light reflecting sub markers are made of light reflecting materials, and the light absorbing sub markers and the light reflecting sub markers are alternately arranged at 90 degrees.
3. The binocular six-degree-of-freedom measurement method based on the monocular vision system according to claim 1 or 2, wherein the acquiring of the image data of the X-type feature collected by the monocular camera comprises the following steps:
controlling the illumination light source to be lightened;
controlling the monocular camera to continuously acquire image data of a plurality of X-type feature identifiers according to a preset sampling rule;
and receiving the image data of the plurality of X-type characteristic identifications through a network port.
4. The monocular vision system based double shield six-degree-of-freedom measurement method according to claim 1 or 2, wherein the determining the two-dimensional coordinates of the X-type feature in the image according to the image data comprises the following steps:
carrying out target initial positioning identification on the image data of the X-shaped feature identifier to obtain a plurality of initial positions of X-shaped corner point features;
extracting sub-pixel central points of the X-type feature identification by adopting a Hessian matrix based on the initial position of the X-type corner feature;
and acquiring the two-dimensional coordinates of the X-type feature identifier according to the feature extraction result.
5. The monocular vision system based double-shield six-degree-of-freedom measurement method according to claim 4, wherein the performing of the target initial positioning recognition on the image data of the X-type feature comprises the following steps:
carrying out gray processing on the image of the X-type feature identifier;
performing threshold segmentation on the processed image according to the gray value to obtain an image segmentation result;
adopting a connected domain marking algorithm to mark a connected domain of the image segmentation result;
and determining the preliminary positions of the X-type characteristic marks according to the connected domain marking result.
6. The monocular vision system-based double-shield six-degree-of-freedom measurement method according to claim 1 or 2, wherein the establishing of the anterior shield coordinate system, the posterior shield coordinate system and the measurement system coordinate system where the monocular camera is located, and the calibration of the anterior shield coordinate system, the posterior shield coordinate system and the measurement system coordinate system comprise the following steps:
acquiring intrinsic parameters of a monocular camera, and establishing a camera matrix according to the intrinsic parameters of the monocular camera;
measuring inherent characteristics of an anterior shield by using a total station, and establishing an anterior shield coordinate system according to a measurement result;
acquiring first three-dimensional calibration coordinates of a plurality of X-type feature identifiers in the anterior shield coordinate system;
measuring inherent characteristics of a rear shield by using a total station, and establishing a rear shield coordinate system according to a measurement result;
acquiring a calibration image of the X-type characteristic mark by adopting the monocular camera, and acquiring two-dimensional calibration coordinates of a plurality of X-type characteristic marks based on the calibration image;
measuring second three-dimensional calibration coordinates of the plurality of X-type feature identifiers in the rear shield coordinate system by using a total station;
calibrating a second rotation and translation relation of the posterior shield coordinate system relative to the measurement system coordinate system according to the two-dimensional calibration coordinate and the second three-dimensional calibration coordinate;
and storing the calibration result.
7. The double-shield six-degree-of-freedom measurement method based on the monocular vision system according to claim 1 or 2, wherein PnP calculation is performed according to a first three-dimensional calibration coordinate of a plurality of X-type feature identifiers on a rear end surface of the anterior shield and the two-dimensional coordinate, a first rotational-translational relationship of the anterior shield coordinate system relative to a measurement system coordinate system is obtained, and first six-degree-of-freedom information of the anterior shield relative to the posterior shield is determined by combining a second rotational-translational relationship between the posterior shield coordinate system obtained by calibration and the measurement system coordinate system, and the method comprises the following steps:
acquiring a space geometric topological relation of an X-shaped characteristic mark of the rear end face of the anterior shield;
matching the X-type characteristic identifier in the anterior shield coordinate system with the two-dimensional coordinate of the X-type characteristic identifier in the image based on the space geometric topological relation;
determining a first rotation and translation relation of the anterior shield coordinate system relative to the measurement system coordinate system according to the three-dimensional coordinates of the X-type feature identifier in the anterior shield coordinate system and the two-dimensional coordinates of the X-type feature identifier in the corresponding image by adopting a PNP algorithm;
obtaining a second rotation and translation relation of the calibrated trailing shield coordinate system relative to the measurement system coordinate system;
and calculating first six-degree-of-freedom information of the anterior shield relative to the posterior shield through a coordinate system transformation algorithm.
8. The double-shield six-degree-of-freedom measurement method based on the monocular vision system according to claim 1 or 2, wherein the plurality of X-type signatures with reflective characteristics includes a first X-type signature, a second X-type signature, a third X-type signature, a fourth X-type signature, a fifth X-type signature, and a sixth X-type signature, the first X-type signature and the sixth X-type signature are located on a first straight line, the second X-type signature, the third X-type signature, the fourth X-type signature, and the fifth X-type signature are located on a second straight line, the first straight line and the second straight line are parallel to the anterior shield posterior end face, and the first straight line intersects the second straight line.
9. The monocular vision system based double-shield six-degree-of-freedom measurement method according to claim 1 or 2, wherein a relative positional relationship between the X-type feature marker and the anterior shield is kept unchanged, a relative positional relationship between the monocular camera and the posterior shield is kept unchanged, the monocular camera has a first angle of view, the illumination light source has a second angle of view, the first angle of view is smaller than or equal to the second angle of view, and a motion section of the X-type feature marker falls into the first angle of view.
10. The utility model provides a two shield six degrees of freedom measurement system based on monocular vision system for measure the relative six degrees of freedom information between the anterior shield and the posterior shield of two shields, its characterized in that includes: the device comprises a plurality of X-shaped characteristic marks with light reflecting characteristics, a monocular camera, an illuminating light source and a control module, a first inclination angle sensor and a second inclination angle sensor, wherein the X-shaped characteristic marks are arranged on the rear end face of a front shield;
the control module is used for controlling the lighting source to be turned on or off, controlling the monocular camera to acquire image data of the X-type characteristic identifier, determining a two-dimensional coordinate of the X-type characteristic identifier in an image according to the image data, performing PnP (PnP) calculation according to a first three-dimensional calibration coordinate and a two-dimensional coordinate of a plurality of X-type characteristic identifiers on the rear end face of the anterior shield, acquiring a first rotational translation relation of the anterior shield coordinate system relative to a measurement system coordinate system, and determining first six-degree-of-freedom information of the anterior shield relative to the posterior shield by combining a second rotational translation relation between the posterior shield coordinate system obtained by calibration and the measurement system coordinate system;
the control module is further configured to acquire first inclination data of the first inclination sensor and second inclination data of the second inclination sensor, correct the first six-degree-of-freedom information according to the first inclination data and the second inclination data, and determine final real-time six-degree-of-freedom information of the anterior shield relative to the posterior shield.
CN202011280179.2A 2020-11-16 2020-11-16 Double-shield six-degree-of-freedom measurement method and system based on monocular vision system Active CN112525158B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011280179.2A CN112525158B (en) 2020-11-16 2020-11-16 Double-shield six-degree-of-freedom measurement method and system based on monocular vision system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011280179.2A CN112525158B (en) 2020-11-16 2020-11-16 Double-shield six-degree-of-freedom measurement method and system based on monocular vision system

