WO2023273023A1 - Method for measuring and calculating four dimensions of motion position parameters of mucking loader - Google Patents

Method for measuring and calculating four dimensions of motion position parameters of mucking loader Download PDF

Info

Publication number
WO2023273023A1
WO2023273023A1 PCT/CN2021/124915 CN2021124915W WO2023273023A1 WO 2023273023 A1 WO2023273023 A1 WO 2023273023A1 CN 2021124915 W CN2021124915 W CN 2021124915W WO 2023273023 A1 WO2023273023 A1 WO 2023273023A1
Authority
WO
WIPO (PCT)
Prior art keywords
slag
image
auxiliary line
scraper
remover
Prior art date
Application number
PCT/CN2021/124915
Other languages
French (fr)
Chinese (zh)
Inventor
吴经纬
黄杏往
Original Assignee
中冶南方工程技术有限公司
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 中冶南方工程技术有限公司 filed Critical 中冶南方工程技术有限公司
Publication of WO2023273023A1 publication Critical patent/WO2023273023A1/en

Links

Images

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
    • G01B11/002—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates

Definitions

  • the invention relates to the metallurgical industry, in particular to a detection and calculation method for four-dimensional movement position parameters of a slag removal machine.
  • the main technical method at present is to directly detect by installing sensors on the slag scraper body, including but not limited to installing encoders on the rotating shaft, Install an inclination sensor on the slag removal arm, a position sensor in the hydraulic cylinder, etc.
  • the slag remover is usually installed in a harsh environment full of high temperature and metal dust; the slag remover moves frequently and vibrates greatly during operation, which is very detrimental to the detection accuracy, stability and reliability of the sensor.
  • the sensor installed in the oil cylinder fails, it is very difficult to overhaul.
  • the embodiment of the present invention provides a method for detecting and calculating the four-dimensional motion position parameters of the slag scraper that overcomes the above problems or at least partially solves the above problems.
  • the specific scheme is as follows:
  • a method for detecting and calculating a four-dimensional movement position parameter of a slag removing machine comprising:
  • Step 1 set a marker as a reference object on the slag remover, set a camera above and on the side of the slag remover, and take a picture of the current movement position of the slag remover through the camera;
  • Step 2 Carry out real-time calculation and analysis on the image to obtain the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper.
  • step 2 the real-time calculation and analysis of the image is carried out to obtain the rotation angle of the slag removal machine, which specifically includes:
  • Step 2.11 rotate the slag scraper to the 0° position and keep it still, and obtain the image taken by the camera above the slag scraper at this moment, and draw an auxiliary line M along the direction of the slag scraper on the image taken by the camera above the slag scraper , the auxiliary line M coincides with the axis of the slag arm of the slag remover, that is, passes through the geometric center point of the marker used for detecting the rotation angle of the slag remover and the center of the rotation axis of the slag remover;
  • Step 2.12 when the slag scraper runs to any position, the image of the slag scraper is collected in real time through the camera above the slag scraper, based on the neural network model of object recognition and tracking, the current position information of the markers in the image is obtained, and the sign is obtained
  • the rotation angle of the slag scraper is corrected by the pitch angle ⁇ Get the rotation angle of the slag remover for:
  • step 2 the image is calculated and analyzed in real time to obtain the pitch angle of the slag removal machine, which specifically includes:
  • Step 2.21 Pitch the slag scraper to the horizontal position and keep it still. At this time, the pitch angle is regarded as 0°, and the image taken by the side camera of the slag scraper at this moment is obtained, and the slag scraper is moved along the image taken by the side camera of the slag scraper.
  • a t y 4 -y 3
  • B t x 3 -x 4
  • C t x 4 ⁇ y 3 -x 3 ⁇ y 4 ;
  • Step 2.22 when the slag scraper runs to any position, the image of the slag scraper is collected in real time through the side camera of the slag scraper, based on the neural network model of object recognition and tracking, the current position information of the marker in the image is obtained, and the The coordinate P t (x 0 ,y 0 ) of the geometric center point of the marker on the image;
  • the pitch angle ⁇ of the slag remover is calculated as:
  • the image is calculated and analyzed in real time, and the lift height of the slag removal machine specifically includes:
  • Step 2.31 lower the body of the slag scraper to the lowest position and keep it still, and obtain the image taken by the camera on the side of the slag scraper at this moment. On the image taken by the side camera of the slag scraper, draw a line parallel to the ground Auxiliary line S of ping foundation;
  • a h y 6 -y 5
  • B h x 5 -x 6
  • C h x 6 ⁇ y 5 -x 5 ⁇ y 6 ;
  • Step 2.32 when the slag scraper runs to any position, collect the image of the slag scraper in real time through the side camera, and obtain the current position information of the marker in the image based on the neural network model of object recognition and tracking, and obtain the geometry of the marker The coordinates Ph (x h , y h ) of the center point on the image;
  • the image is calculated and analyzed in real time, and the front and rear telescopic distances of the slag removal machine specifically include:
  • Step 2.41 retract the slag scraper to the rearmost position, rotate the slag scraper to the 0° position, pitch to the horizontal position, and then keep still, and obtain the image taken by the side camera of the slag scraper at this moment, and then the slag scraper On the image taken by the camera on the side of the machine, draw an auxiliary line W perpendicular to the foundation of the floor at the rear end of the slag picker;
  • Step 2.42 when the slag scraper runs to any position, the image of the slag scraper is collected in real time through the side camera, and based on the neural network model of object recognition and tracking, the current position information of the marker in the image is obtained, and the geometry of the marker is obtained.
  • the method also includes:
  • a preset number of frame pictures are continuously extracted at equal intervals; for each frame of picture, the pitch axis, rotation axis, and markers of the slag scraper in the picture are marked to obtain each frame
  • the motion parameter information of the slag scraper in the picture and a data set is constructed based on the motion parameter information, wherein the motion parameter information includes: the rotation angle, pitch angle, lifting height, front and rear telescopic distance and marker position of the slag scraper;
  • a neural network model for object recognition and tracking is trained through the dataset.
  • markers which are respectively used as reference objects when calculating the rotation angle, pitch angle, lifting height and front and rear telescopic distance, among which, the markers used for calculating the pitch angle and rotation angle of the slag scraper , its geometric center point should be on the central axis of the slag removal arm.
  • the present invention uses a camera installed above the slag remover, on the side of the slag remover, etc., to take images of the current movement position of the slag remover, and perform real-time calculation and analysis on the images, so as to obtain the rotation angle, pitch angle, Parameters such as lifting height and front and rear telescopic distance.
  • the present invention does not need to install sensors on the slag remover, thereby well avoiding the difficulty of sensor maintenance and the problems that the detection accuracy, stability and reliability of the sensor are affected by the frequent movements and vibrations of the slag remover during operation.
  • Fig. 1 is a schematic flowchart of a method for detecting and calculating a four-dimensional motion position parameter of a slag removal machine provided by an embodiment of the present invention
  • Fig. 2 is a schematic diagram of the installation position of the camera above the slag removal machine provided by the embodiment of the present invention
  • Fig. 3 is a schematic diagram of the installation position of the camera on the side of the slag removal machine provided by the embodiment of the present invention
  • Fig. 4 is a schematic diagram of a type of slag removal machine with no pitching action at the tail provided by an embodiment of the present invention
  • Fig. 5 is a schematic diagram of a slag removal machine with a tail that performs both rotating and pitching motions provided by an embodiment of the present invention.
  • a method for detecting and calculating a four-dimensional motion position parameter of a slag removal machine provided by an embodiment of the present invention, the method includes:
  • Step 1 set a marker as a reference object on the slag remover, set a camera above and on the side of the slag remover, and take a picture of the current movement position of the slag remover through the camera;
  • Step 2 Carry out real-time calculation and analysis on the image to obtain the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper.
  • the present invention uses a camera installed above the slag remover, on the side of the slag remover, etc., to take images of the current movement position of the slag remover, and perform real-time calculation and analysis on the images, so as to obtain the rotation angle, pitch angle, Parameters such as lifting height and front and rear telescopic distance.
  • the camera can be a visible light camera, an infrared camera or a far infrared camera.
  • one or more machine vision servers need to be set up for image processing, analysis, and release of calculation results.
  • machine vision servers are not necessary.
  • the detection and calculation methods described in the present invention It can also be directly integrated into the camera and implemented by the camera's own software, so as to directly realize image processing, calculation and release of results.
  • the present invention needs to select appropriate markers on the body of the slag removal machine for image analysis. If there is no suitable marker on the body of the slag remover, a special marker can be made and installed on the slag remover.
  • the marker should be easy to identify and can be distinguished from the background. Among them, the signs used to calculate the pitch angle and rotation angle of the slag remover Object, its geometric center point should be on the central axis of the slag removal arm.
  • the concrete enforcement of the present invention can be divided into two stages:
  • step (4) Use the data set produced in step (4) to train the neural network to obtain the object recognition and tracking model.
  • the object recognition and tracking model does not have to use neural network technology, and the object recognition and tracking model obtained by using other technical solutions and achieving the same results does not deviate from the spirit and scope of the technical solution of this patent.
  • Rotation angle of slag removal machine Calculation of pitch angle ⁇ , lifting height H and front and rear telescopic distance D.
  • the pitch angle ⁇ can be detected and calculated by the method of this patent; for the slag scraper whose tail is both rotating and pitching, the rotation angle
  • the two parameters of the pitch angle ⁇ are not independent, and one of them needs to be obtained by other methods. See Figures 4 and 5 for illustrations of two different types of slag removal machines;
  • auxiliary line N can face any direction, the length of the auxiliary line N on the image is la, and the corresponding actual physical length is L a ; then the object-to-image ratio k b can be calculated;
  • the above steps (1) to (4) are preparatory work.
  • the camera above the slag scraper collects the image of the slag scraper in real time.
  • the image first passes through the object recognition and Track the calculation of the model to obtain the current position information of the marker, and further calculate the coordinate P s (x 0 ,y 0 ) of the geometric center point of the marker on the image;
  • a t y 4 -y 3
  • B t x 3 -x 4
  • C t x 4 ⁇ y 3 -x 3 ⁇ y 4 ;
  • Steps (1) to (4) above are preparatory work.
  • the image of the side camera of the slag scraper is collected in real time.
  • the image first passes through the calculation of the object recognition and tracking model completed in stage one training to obtain the current position information of the marker, and further calculate the coordinate P t (x 0 ,y 0 ) of the geometric center point of the marker on the image;
  • a h y 6 -y 5
  • B h x 5 -x 6
  • C h x 6 ⁇ y 5 -x 5 ⁇ y 6 ;
  • Steps (1) to (4) above are preparatory work.
  • the side camera collects the images of the slag scraper in real time.
  • the image first passes through the calculation of the object recognition and tracking model completed in stage one training, and obtains the current position information of the marker used for the calculation of the lifting height, and can further calculate the coordinate Ph (x h ,y h );
  • Steps (1) to (4) above are preparatory work.
  • the side camera collects the images of the slag scraper in real time.
  • the image first passes through the calculation of the object recognition and tracking model completed in stage one training, and obtains the current position information of the marker used for calculating the front and rear telescopic distances, and can further calculate the coordinate P d (x d ,y d );
  • item (3) in the calculation method of the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper is a method to obtain the image-to-image ratio, and other methods are used to obtain the object-to-image ratio.
  • the method applied to the subsequent calculation process still does not deviate from the spirit and scope of the technical solution of this patent.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Image Analysis (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

