WO2023207257A1 - Tailing dam surface deformation patrol method based on rail-mounted rail robot - Google Patents

Tailing dam surface deformation patrol method based on rail-mounted rail robot Download PDF

Info

Publication number
WO2023207257A1
WO2023207257A1 PCT/CN2023/075438 CN2023075438W WO2023207257A1 WO 2023207257 A1 WO2023207257 A1 WO 2023207257A1 CN 2023075438 W CN2023075438 W CN 2023075438W WO 2023207257 A1 WO2023207257 A1 WO 2023207257A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
dimensional
point cloud
rail
tailings dam
Prior art date
Application number
PCT/CN2023/075438
Other languages
French (fr)
Chinese (zh)
Inventor
聂闻
徐修平
朱天强
代永新
代碧波
周玉新
原粲茗
吴小刚
Original Assignee
中钢集团马鞍山矿山研究总院股份有限公司
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 中钢集团马鞍山矿山研究总院股份有限公司 filed Critical 中钢集团马鞍山矿山研究总院股份有限公司
Publication of WO2023207257A1 publication Critical patent/WO2023207257A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/251—Fusion techniques of input or preprocessed data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration by the use of local operators
    • G06T5/70—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G06T7/55—Depth or shape recovery from multiple images
    • G06T7/593—Depth or shape recovery from multiple images from stereo images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • 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
    • G06T7/85—Stereo 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G06T2207/10012—Stereo images
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/30—Assessment of water resources