Publications (2)

Publication Number Publication Date
CN112525158A true CN112525158A (en) 2021-03-19
CN112525158B CN112525158B (en) 2022-05-13

Family

ID=74980950

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011280179.2A Active CN112525158B (en) 2020-11-16 2020-11-16 Double-shield six-degree-of-freedom measurement method and system based on monocular vision system

Country Status (1)

Country Link
CN (1) CN112525158B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110017810A (en) * 2019-05-16 2019-07-16 湖北工业大学 A kind of photoelectrical position sensor and monocular vision assembled gesture measuring system and method
CN110726373A (en) * 2019-10-30 2020-01-24 上海米度测量技术有限公司 Dynamic real-time measuring device for shield tail clearance of shield machine
CN110864692A (en) * 2019-11-29 2020-03-06 北京龙田华远科技有限公司 Pose determination method of heading machine
CN111457904A (en) * 2020-03-20 2020-07-28 力信测量(上海)有限公司 Be used for two shield TBM vision measurement systems
CN111551152A (en) * 2020-06-04 2020-08-18 江苏集萃智能光电系统研究所有限公司 Monocular vision-based relative pose measurement method and device for near space aircraft

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110017810A (en) * 2019-05-16 2019-07-16 湖北工业大学 A kind of photoelectrical position sensor and monocular vision assembled gesture measuring system and method
CN110726373A (en) * 2019-10-30 2020-01-24 上海米度测量技术有限公司 Dynamic real-time measuring device for shield tail clearance of shield machine
CN110864692A (en) * 2019-11-29 2020-03-06 北京龙田华远科技有限公司 Pose determination method of heading machine
CN111457904A (en) * 2020-03-20 2020-07-28 力信测量(上海)有限公司 Be used for two shield TBM vision measurement systems
CN111551152A (en) * 2020-06-04 2020-08-18 江苏集萃智能光电系统研究所有限公司 Monocular vision-based relative pose measurement method and device for near space aircraft

Also Published As

Publication number Publication date
CN112525158B (en) 2022-05-13

Similar Documents

Publication Publication Date Title
CN109029257B (en) Large-scale workpiece pose measurement system and method based on stereoscopic vision and structured light vision
US8180101B2 (en) Calibration method for structure parameters of structured-light vision sensor
CN111833333A (en) Binocular vision-based boom type tunneling equipment pose measurement method and system
CN108594245A (en) A kind of object movement monitoring system and method
CN102927908B (en) Robot eye-on-hand system structured light plane parameter calibration device and method
CN110728715A (en) Camera angle self-adaptive adjusting method of intelligent inspection robot
CN110146038A (en) The distributed monocular camera laser measuring device for measuring and method of cylindrical member assembly corner
CN110966932B (en) Structured light three-dimensional scanning method based on known mark points
EP3783385A1 (en) Combined point cloud generation using a stationary laser scanner and a mobile scanner
CN109029299A (en) The double camera measuring device and measuring method of bay section pin hole docking corner
CN110736446B (en) Pose identification system and method for cantilever type heading machine
CN111220126A (en) Space object pose measurement method based on point features and monocular camera
CN105783880B (en) A kind of monocular laser assisted bay section docking calculation
CN110966956A (en) Binocular vision-based three-dimensional detection device and method
JPWO2020003497A1 (en) Imaging equipment calibration equipment, monitoring equipment, work machines and calibration methods
CN109238247A (en) A kind of six degree of freedom measurement method towards the complicated scene of large space
CN111879354A (en) Unmanned aerial vehicle measurement system that becomes more meticulous
CN105574812A (en) Multi-angle three-dimensional data registration method and device
CN109917433A (en) A kind of fly able measuring system and flight measurement method
CN107328358B (en) The measuring system and measurement method of aluminium cell pose
CN112284360B (en) Double-shield six-degree-of-freedom measurement method and system based on binocular vision system
CN113781576B (en) Binocular vision detection system, method and device for adjusting pose with multiple degrees of freedom in real time
CN112525158B (en) Double-shield six-degree-of-freedom measurement method and system based on monocular vision system
CN106840137B (en) Automatic positioning and orienting method of four-point type heading machine
CN111044039B (en) Monocular target area self-adaptive high-precision distance measurement device and method based on IMU

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