A method for measuring and calculating four dimensions of motion position parameters of a mucking loader. The method comprises: step 1, providing a marker on a mucking loader as a reference object, and arranging cameras above the mucking loader and at a side of the mucking loader, so as to capture the current motion position image of the mucking loader by means of the cameras; and step 2, performing real-time calculation analysis on the image to obtain a rotation angle, a pitch angle, a lifting height and a front-back extension distance of the mucking loader. By means of the present invention, there is no need to mount a sensor on a mucking loader, such that the problems of the difficulty in the overhaul of the sensor, and the effect on the measurement precision, stability and reliability of the sensor caused by frequent movements and vibrations of the mucking loader during operation are effectively solved and avoided.

Description

一种扒渣机四维度运动位置参量的检测和计算方法A detection and calculation method for four-dimensional motion position parameters of a slag scraper 技术领域technical field
本发明涉及冶金行业，具体涉及一种扒渣机四维度运动位置参量的检测和计算方法。The invention relates to the metallurgical industry, in particular to a detection and calculation method for four-dimensional movement position parameters of a slag removal machine.
背景技术Background technique
扒渣机旋转角度、俯仰角度、升降高度与前后伸缩距离的检测，目前主要的技术手段是采用在扒渣机本体上安装传感器的方式直接检测，包括但不限于在旋转轴上安装编码器、在扒渣臂上安装倾角传感器、在液压油缸内安装位置传感器等。扒渣机通常安装在高温、充满金属粉尘的恶劣环境；扒渣机工作时动作频繁且震动较大，对传感器的检测精度、稳定性和可靠性都有十分不利。此外，安装在油缸中的传感器一旦出现故障，检修非常困难。For the detection of the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper, the main technical method at present is to directly detect by installing sensors on the slag scraper body, including but not limited to installing encoders on the rotating shaft, Install an inclination sensor on the slag removal arm, a position sensor in the hydraulic cylinder, etc. The slag remover is usually installed in a harsh environment full of high temperature and metal dust; the slag remover moves frequently and vibrates greatly during operation, which is very detrimental to the detection accuracy, stability and reliability of the sensor. In addition, once the sensor installed in the oil cylinder fails, it is very difficult to overhaul.
发明内容Contents of the invention
鉴于现有技术中存在的技术缺陷和技术弊端，本发明实施例提供克服上述问题或者至少部分地解决上述问题的一种扒渣机四维度运动位置参量的检测和计算方法，具体方案如下：In view of the technical defects and technical disadvantages in the prior art, the embodiment of the present invention provides a method for detecting and calculating the four-dimensional motion position parameters of the slag scraper that overcomes the above problems or at least partially solves the above problems. The specific scheme is as follows:
一种扒渣机四维度运动位置参量的检测和计算方法，所述方法包括:A method for detecting and calculating a four-dimensional movement position parameter of a slag removing machine, the method comprising:
步骤1，在扒渣机上设置作为参照物的标志物，在扒渣机上方和侧方设置摄像机，通过摄像机拍摄扒渣机当前的运动位置图像；Step 1, set a marker as a reference object on the slag remover, set a camera above and on the side of the slag remover, and take a picture of the current movement position of the slag remover through the camera;
步骤2，对图像进行实时的计算分析，得到扒渣机的旋转角度、俯仰角度、升降高度与前后伸缩距离。Step 2: Carry out real-time calculation and analysis on the image to obtain the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper.
进一步地，步骤2中，对图像进行实时的计算分析，得到扒渣机的旋转 角度具体包括：Further, in step 2, the real-time calculation and analysis of the image is carried out to obtain the rotation angle of the slag removal machine, which specifically includes:
步骤2.11，将扒渣机旋转至0°位置后保持静止，并获取该时刻扒渣机上方摄像机拍摄的图像，在扒渣机上方的摄像机拍摄的图像上沿扒渣机方向画一条辅助线M，辅助线M与扒渣机渣臂轴线重合，即穿过扒渣机旋转角度检测用的标志物的几何中心点和扒渣机旋转轴中心；Step 2.11, rotate the slag scraper to the 0° position and keep it still, and obtain the image taken by the camera above the slag scraper at this moment, and draw an auxiliary line M along the direction of the slag scraper on the image taken by the camera above the slag scraper , the auxiliary line M coincides with the axis of the slag arm of the slag remover, that is, passes through the geometric center point of the marker used for detecting the rotation angle of the slag remover and the center of the rotation axis of the slag remover;
在扒渣机上方的摄像机拍摄的图像上，测量出旋转角度计算用的标志物的几何中心点至扒渣机旋转轴中心的物理距离L sw； On the image taken by the camera above the slag scraper, measure the physical distance L sw from the geometric center point of the marker used for calculating the rotation angle to the center of the slag scraper rotation axis;
在扒渣机上方摄像机拍摄的图像上，于扒渣机尾部另画一条辅助线N，令辅助线N在图像上的长度为l a，对应的实际物理长度为L a，进而可计算出物像比k a； On the image taken by the camera above the slag remover, draw another auxiliary line N at the tail of the slag remover, let the length of the auxiliary line N on the image be l a , and the corresponding actual physical length is L a , and then the object can be calculated Like than k a ;
Figure PCTCN2021124915-appb-000001
Figure PCTCN2021124915-appb-000001
任取辅助线M上的两个点P(x 1,y 1)，P(x 2,y 2)，可得到辅助线M在图像上的直线方程； Take any two points P(x 1 , y 1 ) and P(x 2 , y 2 ) on the auxiliary line M to obtain the straight line equation of the auxiliary line M on the image;
A s·x+B s·y+C s＝0； A s x+B s y+C s = 0;
其中，A s＝y 2–y 1；B s＝x 1–x 2；C s＝x 2·y 1–x 1·y 2； Among them, A s =y 2 -y 1 ; B s =x 1 -x 2 ; C s =x 2 ·y 1 -x 1 ·y 2 ;
步骤2.12，当扒渣机运行到任意位置时，通过扒渣机上方摄像机实时采集扒渣机的图像，基于物体识别与跟踪的神经网络模型，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P s(x 0,y 0)； Step 2.12, when the slag scraper runs to any position, the image of the slag scraper is collected in real time through the camera above the slag scraper, based on the neural network model of object recognition and tracking, the current position information of the markers in the image is obtained, and the sign is obtained The coordinates P s (x 0 ,y 0 ) of the geometric center point of the object on the image;
计算P s(x 0,y 0)至辅助线M的实际物理距离L p； Calculate the actual physical distance L p from P s (x 0 ,y 0 ) to the auxiliary line M;
Figure PCTCN2021124915-appb-000002
Figure PCTCN2021124915-appb-000002
对于尾部不做俯仰动作的扒渣机，计算扒渣机旋转角度
Figure PCTCN2021124915-appb-000003
为： For the slag scraper without pitching action at the tail, calculate the rotation angle of the slag scraper
Figure PCTCN2021124915-appb-000003
for:
Figure PCTCN2021124915-appb-000004
Figure PCTCN2021124915-appb-000004
对于尾部既做旋转动作又作俯仰动作的扒渣机，通过俯仰角度θ修正扒渣机旋转角度
Figure PCTCN2021124915-appb-000005
得到扒渣机旋转角度
Figure PCTCN2021124915-appb-000006
为： For the slag scraper whose tail is both rotating and pitching, the rotation angle of the slag scraper is corrected by the pitch angle θ
Figure PCTCN2021124915-appb-000005
Get the rotation angle of the slag remover
Figure PCTCN2021124915-appb-000006
for:
Figure PCTCN2021124915-appb-000007
Figure PCTCN2021124915-appb-000007
计算出扒渣机旋转角度
Figure PCTCN2021124915-appb-000008
后，根据P s(x 0,y 0)与辅助线M在图像上的相对位置，修正扒渣机旋转角度
Figure PCTCN2021124915-appb-000009
的正负符号。 Calculate the rotation angle of the slag scraper
Figure PCTCN2021124915-appb-000008
Finally, according to the relative position of P s (x 0 ,y 0 ) and the auxiliary line M on the image, correct the rotation angle of the slag remover
Figure PCTCN2021124915-appb-000009
positive and negative signs.
进一步地，步骤2中，对图像进行实时的计算分析，得到扒渣机的俯仰角度具体包括：Further, in step 2, the image is calculated and analyzed in real time to obtain the pitch angle of the slag removal machine, which specifically includes:
步骤2.21，将扒渣机俯仰至水平位置后保持静止，此时俯仰角认为是0°，并获取该时刻扒渣机侧面摄像机拍摄的图像，在扒渣机侧面摄像机拍摄的图像上沿扒渣机方向画一条辅助线J，辅助线J与扒渣机渣臂轴线重合，即穿过扒渣机俯仰角度检测用标志物的几何中心点和扒渣机俯仰轴中心；Step 2.21: Pitch the slag scraper to the horizontal position and keep it still. At this time, the pitch angle is regarded as 0°, and the image taken by the side camera of the slag scraper at this moment is obtained, and the slag scraper is moved along the image taken by the side camera of the slag scraper. Draw an auxiliary line J in the machine direction, and the auxiliary line J coincides with the axis of the slag arm of the slag remover, that is, passes through the geometric center point of the marker for detecting the pitch angle of the slag remover and the center of the pitch axis of the slag remover;
在扒渣机侧面摄像机拍摄的图像上，测量出俯仰角度计算用的标志物的几何中心点至扒渣机俯仰轴中心的物理距离；Measure the physical distance from the geometric center point of the marker used for pitch angle calculation to the center of the pitch axis of the slag remover on the image taken by the side camera of the slag remover;
在扒渣机侧面摄像机拍摄的图像上，于扒渣机上另画一条辅助线K，令辅助线K在图像上的长度为l b，对应的实际物理长度为L b；进而可计算出物像比k b； On the image taken by the camera on the side of the slag remover, draw another auxiliary line K on the slag remover, so that the length of the auxiliary line K on the image is l b , and the corresponding actual physical length is L b ; then the object image can be calculated than k b ;
Figure PCTCN2021124915-appb-000010
Figure PCTCN2021124915-appb-000010
任取辅助线J上的两个点P(x 3,y 3)，P(x 4,y 4)，可得到辅助线J在图像上的直线方程： Taking any two points P(x 3 ,y 3 ) and P(x 4 ,y 4 ) on the auxiliary line J, the straight line equation of the auxiliary line J on the image can be obtained:
A t·x+B t·y+C t＝0； A t x+B t y+C t = 0;
其中，A t＝y 4–y 3，B t＝x 3–x 4，C t＝x 4·y 3–x 3·y 4； Among them, A t =y 4 -y 3 , B t =x 3 -x 4 , C t =x 4 ·y 3 -x 3 ·y 4 ;