Definitions

  • the invention belongs to the field of deformation inspection, and in particular relates to a tailings dam surface deformation inspection method based on a rail-mounted orbital robot.
  • the present invention provides a tailings dam surface deformation inspection method based on a rail-mounted orbital robot, which includes:
  • the three-dimensional monitoring model is processed and analyzed to obtain the deformation condition of the dam body.
  • the process of collecting data on the surface of the tailings dam body includes collecting data on the tailings dam body based on an orbital slide robot, and obtaining image data through binocular camera collection of the orbital slide robot. , three-dimensional point cloud data is obtained through lidar collection of the orbital slide robot.
  • the rail slide robot builds a data collection layer and is monitored by a host computer. layer, communication layer, and control layer acquisition; the data collection layer is used to collect the image data and the three-dimensional point cloud data; the host computer monitoring layer is used to view the position information of the track slide robot and the operation of the equipment in real time status; the communication layer is used for data transmission with the host computer; the control layer is remotely controlled by the host computer and used to control the motor operation of the equipment.
  • the process of preprocessing the data includes denoising the image data based on super pixels and intelligent optimization methods, performing data fusion processing based on the denoised image data and the three-dimensional point cloud data, and obtaining The three-dimensional monitoring data.
  • the process of denoising the image data based on super pixels and intelligent optimization methods includes using a mask to scan each pixel in the image through Gaussian filtering, and replacing it with the weighted average gray value of the pixels in the mask neighborhood.
  • the value of the pixel in the center of the mask is used to obtain the denoised image data.
  • the data fusion process includes processing the image data and the three-dimensional point cloud data based on the LVI-SAM algorithm to generate a two-phase calibrated three-dimensional point cloud at each time of the tailings dam body. .
  • modeling is performed based on the three-dimensional monitoring data
  • the process of obtaining the three-dimensional monitoring model includes: obtaining monitoring images based on the three-dimensional monitoring data, preprocessing the monitoring images using the Sobel operator; and preprocessing the preprocessed images.
  • Perform cost calculation perform dynamic programming on the image after cost calculation, and obtain a disparity map through stereo calibration and stereo matching; obtain a depth map based on the disparity map; draw a point cloud map based on the depth map; based on the point cloud map, Construct the three-dimensional monitoring model.
  • the process of processing and analyzing the three-dimensional monitoring model includes importing the three-dimensional point clouds at different times into the three-dimensional monitoring model to align the point clouds, and calculate and obtain the former after the alignment.
  • rear point cloud distance visualize the difference in distance between the front and rear point clouds through a preset threshold to obtain a visualization result; calculate the area and volume of the three-dimensional monitoring model, and obtain the dam based on the area and volume and the visualization result Body shape deformation.
  • the process of importing three-dimensional point clouds at different times into the three-dimensional monitoring model for point cloud alignment includes calculating the chamfer distance between the three-dimensional point clouds, and comparing and aligning the generated point cloud and the original point cloud based on the chamfer distance.
  • the comparison and alignment of the generated point cloud and the original point cloud based on the chamfer distance is completed through the ICP point cloud registration algorithm;
  • the process of the ICP point cloud registration algorithm includes preprocessing the three-dimensional point cloud to obtain the original transformation; matching the original transformation to obtain the closest point; adjusting the weight of the corresponding point pairs through weighting, and eliminating incorrect points.
  • Reasonable corresponding point pairs by calculating the loss, the minimized loss is obtained; based on the minimized loss, the optimal transformation is obtained.
  • the present invention provides a tailings dam surface deformation inspection method based on a rail-mounted orbital robot.
  • the main body of monitoring is the entire tailings dam body.
  • the constructed track environment is outdoors.
  • the radar is used to scan the entire dam body to generate a three-dimensional point cloud, and then capture the deformation and attributes of the tailings dam surface, including the specific volume, position and other information of the deformation, to achieve three-dimensional monitoring and control of the tailings dam surface deformation data. Refactor.
  • Figure 1 is a method flow chart according to an embodiment of the present invention
  • Figure 2 is a device model diagram of an embodiment of the present invention.
  • Figure 3 is a schematic diagram of the track architecture according to the embodiment of the present invention.
  • Figure 4 is an example diagram of a device information capture area according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of the coincidence rate of photos taken at the same horizontal position at different levels according to the embodiment of the present invention.
  • the present invention provides a tailings dam surface deformation inspection method based on a rail-mounted orbital robot, which includes:
  • a stepped operating track of the rail-mounted robot is constructed around the monitoring target object.
  • the track is built-in
  • the energy supply circuit allows the rail-mounted robot to operate normally. Through the stepped installation, the area of information collected by the equipment can be improved.
  • the collection equipment normally does not change the working orientation, but for some special needs, the working orientation can be changed by changing the orbital running direction (as shown in Figure 4).
  • the host computer monitoring layer is used to view the specific position of the robot and the operating status of related equipment in real time, and feedback to the host computer in real time through the communication layer;
  • the communication layer acts as a device to communicate with the host computer.
  • the control layer is remotely controlled by the host computer and used to control the motor operation of the equipment;
  • the data collection layer collects image data through cameras and three-dimensional point cloud data through lidar.
  • users can flexibly control the running speed of the track slide robot according to the preset collection area.
  • a double-circular track inspection robot for tailings dams driven by high-speed servo motors to assist in the initial collection of monitoring data.
  • a fill light is installed on the track slide robot to compensate for the failure of the night vision mode.
  • Use Gaussian filtering for denoising The specific operation is: use a template (or convolution, mask) to scan each pixel in the image, and use the weighted average gray value of the pixels in the neighborhood determined by the template to replace the central pixel of the template. point value.
  • the image data is preprocessed through the SGBM algorithm, a three-dimensional monitoring model of the tailings pond is established, and the three-dimensional monitoring data is stored in the database.
  • LiDAR fuses the collected radar data with the image data collected by the binocular camera.
  • the specific method is to use the LVI-SAM algorithm to process the image data and radar data to generate a two-phase calibrated entire tailings dam. 3D point cloud data of the volume.
  • the specific construction method of the three-dimensional monitoring model of the tailings pond is to use the Sobel operator to preprocess the image; perform cost calculation on the preprocessed image; perform dynamic programming on the cost-calculated image, and obtain it through stereo calibration and stereo matching.
  • Disparity map as described The disparity map is used to obtain a depth map; a point cloud map is drawn based on the depth map; and the three-dimensional monitoring model is constructed based on the point cloud map.
  • the process of point cloud alignment includes calculating the chamfer distance between three-dimensional point clouds, that is, calculating the average shortest point distance between the generated point cloud and the groundtruth point cloud.
  • the difference with the origin cloud for comparison is determined by the size of the chamfer distance.
  • the chamfer distance compares the generated point cloud with the original point cloud and the alignment is completed through point cloud registration.
  • Point Cloud Registration refers to inputting two point clouds (source) and (target), and outputting a transformation to make the degree of overlap between (source) and (target) as high as possible.
  • This invention only considers rigid transformation, that is, the transformation only includes rotation and translation.
  • Point cloud registration can be divided into two steps: Coarse Registration and Fine Registration.
  • Coarse registration refers to a relatively rough registration when the transformation between two point clouds is completely unknown. The main purpose is to provide a better initial transformation value for fine registration; the fine registration criterion is to give an initial value. Transform and further optimize to obtain a more accurate transformation.
  • the most widely used point cloud precision registration algorithm at present is the iterative closest point algorithm (ICP) and various variants of the ICP algorithm.
  • ICP iterative closest point algorithm
  • This invention is based on Point cloud alignment is performed using the ICP algorithm.
  • p s and p t are the corresponding points in the source point cloud and target point cloud.
  • the general algorithm flow of ICP is: preprocess the three-dimensional point cloud to obtain the original transformation; match the original transformation to obtain the closest point; adjust the weight of the corresponding point pairs through weighting, and eliminate unreasonable corresponding point pairs; By calculating the loss, the minimized loss is obtained; based on the minimized loss, the optimal transformation is obtained; the above steps are iterated until convergence.