步骤2.22，当扒渣机运行到任意位置时，通过扒渣机侧方摄像机实时采集扒渣机的图像，基于物体识别与跟踪的神经网络模型，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P t(x 0,y 0)； Step 2.22, when the slag scraper runs to any position, the image of the slag scraper is collected in real time through the side camera of the slag scraper, based on the neural network model of object recognition and tracking, the current position information of the marker in the image is obtained, and the The coordinate P t (x 0 ,y 0 ) of the geometric center point of the marker on the image;
计算P t(x 0,y 0)至辅助线J的实际距离L q； Calculate the actual distance L q from P t (x 0 ,y 0 ) to the auxiliary line J;
Figure PCTCN2021124915-appb-000011
Figure PCTCN2021124915-appb-000011
对于尾部不做俯仰动作的扒渣机，计算扒渣机俯仰角度θ为：For the slag remover without pitching action at the tail, the pitch angle θ of the slag remover is calculated as:
Figure PCTCN2021124915-appb-000012
Figure PCTCN2021124915-appb-000012
对于尾部既做旋转动作又作俯仰动作的扒渣机，通过旋转角度
Figure PCTCN2021124915-appb-000013
修正扒渣机俯仰角度θ为： For the slag scraper whose tail is both rotating and pitching, the rotation angle
Figure PCTCN2021124915-appb-000013
Correct the pitch angle θ of the slag remover as:
Figure PCTCN2021124915-appb-000014
Figure PCTCN2021124915-appb-000014
计算出扒渣机俯仰角度θ后，根据P t(x 0,y 0)与辅助线J在图像上的相对位置，修正扒渣机旋转角度
Figure PCTCN2021124915-appb-000015
的正负符号。 After calculating the pitch angle θ of the slag scraper, correct the rotation angle of the slag scraper according to the relative position of P t (x 0 , y 0 ) and the auxiliary line J on the image
Figure PCTCN2021124915-appb-000015
positive and negative signs.
进一步地，对图像进行实时的计算分析，得到扒渣机的升降高度具体包括：Further, the image is calculated and analyzed in real time, and the lift height of the slag removal machine specifically includes:
步骤2.31，将扒渣机本体降至最低位置后保持静止，并获取该时刻扒渣机侧面摄像机拍摄的图像，在扒渣机侧面摄像机拍摄的图像上，沿扒渣机底座画一条平行于地坪基础的辅助线S；Step 2.31, lower the body of the slag scraper to the lowest position and keep it still, and obtain the image taken by the camera on the side of the slag scraper at this moment. On the image taken by the side camera of the slag scraper, draw a line parallel to the ground Auxiliary line S of ping foundation;
在扒渣机侧面摄像机拍摄的图像上，测量出升降高度计算用的标志物的几何中心点至辅助线S的物理距离；Measure the physical distance from the geometric center point of the marker used for lifting height calculation to the auxiliary line S on the image taken by the side camera of the slag removal machine;
在扒渣机侧面摄像机拍摄的图像上，于扒渣机上另画一条辅助线T，令辅助线T在图像上的长度为l c，对应的实际物理长度为L c；进而可计算出物像比k c； On the image taken by the camera on the side of the slag remover, draw another auxiliary line T on the slag remover, so that the length of the auxiliary line T on the image is l c , and the corresponding actual physical length is L c ; then the object image can be calculated than k c ;
Figure PCTCN2021124915-appb-000016
Figure PCTCN2021124915-appb-000016
任取辅助线S上的两个点P(x 5,y 5)，P(x 6,y 6)，可得到辅助线S在图像上的直线方程； Take any two points P(x 5 , y 5 ) and P(x 6 , y 6 ) on the auxiliary line S to obtain the straight line equation of the auxiliary line S on the image;
A h·x+B h·y+C h＝0； A h x+B h y+C h = 0;
其中，A h＝y 6–y 5，B h＝x 5–x 6，C h＝x 6·y 5–x 5·y 6； Among them, A h =y 6 -y 5 , B h =x 5 -x 6 , C h =x 6 ·y 5 -x 5 ·y 6 ;
步骤2.32，当扒渣机运行到任意位置时，通过侧面摄像机实时采集扒渣机的图像，基于物体识别与跟踪的神经网络模型，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P h(x h,y h)； Step 2.32, when the slag scraper runs to any position, collect the image of the slag scraper in real time through the side camera, and obtain the current position information of the marker in the image based on the neural network model of object recognition and tracking, and obtain the geometry of the marker The coordinates Ph (x h , y h ) of the center point on the image;
计算扒渣机实际抬升高度H；Calculate the actual lifting height H of the slag remover;
Figure PCTCN2021124915-appb-000017
Figure PCTCN2021124915-appb-000017
进一步地，对图像进行实时的计算分析，得到扒渣机的前后伸缩距离具体包括：Further, the image is calculated and analyzed in real time, and the front and rear telescopic distances of the slag removal machine specifically include:
步骤2.41，将扒渣机缩回至最后端位置，并将扒渣机旋转至0°位置，俯仰至水平位置，然后保持静止，并获取该时刻扒渣机侧面摄像机拍摄的图像，在扒渣机侧面摄像机拍摄图像上，在扒渣机最后端画一条垂直于地坪基础的辅助线W；Step 2.41, retract the slag scraper to the rearmost position, rotate the slag scraper to the 0° position, pitch to the horizontal position, and then keep still, and obtain the image taken by the side camera of the slag scraper at this moment, and then the slag scraper On the image taken by the camera on the side of the machine, draw an auxiliary line W perpendicular to the foundation of the floor at the rear end of the slag picker;
在扒渣机侧面摄像机拍摄图像上，测量出前后伸缩距离计算用的标志物的几何中心点至辅助线W的物理距离L rd； Measure the physical distance L rd from the geometric center point of the marker used for the calculation of the front and rear telescopic distance to the auxiliary line W on the image taken by the side camera of the slag removal machine;
在扒渣机侧面摄像机拍摄的图像上，于扒渣机上另画一条辅助线V，辅助线V在图像上的长度为l d，对应的实际物理长度为L d；进而可计算出物像比k d： On the image taken by the side camera of the slag remover, draw another auxiliary line V on the slag remover. The length of the auxiliary line V on the image is l d , and the corresponding actual physical length is L d ; then the object-to-image ratio can be calculated k d :
Figure PCTCN2021124915-appb-000018
Figure PCTCN2021124915-appb-000018
任取辅助线W上的两个点P(x 7,y 7)，P(x 8,y 8)，可得到辅助线W在图像上的直线方程如下： Taking any two points P(x 7 , y 7 ) and P(x 8 , y 8 ) on the auxiliary line W, the linear equation of the auxiliary line W on the image can be obtained as follows:
A d·x+B d·y+C d＝0； A d x + B d y + C d = 0;
其中，A d＝y 8–y 7；B d＝x 7–x 8；C d＝x 8·y 7–x 7·y 8； Wherein, A d =y 8 -y 7 ; B d =x 7 -x 8 ; C d =x 8 ·y 7 -x 7 ·y 8 ;
步骤2.42，当扒渣机运行到任意位置时，通过侧面摄像机实时采集扒渣机的图像，基于物体识别与跟踪的神经网络模型，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P d(x d,y d)； Step 2.42, when the slag scraper runs to any position, the image of the slag scraper is collected in real time through the side camera, and based on the neural network model of object recognition and tracking, the current position information of the marker in the image is obtained, and the geometry of the marker is obtained. The coordinates P d (x d , y d ) of the center point on the image;
计算扒渣机实际前后伸缩距离D为：Calculate the actual front and rear telescopic distance D of the slag scraper as:
Figure PCTCN2021124915-appb-000019
Figure PCTCN2021124915-appb-000019
进一步地，所述方法还包括：Further, the method also includes:
通过摄像机采集预设时长的扒渣机动作视频；Capture the action video of the slag removal machine with a preset duration through the camera;
从采集的扒渣机动作视频中，连续等间隔提取预设数量的帧图片；对每一帧图片，均对图片中扒渣机的俯仰轴、旋转轴、标志物进行标注，获取每一帧图片中扒渣机运动参量信息，并基于运动参量信息构建数据集，其中，所述运动参量信息包括：扒渣机的旋转角度、俯仰角度、升降高度、前后伸缩距离以及标志物位置；From the collected slag scraper action video, a preset number of frame pictures are continuously extracted at equal intervals; for each frame of picture, the pitch axis, rotation axis, and markers of the slag scraper in the picture are marked to obtain each frame The motion parameter information of the slag scraper in the picture, and a data set is constructed based on the motion parameter information, wherein the motion parameter information includes: the rotation angle, pitch angle, lifting height, front and rear telescopic distance and marker position of the slag scraper;
通过所述数据集训练用于物体识别与跟踪的神经网络模型。A neural network model for object recognition and tracking is trained through the dataset.
进一步地，所述标志物为多个，分别用于在计算旋转角度、俯仰角度、升降高度与前后伸缩距离的时候作为参照的物体，其中，计算扒渣机俯仰角度与旋转角度用的标志物，其几何中心点应在扒渣臂的中心轴线上。Further, there are multiple markers, which are respectively used as reference objects when calculating the rotation angle, pitch angle, lifting height and front and rear telescopic distance, among which, the markers used for calculating the pitch angle and rotation angle of the slag scraper , its geometric center point should be on the central axis of the slag removal arm.
本发明具有以下有益效果：The present invention has the following beneficial effects:
本发明借助安装在扒渣机上方、扒渣机侧面等位置的摄像机，拍摄扒渣机当前的运动位置图像，并对图像进行实时的计算分析，从而得到扒渣机的旋转角度、俯仰角度、升降高度与前后伸缩距离等参数。本发明不需要在扒渣机上安装传感器，从而很好的解决避免了传感器检修困难，以及因扒渣机工作时动作频繁且震动，对传感器的检测精度、稳定性和可靠性造成影响的问题。The present invention uses a camera installed above the slag remover, on the side of the slag remover, etc., to take images of the current movement position of the slag remover, and perform real-time calculation and analysis on the images, so as to obtain the rotation angle, pitch angle, Parameters such as lifting height and front and rear telescopic distance. The present invention does not need to install sensors on the slag remover, thereby well avoiding the difficulty of sensor maintenance and the problems that the detection accuracy, stability and reliability of the sensor are affected by the frequent movements and vibrations of the slag remover during operation.
附图说明Description of drawings
图1为本发明实施例提供的扒渣机四维度运动位置参量的检测和计算方法的流程示意图；Fig. 1 is a schematic flowchart of a method for detecting and calculating a four-dimensional motion position parameter of a slag removal machine provided by an embodiment of the present invention;
图2为本发明实施例提供的在扒渣机上方摄像机安装位置示意图；Fig. 2 is a schematic diagram of the installation position of the camera above the slag removal machine provided by the embodiment of the present invention;
图3为本发明实施例提供的在扒渣机侧面摄像机安装位置示意图；Fig. 3 is a schematic diagram of the installation position of the camera on the side of the slag removal machine provided by the embodiment of the present invention;
图4为本发明实施例提供的一种尾部不做俯仰动作的扒渣机型式示意 图；Fig. 4 is a schematic diagram of a type of slag removal machine with no pitching action at the tail provided by an embodiment of the present invention;
图5为本发明实施例提供的一种尾部既做旋转动作又作俯仰动作的扒渣机型式示意图。Fig. 5 is a schematic diagram of a slag removal machine with a tail that performs both rotating and pitching motions provided by an embodiment of the present invention.
具体实施方式detailed description
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其它实施例，都属于本发明保护的范围。The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
如图1所示，本发明实施例提供的一种扒渣机四维度运动位置参量的检测和计算方法，所述方法包括:As shown in Figure 1, a method for detecting and calculating a four-dimensional motion position parameter of a slag removal machine provided by an embodiment of the present invention, the method includes:
步骤1，在扒渣机上设置作为参照物的标志物，在扒渣机上方和侧方设置摄像机，通过摄像机拍摄扒渣机当前的运动位置图像；Step 1, set a marker as a reference object on the slag remover, set a camera above and on the side of the slag remover, and take a picture of the current movement position of the slag remover through the camera;
步骤2，对图像进行实时的计算分析，得到扒渣机的旋转角度、俯仰角度、升降高度与前后伸缩距离。Step 2: Carry out real-time calculation and analysis on the image to obtain the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper.
本发明借助安装在扒渣机上方、扒渣机侧面等位置的摄像机，拍摄扒渣机当前的运动位置图像，并对图像进行实时的计算分析，从而得到扒渣机的旋转角度、俯仰角度、升降高度与前后伸缩距离等参数。根据现场环境的不同，摄像机可选用可见光摄像机、红外摄像机或远红外摄像机等。除现场安装摄像机外，还需设置一台或数台机器视觉服务器，用于图像的处理、分析和计算结果的发布，另外，机器视觉服务器并不是必须的，本发明所描述的检测和计算方法也可以直接集成到摄像机中，由摄像机自身的软件实现，从而直接实现图像的处理、计算和结果的发布。The present invention uses a camera installed above the slag remover, on the side of the slag remover, etc., to take images of the current movement position of the slag remover, and perform real-time calculation and analysis on the images, so as to obtain the rotation angle, pitch angle, Parameters such as lifting height and front and rear telescopic distance. Depending on the site environment, the camera can be a visible light camera, an infrared camera or a far infrared camera. In addition to installing cameras on site, one or more machine vision servers need to be set up for image processing, analysis, and release of calculation results. In addition, machine vision servers are not necessary. The detection and calculation methods described in the present invention It can also be directly integrated into the camera and implemented by the camera's own software, so as to directly realize image processing, calculation and release of results.
本发明需要在扒渣机本体上选择合适的标志物，用于图像分析。如果扒渣机本体没有合适的标志物，可以制作专门的标志物并安装在扒渣机上，标 志物应方便辨认、能与背景区分开，其中，计算扒渣机俯仰角度与旋转角度用的标志物，其几何中心点应在扒渣臂的中心轴线上。The present invention needs to select appropriate markers on the body of the slag removal machine for image analysis. If there is no suitable marker on the body of the slag remover, a special marker can be made and installed on the slag remover. The marker should be easy to identify and can be distinguished from the background. Among them, the signs used to calculate the pitch angle and rotation angle of the slag remover Object, its geometric center point should be on the central axis of the slag removal arm.
本发明的具体实施可分为两个阶段：The concrete enforcement of the present invention can be divided into two stages:
一.标志物位置的检测与跟踪，具体如下：1. Detection and tracking of the position of markers, as follows:
(1)(如有必要)在扒渣机上安装标志物。(1) (If necessary) Install markers on the slag removal machine.
(2)在扒渣机上方、扒渣机侧面等位置安装摄像机，摄像机的参考位置如附图2-3所示，需要说明的是，摄像机的安装数量和位置会根据现场实际情况调整，附图仅为示意。(2) Install cameras on the top of the slag remover, on the side of the slag remover, etc. The reference positions of the cameras are shown in Figure 2-3. It should be noted that the number and positions of the cameras will be adjusted according to the actual situation on site. The picture is for illustration only.
(3)用摄像机采集预设时长的扒渣机动作视频；(3) Use the video camera to collect the action video of the slag removal machine with a preset duration;
(4)从采集的扒渣机动作视频中，提取预设数量的图片，对提取出的每一幅图片中的旋转角度、俯仰角度、升降高度与前后伸缩距离计算用的标志物，和扒渣机俯仰轴、旋转轴等物体进行标注，构建数据集；(4) From the action video of the slag removal machine collected, extract a preset number of pictures, and calculate the rotation angle, pitch angle, lifting height and front and rear telescopic distance calculation markers in each picture extracted, and pick up Mark the pitch axis, rotation axis and other objects of the slag machine to build a data set;
(5)用第(4)步制作的数据集训练神经网络，得到物体识别与跟踪模型。(5) Use the data set produced in step (4) to train the neural network to obtain the object recognition and tracking model.
需要说明的是，实际应用中，物体识别与跟踪模型并不是必须采用神经网络技术，采用其他技术方案获得物体识别与跟踪模型并取得相同结果也不脱离本专利技术方案的精神和范围。It should be noted that in practical applications, the object recognition and tracking model does not have to use neural network technology, and the object recognition and tracking model obtained by using other technical solutions and achieving the same results does not deviate from the spirit and scope of the technical solution of this patent.
二、扒渣机旋转角度
Figure PCTCN2021124915-appb-000020
俯仰角度θ、升降高度H与前后伸缩距离D的计算。对于尾部不做俯仰动作的扒渣机，旋转角度
Figure PCTCN2021124915-appb-000021
俯仰角度θ均可以由本专利的方法检测和计算；对于尾部既做旋转动作又作俯仰动作的扒渣机，旋转角度
Figure PCTCN2021124915-appb-000022
俯仰角度θ两个参量不是独立的，需要其中一个参量用其它方法获得。两种不同型式扒渣机的示意见附图4和5； 2. Rotation angle of slag removal machine
Figure PCTCN2021124915-appb-000020
Calculation of pitch angle θ, lifting height H and front and rear telescopic distance D. For the slag scraper without pitching action at the tail, the rotation angle
Figure PCTCN2021124915-appb-000021
The pitch angle θ can be detected and calculated by the method of this patent; for the slag scraper whose tail is both rotating and pitching, the rotation angle
Figure PCTCN2021124915-appb-000022
The two parameters of the pitch angle θ are not independent, and one of them needs to be obtained by other methods. See Figures 4 and 5 for illustrations of two different types of slag removal machines;
其中，扒渣机旋转角度
Figure PCTCN2021124915-appb-000023
的计算具体如下： Among them, the rotation angle of the slag remover
Figure PCTCN2021124915-appb-000023
The calculation is as follows:
(1)将扒渣机旋转至0°位置后保持静止，在扒渣机上方摄像机拍摄的图像上沿扒渣机方向(即臂长方向)画一条辅助线M，辅助线M与扒渣机渣臂轴线重合，即穿过扒渣机旋转角度检测用的标志物的几何中心点和扒渣 机旋转轴中心；(1) Rotate the slag remover to the 0° position and keep it still, draw an auxiliary line M along the direction of the slag remover (that is, the arm length direction) on the image captured by the camera above the slag remover, and the auxiliary line M is in line with the slag remover The axis of the slag arm coincides, that is, the geometric center point passing through the marker for detecting the rotation angle of the slag scraper and the center of the rotation axis of the slag scraper;
(2)测量出旋转角度计算用的标志物的几何中心点至扒渣机旋转轴中心的物理距离L sw； (2) Measure the physical distance L sw from the geometric center point of the marker used for calculating the rotation angle to the center of the rotating shaft of the slag picker;
(3)在扒渣机上方摄像机拍摄的图像上，在扒渣机尾部另画一条辅助线N。该辅助线N可以朝任意方向，辅助线N在图像上的长度为l a，对应的实际物理长度为L a；进而可计算出物像比k b； (3) On the image captured by the camera above the slag remover, draw another auxiliary line N at the tail of the slag remover. The auxiliary line N can face any direction, the length of the auxiliary line N on the image is la, and the corresponding actual physical length is L a ; then the object-to-image ratio k b can be calculated;
Figure PCTCN2021124915-appb-000024
Figure PCTCN2021124915-appb-000024
(4)任取辅助线M上的两个点P(x 1,y 1)，P(x 2,y 2)，可得到辅助线M在图像上的直线方程： (4) Take any two points P(x 1 ,y 1 ) and P(x 2 ,y 2 ) on the auxiliary line M to obtain the straight line equation of the auxiliary line M on the image:
A s·x+B s·y+C s＝0； A s x+B s y+C s = 0;
其中，A s＝y 2–y 1；B s＝x 1–x 2；C s＝x 2·y 1–x 1·y 2； Among them, A s =y 2 -y 1 ; B s =x 1 -x 2 ; C s =x 2 ·y 1 -x 1 ·y 2 ;
(5)上述(1)～(4)步为准备工作，当扒渣机运行到任意位置时，扒渣机上方摄像机实时采集扒渣机的图像，图像首先经过阶段一训练完成的物体识别与跟踪模型的计算，获得标志物当前的位置信息，并可进一步计算出标志物的几何中心点在图像上的坐标P s(x 0,y 0)； (5) The above steps (1) to (4) are preparatory work. When the slag scraper moves to any position, the camera above the slag scraper collects the image of the slag scraper in real time. The image first passes through the object recognition and Track the calculation of the model to obtain the current position information of the marker, and further calculate the coordinate P s (x 0 ,y 0 ) of the geometric center point of the marker on the image;
(6)用下式计算P s(x 0,y 0)至辅助线M的实际物理距离L p； (6) Use the following formula to calculate the actual physical distance L p from P s (x 0 , y 0 ) to the auxiliary line M;
Figure PCTCN2021124915-appb-000025
Figure PCTCN2021124915-appb-000025
(7)对于尾部不做俯仰动作的扒渣机，计算扒渣机旋转角度
Figure PCTCN2021124915-appb-000026
为： (7) For the slag scraper without pitching action at the tail, calculate the rotation angle of the slag scraper
Figure PCTCN2021124915-appb-000026
for:
Figure PCTCN2021124915-appb-000027
Figure PCTCN2021124915-appb-000027
(8)对于尾部既做旋转动作又作俯仰动作的扒渣机，通过俯仰角度θ修正扒渣机旋转角度
Figure PCTCN2021124915-appb-000028
得到扒渣机旋转角度
Figure PCTCN2021124915-appb-000029
为： (8) For the slag scraper whose tail is both rotating and pitching, the rotation angle of the slag scraper is corrected by the pitch angle θ
Figure PCTCN2021124915-appb-000028
Get the rotation angle of the slag remover
Figure PCTCN2021124915-appb-000029
for:
Figure PCTCN2021124915-appb-000030
Figure PCTCN2021124915-appb-000030
计算出扒渣机旋转角度
Figure PCTCN2021124915-appb-000031
后，根据P s(x 0,y 0)与辅助线M在图像上的相对位置，修正扒渣机旋转角度
Figure PCTCN2021124915-appb-000032
的正负符号。 Calculate the rotation angle of the slag scraper
Figure PCTCN2021124915-appb-000031
Finally, according to the relative position of P s (x 0 ,y 0 ) and the auxiliary line M on the image, correct the rotation angle of the slag remover
Figure PCTCN2021124915-appb-000032
positive and negative signs.
其中，计算扒渣机俯仰角度θ具体如下：Among them, the calculation of the pitch angle θ of the slag picker is as follows:
(1)将扒渣机俯仰至水平位置(此时俯仰角认为是0°)后保持静止。在扒渣机侧面摄像机拍摄的图像上沿扒渣机方向画一条辅助线J，辅助线J与扒渣机渣臂轴线重合，即穿过扒渣机俯仰角度检测用标志物的几何中心点和扒渣机俯仰轴中心；(1) Pitch the slag scraper to the horizontal position (at this time, the pitch angle is considered to be 0°) and keep it still. Draw an auxiliary line J along the direction of the slag remover on the image captured by the camera on the side of the slag remover. The auxiliary line J coincides with the axis of the slag remover arm, that is, passes through the geometric center point of the marker for pitch angle detection of the slag remover and The center of the pitch axis of the slag scraper;
(2)测量出俯仰角度计算用的标志物的几何中心点，至扒渣机俯仰轴中心的物理距离L tilt； (2) Measure the geometric center point of the marker used for pitch angle calculation, and the physical distance L tilt to the center of the pitch axis of the slag picker;
(3)在扒渣机侧面摄像机拍摄的图像上，在扒渣机上另画一条辅助线K。辅助线K在图像上的长度为l b，对应的实际物理长度为L b；进而可计算出物像比k b； (3) On the image taken by the camera on the side of the slag remover, draw another auxiliary line K on the slag remover. The length of the auxiliary line K on the image is l b , and the corresponding actual physical length is L b ; then the object-to-image ratio k b can be calculated;
Figure PCTCN2021124915-appb-000033
Figure PCTCN2021124915-appb-000033
(4)任取辅助线J上的两个点P(x 3,y 3)，P(x 4,y 4)，可得到辅助线J在图像上的直线方程： (4) Take any two points P(x 3 , y 3 ) and P(x 4 , y 4 ) on the auxiliary line J to obtain the straight line equation of the auxiliary line J on the image:
A t·x+B t·y+C t＝0； A t x+B t y+C t = 0;
其中，A t＝y 4–y 3，B t＝x 3–x 4，C t＝x 4·y 3–x 3·y 4； Among them, A t =y 4 -y 3 , B t =x 3 -x 4 , C t =x 4 ·y 3 -x 3 ·y 4 ;
(5)上述(1)～(4)步为准备工作。当扒渣机运行到任意位置时，实时采集扒渣机侧面摄像机的图像。图像首先经过阶段一训练完成的物体识别与跟踪模型的计算，获得标志物当前的位置信息，并可进一步计算出标志物的几何中心点在图像上的坐标P t(x 0,y 0)； (5) Steps (1) to (4) above are preparatory work. When the slag scraper runs to any position, the image of the side camera of the slag scraper is collected in real time. The image first passes through the calculation of the object recognition and tracking model completed in stage one training to obtain the current position information of the marker, and further calculate the coordinate P t (x 0 ,y 0 ) of the geometric center point of the marker on the image;
(6)用下式计算P t(x 0,y 0)至辅助线J的实际距离L q； (6) Use the following formula to calculate the actual distance L q from P t (x 0 ,y 0 ) to the auxiliary line J;
Figure PCTCN2021124915-appb-000034
Figure PCTCN2021124915-appb-000034
(7)对于尾部不做俯仰动作的扒渣机，计算扒渣机俯仰角度θ为：(7) For the slag scraper without pitching action at the tail, the pitch angle θ of the slag scraper is calculated as:
Figure PCTCN2021124915-appb-000035
Figure PCTCN2021124915-appb-000035
(8)对于尾部既做旋转动作又作俯仰动作的扒渣机，通过旋转角度
Figure PCTCN2021124915-appb-000036
修正扒渣机俯仰角度θ为： (8) For the slag removal machine with both rotating and pitching actions at the tail, the rotation angle
Figure PCTCN2021124915-appb-000036
Correct the pitch angle θ of the slag remover as:
Figure PCTCN2021124915-appb-000037
Figure PCTCN2021124915-appb-000037
计算出扒渣机俯仰角度θ后，根据P t(x 0,y 0)与辅助线J在图像上的相对位置，修正扒渣机旋转角度
Figure PCTCN2021124915-appb-000038
的正负符号。 After calculating the pitch angle θ of the slag scraper, correct the rotation angle of the slag scraper according to the relative position of P t (x 0 , y 0 ) and the auxiliary line J on the image
Figure PCTCN2021124915-appb-000038
positive and negative signs.
其中，计算扒渣机升降高度H具体如下：Among them, the calculation of the lifting height H of the slag removal machine is as follows:
(1)将扒渣机本体降至最低位置后保持静止。在扒渣机侧面摄像机拍摄的图像上，沿扒渣机底座画一条平行于地坪基础的辅助线S；(1) Lower the body of the slag scraper to the lowest position and keep it still. On the image taken by the side camera of the slag remover, draw an auxiliary line S parallel to the floor foundation along the base of the slag remover;
(2)测量出升降高度计算用的标志物的几何中心点至辅助线S的物理距离L rh； (2) Measure the physical distance L rh from the geometric center point of the landmark used for calculating the lifting height to the auxiliary line S;
(3)在扒渣机侧面摄像机拍摄的图像上，在扒渣机上另画一条辅助线T。辅助线T在图像上的长度为l c，对应的实际物理长度为L c；进而可计算出物像比k c； (3) On the image taken by the camera on the side of the slag remover, draw another auxiliary line T on the slag remover. The length of the auxiliary line T on the image is l c , and the corresponding actual physical length is L c ; then the object-to-image ratio k c can be calculated;
Figure PCTCN2021124915-appb-000039
Figure PCTCN2021124915-appb-000039
(4)任取辅助线S上的两个点P(x 5,y 5)，P(x 6,y 6)，可得到辅助线S在图像上的直线方程： (4) Take any two points P(x 5 , y 5 ) and P(x 6 , y 6 ) on the auxiliary line S to obtain the straight line equation of the auxiliary line S on the image:
A h·x+B h·y+C h＝0； A h x+B h y+C h = 0;
其中，A h＝y 6–y 5，B h＝x 5–x 6，C h＝x 6·y 5–x 5·y 6； Among them, A h =y 6 -y 5 , B h =x 5 -x 6 , C h =x 6 ·y 5 -x 5 ·y 6 ;
(5)上述(1)～(4)步为准备工作。当扒渣机运行到任意位置时，侧面摄像机实时采集扒渣机的图像。图像首先经过阶段一训练完成的物体识别与跟踪模型的计算，获得升降高度计算用标志物当前的位置信息，并可进一步计算出该标志物的几何中心点在图像上的坐标P h(x h,y h)； (5) Steps (1) to (4) above are preparatory work. When the slag scraper moves to any position, the side camera collects the images of the slag scraper in real time. The image first passes through the calculation of the object recognition and tracking model completed in stage one training, and obtains the current position information of the marker used for the calculation of the lifting height, and can further calculate the coordinate Ph (x h ,y h );
(6)用下式计算扒渣机实际抬升高度H；(6) Use the following formula to calculate the actual lifting height H of the slag removal machine;
Figure PCTCN2021124915-appb-000040
Figure PCTCN2021124915-appb-000040
其中，扒渣机前后伸缩距离DAmong them, the front and rear telescopic distance D of the slag remover
(1)将扒渣机缩回至最后端位置并旋转至0°位置，俯仰至水平位置， 然后保持静止，在扒渣机侧面摄像机拍摄的图像上，沿扒渣机最后端画一条垂直于地坪基础的辅助线W；(1) Retract the slag scraper to the rearmost position and rotate to 0°, pitch to the horizontal position, and then keep it still. On the image taken by the side camera of the slag scraper, draw a vertical line along the rearmost end of the slag scraper. Auxiliary line W of the floor foundation;
(2)测量出前后伸缩距离计算用的标志物的几何中心点至辅助线W的物理距离L rd； (2) Measure the physical distance L rd from the geometric center point of the marker used for calculating the front and rear telescopic distance to the auxiliary line W;
(3)在扒渣机侧面摄像机拍摄的图像上，在扒渣机上另画一条辅助线V，辅助线V在图像上的长度为l d，对应的实际物理长度为L d；进而可计算出物像比k d； (3) On the image taken by the side camera of the slag remover, another auxiliary line V is drawn on the slag remover. The length of the auxiliary line V on the image is l d , and the corresponding actual physical length is L d ; then it can be calculated Object-to-image ratio k d ;
Figure PCTCN2021124915-appb-000041
Figure PCTCN2021124915-appb-000041
(4)任取辅助线W上的两个点P(x 7,y 7)，P(x 8,y 8)，可得到辅助线W在图像上的直线方程如下： (4) Take any two points P(x 7 , y 7 ) and P(x 8 , y 8 ) on the auxiliary line W, and the straight line equation of the auxiliary line W on the image can be obtained as follows:
A d·x+B d·y+C d＝0； A d x + B d y + C d = 0;
其中，A d＝y 8–y 7；B d＝x 7–x 8；C d＝x 8·y 7–x 7·y 8； Wherein, A d =y 8 -y 7 ; B d =x 7 -x 8 ; C d =x 8 ·y 7 -x 7 ·y 8 ;
(5)上述(1)～(4)步为准备工作。当扒渣机运行到任意位置时，侧面摄像机实时采集扒渣机的图像。图像首先经过阶段一训练完成的物体识别与跟踪模型的计算，获得前后伸缩距离计算用标志物当前的位置信息，并可进一步计算出该标志物的几何中心点在图像上的坐标P d(x d,y d)； (5) Steps (1) to (4) above are preparatory work. When the slag scraper moves to any position, the side camera collects the images of the slag scraper in real time. The image first passes through the calculation of the object recognition and tracking model completed in stage one training, and obtains the current position information of the marker used for calculating the front and rear telescopic distances, and can further calculate the coordinate P d (x d ,y d );
(6)用下式计算扒渣机实际前后伸缩距离D；(6) Use the following formula to calculate the actual front and rear telescopic distance D of the slag scraper;
Figure PCTCN2021124915-appb-000042
Figure PCTCN2021124915-appb-000042
需要说明的是，扒渣机旋转角度、俯仰角度、升降高度与前后伸缩距离的计算方法中的第(3)条，是一种获得图像物像比的方法，采用其它方式得到物像比，应用于后续的计算过程的方法，仍不脱离本专利技术方案的精神和范围。It should be noted that item (3) in the calculation method of the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper is a method to obtain the image-to-image ratio, and other methods are used to obtain the object-to-image ratio. The method applied to the subsequent calculation process still does not deviate from the spirit and scope of the technical solution of this patent.
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发 明的保护范围之内。The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (7)

  1. 一种扒渣机四维度运动位置参量的检测和计算方法，其特征在于，所述方法包括:A detection and calculation method for a four-dimensional movement position parameter of a slag removal machine, characterized in that the method comprises:
    步骤1，在扒渣机上设置作为参照物的标志物，在扒渣机上方和侧方设置摄像机，通过摄像机拍摄扒渣机当前的运动位置图像；Step 1, set a marker as a reference object on the slag remover, set a camera above and on the side of the slag remover, and take a picture of the current movement position of the slag remover through the camera;
    步骤2，对图像进行实时的计算分析，得到扒渣机的旋转角度、俯仰角度、升降高度与前后伸缩距离。Step 2: Carry out real-time calculation and analysis on the image to obtain the rotation angle, pitch angle, lifting height and front and rear telescopic distance of the slag scraper.
  2. 根据权利要求1所述的扒渣机四维度运动位置参量的检测和计算方法，其特征在于，步骤2中，对图像进行实时的计算分析，得到扒渣机的旋转角度具体包括：The method for detecting and calculating the four-dimensional movement position parameters of the slag scraper according to claim 1, characterized in that in step 2, real-time calculation and analysis is performed on the image to obtain the rotation angle of the slag scraper specifically includes:
    步骤2.11，将扒渣机旋转至0°位置后保持静止，并获取该时刻扒渣机上方摄像机拍摄的图像，在扒渣机上方的摄像机拍摄的图像上沿扒渣机方向画一条辅助线M，辅助线M与扒渣机渣臂轴线重合，即穿过扒渣机旋转角度检测用的标志物的几何中心点和扒渣机旋转轴中心；Step 2.11, rotate the slag scraper to the 0° position and keep it still, and obtain the image taken by the camera above the slag scraper at this moment, and draw an auxiliary line M along the direction of the slag scraper on the image taken by the camera above the slag scraper , the auxiliary line M coincides with the axis of the slag arm of the slag remover, that is, passes through the geometric center point of the marker used for detecting the rotation angle of the slag remover and the center of the rotation axis of the slag remover;
    在扒渣机上方的摄像机拍摄的图像上，测量出标志物的几何中心点至扒渣机旋转轴中心的物理距离L sw； Measure the physical distance L sw from the geometric center point of the marker to the center of the rotation axis of the slag remover on the image taken by the camera above the slag remover;
    在扒渣机上方摄像机拍摄的图像上，于扒渣机尾部另画一条辅助线N，令辅助线N在图像上的长度为l a，对应的实际物理长度为L a，进而可计算出物像比k a； On the image taken by the camera above the slag remover, draw another auxiliary line N at the tail of the slag remover, let the length of the auxiliary line N on the image be l a , and the corresponding actual physical length is L a , and then the object can be calculated Like than k a ;
    Figure PCTCN2021124915-appb-100001
    Figure PCTCN2021124915-appb-100001
    任取辅助线M上的两个点P(x 1,y 1)，P(x 2,y 2)，可得到辅助线M在图像上的直线方程； Take any two points P(x 1 , y 1 ) and P(x 2 , y 2 ) on the auxiliary line M to obtain the straight line equation of the auxiliary line M on the image;
    A s·x+B s·y+C s＝0； A s x+B s y+C s = 0;
    其中，A s＝y 2–y 1；B s＝x 1–x 2；C s＝x 2·y 1–x 1·y 2； Among them, A s =y 2 -y 1 ; B s =x 1 -x 2 ; C s =x 2 ·y 1 -x 1 ·y 2 ;
    步骤2.12，当扒渣机运行到任意位置时，通过扒渣机上方摄像机实时采集扒渣机的图像，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P s(x 0,y 0)； Step 2.12, when the slag remover runs to any position, collect the image of the slag remover in real time through the camera above the slag remover, obtain the current position information of the marker in the image, and obtain the coordinates of the geometric center point of the marker on the image P s (x 0 ,y 0 );
    计算P s(x 0,y 0)至辅助线M的实际物理距离L p； Calculate the actual physical distance L p from P s (x 0 ,y 0 ) to the auxiliary line M;
    Figure PCTCN2021124915-appb-100002
    Figure PCTCN2021124915-appb-100002
    对于尾部不做俯仰动作的扒渣机，计算扒渣机旋转角度
    Figure PCTCN2021124915-appb-100003
    为：     For the slag scraper without pitching action at the tail, calculate the rotation angle of the slag scraper
    Figure PCTCN2021124915-appb-100003
    for:
    Figure PCTCN2021124915-appb-100004
    Figure PCTCN2021124915-appb-100004
    对于尾部既做旋转动作又作俯仰动作的扒渣机，通过俯仰角度θ修正扒渣机旋转角度
    Figure PCTCN2021124915-appb-100005
    得到扒渣机旋转角度
    Figure PCTCN2021124915-appb-100006
    为：     For the slag scraper whose tail is both rotating and pitching, the rotation angle of the slag scraper is corrected by the pitch angle θ
    Figure PCTCN2021124915-appb-100005
    Get the rotation angle of the slag remover
    Figure PCTCN2021124915-appb-100006
    for:
    Figure PCTCN2021124915-appb-100007
    Figure PCTCN2021124915-appb-100007
    计算出扒渣机旋转角度
    Figure PCTCN2021124915-appb-100008
    后，根据P s(x 0,y 0)与辅助线M在图像上的相对位置，修正扒渣机旋转角度
    Figure PCTCN2021124915-appb-100009
    的正负符号。     Calculate the rotation angle of the slag scraper
    Figure PCTCN2021124915-appb-100008
    Finally, according to the relative position of P s (x 0 ,y 0 ) and the auxiliary line M on the image, correct the rotation angle of the slag remover
    Figure PCTCN2021124915-appb-100009
    positive and negative signs.
  3. 根据权利要求1所述的扒渣机四维度运动位置参量的检测和计算方法，其特征在于，步骤2中，对图像进行实时的计算分析，得到扒渣机的俯仰角度具体包括：The method for detecting and calculating the four-dimensional movement position parameters of the slag scraper according to claim 1, characterized in that, in step 2, real-time calculation and analysis is performed on the image to obtain the pitch angle of the slag scraper specifically includes:
    步骤2.21，将扒渣机俯仰至水平位置后保持静止，并获取该时刻扒渣机侧面摄像机拍摄的图像，在扒渣机侧面摄像机拍摄的图像上沿扒渣机方向画一条辅助线J，辅助线J与扒渣机渣臂轴线重合，即穿过扒渣机俯仰角度检测用标志物的几何中心点和扒渣机俯仰轴中心；Step 2.21: Pitch the slag scraper to the horizontal position and keep it still, and obtain the image taken by the side camera of the slag scraper at this moment, and draw an auxiliary line J along the direction of the slag scraper on the image taken by the side camera of the slag scraper to assist Line J coincides with the slag arm axis of the slag remover, that is, passes through the geometric center point of the marker for detecting the pitch angle of the slag remover and the center of the slag remover pitch axis;
    在扒渣机侧面摄像机拍摄的图像上，测量出标志物的几何中心点至扒渣机俯仰轴中心的物理距离；Measure the physical distance from the geometric center point of the marker to the center of the pitch axis of the slag remover on the image taken by the side camera of the slag remover;
    在扒渣机侧面摄像机拍摄的图像上，于扒渣机上另画一条辅助线K，令辅助线K在图像上的长度为l b，对应的实际物理长度为L b；进而可计算出物像比k b； On the image taken by the camera on the side of the slag remover, draw another auxiliary line K on the slag remover, so that the length of the auxiliary line K on the image is l b , and the corresponding actual physical length is L b ; then the object image can be calculated than k b ;
    Figure PCTCN2021124915-appb-100010
    Figure PCTCN2021124915-appb-100010
    任取辅助线J上的两个点P(x 3,y 3)，P(x 4,y 4)，可得到辅助线J在图像上的直线方程： Taking any two points P(x 3 ,y 3 ) and P(x 4 ,y 4 ) on the auxiliary line J, the straight line equation of the auxiliary line J on the image can be obtained:
    A t·x+B t·y+C t＝0； A t x+B t y+C t = 0;
    其中，A t＝y 4–y 3，B t＝x 3–x 4，C t＝x 4·y 3–x 3·y 4； Among them, A t =y 4 -y 3 , B t =x 3 -x 4 , C t =x 4 ·y 3 -x 3 ·y 4 ;
    步骤2.22，当扒渣机运行到任意位置时，通过扒渣机侧方摄像机实时采集扒渣机的图像，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P t(x 0,y 0)； Step 2.22, when the slag remover runs to any position, collect the image of the slag remover in real time through the camera on the side of the slag remover, obtain the current position information of the marker in the image, and obtain the position of the geometric center point of the marker on the image Coordinate P t (x 0 ,y 0 );
    计算P t(x 0,y 0)至辅助线J的实际距离L q； Calculate the actual distance L q from P t (x 0 ,y 0 ) to the auxiliary line J;
    Figure PCTCN2021124915-appb-100011
    Figure PCTCN2021124915-appb-100011
    对于尾部不做俯仰动作的扒渣机，计算扒渣机俯仰角度θ为：For the slag remover without pitching action at the tail, the pitch angle θ of the slag remover is calculated as:
    Figure PCTCN2021124915-appb-100012
    Figure PCTCN2021124915-appb-100012
    对于尾部既做旋转动作又作俯仰动作的扒渣机，通过旋转角度
    Figure PCTCN2021124915-appb-100013
    修正扒渣机俯仰角度θ为：     For the slag scraper whose tail is both rotating and pitching, the rotation angle
    Figure PCTCN2021124915-appb-100013
    Correct the pitch angle θ of the slag remover as:
    Figure PCTCN2021124915-appb-100014
    Figure PCTCN2021124915-appb-100014
    计算出扒渣机俯仰角度θ后，根据P t(x 0,y 0)与辅助线J在图像上的相对位置，修正扒渣机旋转角度
    Figure PCTCN2021124915-appb-100015
    的正负符号。     After calculating the pitch angle θ of the slag scraper, correct the rotation angle of the slag scraper according to the relative position of P t (x 0 , y 0 ) and the auxiliary line J on the image
    Figure PCTCN2021124915-appb-100015
    positive and negative signs.
  4. 根据权利要求1所述的扒渣机四维度运动位置参量的检测和计算方法，其特征在于，对图像进行实时的计算分析，得到扒渣机的升降高度具体包括：The method for detecting and calculating the four-dimensional movement position parameters of the slag scraper according to claim 1, wherein the real-time calculation and analysis of the image to obtain the lifting height of the slag scraper specifically includes:
    步骤2.31，将扒渣机本体降至最低位置后保持静止，并获取该时刻扒渣机侧面摄像机拍摄的图像，在扒渣机侧面摄像机拍摄的图像上，沿扒渣机底座画一条平行于地坪基础的辅助线S；Step 2.31, lower the body of the slag scraper to the lowest position and keep it still, and obtain the image taken by the camera on the side of the slag scraper at this moment. On the image taken by the side camera of the slag scraper, draw a line parallel to the ground Auxiliary line S of ping foundation;
    在扒渣机侧面摄像机拍摄的图像上，测量出标志物的几何中心点至辅助 线S的物理距离；Measure the physical distance from the geometric center point of the marker to the auxiliary line S on the image taken by the side camera of the slag removal machine;
    在扒渣机侧面摄像机拍摄的图像上，于扒渣机上另画一条辅助线T，令辅助线T在图像上的长度为l c，对应的实际物理长度为L c；进而可计算出物像比k c； On the image taken by the camera on the side of the slag remover, draw another auxiliary line T on the slag remover, so that the length of the auxiliary line T on the image is l c , and the corresponding actual physical length is L c ; then the object image can be calculated than k c ;
    Figure PCTCN2021124915-appb-100016
    Figure PCTCN2021124915-appb-100016
    任取辅助线S上的两个点P(x 5,y 5)，P(x 6,y 6)，可得到辅助线S在图像上的直线方程； Take any two points P(x 5 , y 5 ) and P(x 6 , y 6 ) on the auxiliary line S to obtain the straight line equation of the auxiliary line S on the image;
    A h·x+B h·y+C h＝0； A h x+B h y+C h = 0;
    其中，A h＝y 6–y 5，B h＝x 5–x 6，C h＝x 6·y 5–x 5·y 6； Among them, A h =y 6 -y 5 , B h =x 5 -x 6 , C h =x 6 ·y 5 -x 5 ·y 6 ;
    步骤2.32，当扒渣机运行到任意位置时，通过侧面摄像机实时采集扒渣机的图像，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P h(x h,y h)； Step 2.32, when the slag removal machine runs to any position, collect the image of the slag removal machine in real time through the side camera, obtain the current position information of the marker in the image, and obtain the coordinate Ph ( x h ,y h );
    计算扒渣机实际抬升高度H；Calculate the actual lifting height H of the slag remover;
    Figure PCTCN2021124915-appb-100017
    Figure PCTCN2021124915-appb-100017
  5. 根据权利要求1所述的扒渣机四维度运动位置参量的检测和计算方法，其特征在于，对图像进行实时的计算分析，得到扒渣机的前后伸缩距离具体包括：The method for detecting and calculating the four-dimensional movement position parameters of the slag scraper according to claim 1, characterized in that the image is calculated and analyzed in real time to obtain the front and rear telescopic distances of the slag scraper specifically include:
    步骤2.41，将扒渣机缩回至最后端位置，并将扒渣机旋转至0°位置，俯仰至水平位置，然后保持静止，并获取该时刻扒渣机侧面摄像机拍摄的图像，在扒渣机侧面摄像机拍摄图像上，在扒渣机最后端画一条垂直于地坪基础的辅助线W；Step 2.41, retract the slag scraper to the rearmost position, rotate the slag scraper to the 0° position, pitch to the horizontal position, and then keep still, and obtain the image taken by the side camera of the slag scraper at this moment, and then the slag scraper On the image taken by the camera on the side of the machine, draw an auxiliary line W perpendicular to the foundation of the floor at the rear end of the slag picker;
    在扒渣机侧面摄像机拍摄图像上，测量出标志物的几何中心点至辅助线W的物理距离L rd； Measure the physical distance L rd from the geometric center point of the marker to the auxiliary line W on the image taken by the side camera of the slag removal machine;
    在扒渣机侧面摄像机拍摄的图像上，于扒渣机上另画一条辅助线V，辅 助线V在图像上的长度为l d，对应的实际物理长度为L d；进而可计算出物像比k d： On the image taken by the side camera of the slag remover, draw another auxiliary line V on the slag remover. The length of the auxiliary line V on the image is l d , and the corresponding actual physical length is L d ; then the object-to-image ratio can be calculated k d :
    Figure PCTCN2021124915-appb-100018
    Figure PCTCN2021124915-appb-100018
    任取辅助线W上的两个点P(x 7,y 7)，P(x 8,y 8)，可得到辅助线W在图像上的直线方程如下： Taking any two points P(x 7 , y 7 ) and P(x 8 , y 8 ) on the auxiliary line W, the linear equation of the auxiliary line W on the image can be obtained as follows:
    A d·x+B d·y+C d＝0； A d x + B d y + C d = 0;
    其中，A d＝y 8–y 7；B d＝x 7–x 8；C d＝x 8·y 7–x 7·y 8； Wherein, A d =y 8 -y 7 ; B d =x 7 -x 8 ; C d =x 8 ·y 7 -x 7 ·y 8 ;
    步骤2.42，当扒渣机运行到任意位置时，通过侧面摄像机实时采集扒渣机的图像，获得图像中标志物当前的位置信息，并获取标志物的几何中心点在图像上的坐标P d(x d,y d)； Step 2.42, when the slag removal machine runs to any position, collect the image of the slag removal machine in real time through the side camera, obtain the current position information of the marker in the image, and obtain the coordinate P d ( x d , y d );
    计算扒渣机实际前后伸缩距离D为：Calculate the actual front and rear telescopic distance D of the slag scraper as:
    Figure PCTCN2021124915-appb-100019
    Figure PCTCN2021124915-appb-100019
  6. 根据权利要求1所述的扒渣机四维度运动位置参量的检测和计算方法，其特征在于，所述方法还包括：The detection and calculation method of the four-dimensional motion position parameter of the slag removal machine according to claim 1, characterized in that the method further comprises:
    通过摄像机采集预设时长的扒渣机动作视频；Capture the action video of the slag removal machine with a preset duration through the camera;
    从采集的扒渣机动作视频中，连续等间隔提取预设数量的帧图片；对每一帧图片，均对图片中扒渣机的俯仰轴、旋转轴、标志物进行标注，获取每一帧图片中扒渣机运动参量信息，并基于运动参量信息构建数据集，其中，所述运动参量信息包括：扒渣机的旋转角度、俯仰角度、升降高度、前后伸缩距离以及标志物位置；From the collected slag scraper action video, a preset number of frame pictures are continuously extracted at equal intervals; for each frame of picture, the pitch axis, rotation axis, and markers of the slag scraper in the picture are marked to obtain each frame The motion parameter information of the slag scraper in the picture, and a data set is constructed based on the motion parameter information, wherein the motion parameter information includes: the rotation angle, pitch angle, lifting height, front and rear telescopic distance and marker position of the slag scraper;
    通过所述数据集训练用于物体识别与跟踪的神经网络模型。A neural network model for object recognition and tracking is trained through the dataset.
  7. 根据权利要求1所述的扒渣机四维度运动位置参量的检测和计算方法，其特征在于，所述标志物为多个，分别用于在计算旋转角度、俯仰角度、升降高度与前后伸缩距离的时候作为参照的物体，其中，计算扒渣机俯仰角度与旋转角度用的标志物，其几何中心点应在扒渣臂的中心轴线上。The detection and calculation method for the four-dimensional movement position parameters of the slag scraper according to claim 1, wherein there are multiple markers, which are used to calculate the rotation angle, pitch angle, lifting height and front and rear telescopic distance respectively. When used as a reference object, the geometric center point of the marker used to calculate the pitch angle and rotation angle of the slag scraper should be on the central axis of the slag scraper arm.
PCT/CN2021/124915 2021-07-01 2021-10-20 Method for measuring and calculating four dimensions of motion position parameters of mucking loader WO2023273023A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110741448.9 2021-07-01
CN202110741448.9A CN113532272B (en) 2021-07-01 2021-07-01 Detection and calculation method for four-dimensional motion position parameters of slag raking machine

Publications (1)

Publication Number Publication Date
WO2023273023A1 true WO2023273023A1 (en) 2023-01-05

Family

ID=78126434

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/124915 WO2023273023A1 (en) 2021-07-01 2021-10-20 Method for measuring and calculating four dimensions of motion position parameters of mucking loader

Country Status (2)

Country Link
CN (1) CN113532272B (en)
WO (1) WO2023273023A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1078311A (en) * 1996-09-04 1998-03-24 Matsushita Electric Ind Co Ltd Method and device for measuring rotating angle of motor
CN103323229A (en) * 2013-07-08 2013-09-25 重庆工业职业技术学院 Rotation axis error detection method of five-axis numerical control machine tool based on machine vision
CN112093075A (en) * 2020-08-06 2020-12-18 北京航空航天大学 Synchronous measurement platform for average lift force and wing motion of flapping wing micro aircraft
CN112432612A (en) * 2020-10-22 2021-03-02 中国计量科学研究院 High-precision micro rotation angle measuring method based on monocular vision

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101242847B1 (en) * 2009-12-03 2013-03-12 주식회사 포스코 Method for correcting long nozzle distortion of continuous casting process
JP5961472B2 (en) * 2012-07-27 2016-08-02 日立建機株式会社 Work machine ambient monitoring device
CN207632835U (en) * 2017-12-26 2018-07-20 河北工程大学 A kind of novel jet mucking machine
CN111522294A (en) * 2020-05-08 2020-08-11 衡阳镭目科技有限责任公司 Automatic control system and method of slag raking machine

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1078311A (en) * 1996-09-04 1998-03-24 Matsushita Electric Ind Co Ltd Method and device for measuring rotating angle of motor
CN103323229A (en) * 2013-07-08 2013-09-25 重庆工业职业技术学院 Rotation axis error detection method of five-axis numerical control machine tool based on machine vision
CN112093075A (en) * 2020-08-06 2020-12-18 北京航空航天大学 Synchronous measurement platform for average lift force and wing motion of flapping wing micro aircraft
CN112432612A (en) * 2020-10-22 2021-03-02 中国计量科学研究院 High-precision micro rotation angle measuring method based on monocular vision

Also Published As

Publication number Publication date
CN113532272B (en) 2022-08-05
CN113532272A (en) 2021-10-22

Similar Documents

Publication Publication Date Title
JP5227139B2 (en) Construction machinery
CN111272366B (en) Bridge displacement high-precision measurement method based on multi-sensor data fusion
CN106989683B (en) A kind of shield tail clearance of shield machine vision measuring method
JP5075672B2 (en) Object detection apparatus and method
CN206420439U (en) Contact net geometric parameter detection means based on machine vision
CN108564628B (en) Cutting head vision positioning and orientation system for heading machine automation
CN106996748A (en) Wheel diameter measuring method based on binocular vision
CN109060820B (en) Tunnel disease detection method and tunnel disease detection device based on laser detection
CN112050732B (en) Method and system for automatically detecting spatial pose of cantilever type heading machine
CN110991360B (en) Robot inspection point position intelligent configuration method based on visual algorithm
CN111396133B (en) Digital camera technology-based roadway full-section deformation real-time monitoring device and method
CN111967360A (en) Target vehicle attitude detection method based on wheels
CN107316288B (en) A kind of boom-type roadheader cutterhead pose vision measuring method
CN107316330A (en) Underground hydraulic support frame group pose and verticality measuring method based on many image sequences
CN115063579B (en) Train positioning pin looseness detection method based on two-dimensional image and three-dimensional point cloud projection
CN114964007A (en) Visual measurement and surface defect detection method for weld size
CN115717867A (en) Bridge deformation measurement method based on airborne double cameras and target tracking
CN113848209B (en) Dam crack detection method based on unmanned aerial vehicle and laser ranging
WO2023273023A1 (en) Method for measuring and calculating four dimensions of motion position parameters of mucking loader
CN115962783B (en) Positioning method of cutting head of heading machine and heading machine
CN109879170B (en) Crane jib sidewise bending displacement real-time detection system
CN110726402A (en) Laser point vision guiding method of non-orthogonal shafting laser total station
CN107316302A (en) Driving automatic positioning equipment and its corresponding localization method based on binocular vision
CN114913131A (en) Vertical absolute and relative vertical deformation detection method for ballastless track structure
CN207115499U (en) Driving automatic positioning equipment based on binocular vision

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21947950

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2024101854

Country of ref document: RU

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21947950

Country of ref document: EP

Kind code of ref document: A1