Abstract

Disclosed in the present invention is a tailing dam surface deformation patrol method based on a rail-mounted rail robot, comprising: collecting data of the surface of a tailing pond dam body, and preprocessing the data to obtain three-dimensional monitoring data; performing modeling on the basis of the three-dimensional monitoring data to obtain a three-dimensional monitoring model; and processing and analyzing the three-dimensional monitoring model to obtain a dam body deformation condition. The monitored main body in the present invention is the dam body of the whole tailing dam, a constructed rail environment is outdoor, and the whole dam body is scanned by mounting a binocular camera and a laser radar on the robot, so as to generate three-dimensional point cloud and capture the deformation of the surface of the tailing dam and the attributes thereof, comprising the specific size, position and other information of the deformation, thereby realizing three-dimensional monitoring and reconstruction of the deformation data of the surface of the tailing dam body.

Description

基于挂轨轨道机器人的尾矿坝表面形变巡检方法Tailings dam surface deformation inspection method based on rail-mounted orbital robot 技术领域Technical field
本发明属于形变巡检领域，尤其涉及一种基于挂轨轨道机器人的尾矿坝表面形变巡检方法。The invention belongs to the field of deformation inspection, and in particular relates to a tailings dam surface deformation inspection method based on a rail-mounted orbital robot.
背景技术Background technique
目前市面上用的比较多的轨道挂轨机器人的应用场景主要是在一些车间以及一些封闭环境内，通过摄像头进行一个图像的采集，通过对采集的二维图像的处理，对异常情况进行监控和监测。现有技术的局限性在于只能采集到基础的二维图像数据，在图像的数据的处理上也只是简单的在二维平面上的识别，并不能观测出异常处具体的形变变化。Currently, the application scenarios of many rail-mounted robots on the market are mainly in some workshops and some closed environments. They collect an image through a camera and process the collected two-dimensional image to monitor and detect abnormal situations. monitor. The limitation of the existing technology is that it can only collect basic two-dimensional image data, and the image data processing is only simple recognition on the two-dimensional plane, and it cannot observe the specific deformation changes of the abnormality.
发明内容Contents of the invention
为解决上述问题，本发明提供一种基于挂轨轨道机器人的尾矿坝表面形变巡检方法，包括：In order to solve the above problems, the present invention provides a tailings dam surface deformation inspection method based on a rail-mounted orbital robot, which includes:
采集尾矿库坝体表面的数据，对所述数据进行预处理，获得三维监测数据；Collect data on the surface of the tailings reservoir dam, preprocess the data, and obtain three-dimensional monitoring data;
基于所述三维监测数据进行建模，获得三维监测模型；Perform modeling based on the three-dimensional monitoring data to obtain a three-dimensional monitoring model;
对所述三维监测模型进行处理分析，获得坝体形变情况。The three-dimensional monitoring model is processed and analyzed to obtain the deformation condition of the dam body.
优选的，采集所述尾矿库坝体表面的数据的过程包括，基于轨道滑轨机器人对所述尾矿库坝体进行数据采集，通过所述轨道滑轨机器人的双目相机采集获得图像数据，通过所述轨道滑轨机器人的激光雷达采集获得三维点云数据。Preferably, the process of collecting data on the surface of the tailings dam body includes collecting data on the tailings dam body based on an orbital slide robot, and obtaining image data through binocular camera collection of the orbital slide robot. , three-dimensional point cloud data is obtained through lidar collection of the orbital slide robot.
优选的，所述轨道滑轨机器人通过搭建数据采集层，上位机监控 层，通讯层，控制层获得；所述数据采集层用于采集所述图像数据和所述三维点云数据；所述上位机监控层用于实时查看轨道滑轨机器人的位置信息和设备的运行状态；所述通讯层用于与上位机进行数据传输；所述控制层由上位机远程控制，用于控制设备的电机运转。Preferably, the rail slide robot builds a data collection layer and is monitored by a host computer. layer, communication layer, and control layer acquisition; the data collection layer is used to collect the image data and the three-dimensional point cloud data; the host computer monitoring layer is used to view the position information of the track slide robot and the operation of the equipment in real time status; the communication layer is used for data transmission with the host computer; the control layer is remotely controlled by the host computer and used to control the motor operation of the equipment.
优选的，对所述数据进行预处理的过程包括，基于超像素和智能优化方法对所述图像数据进行去噪，基于去噪后的图像数据和所述三维点云数据进行数据融合处理，获得所述三维监测数据。Preferably, the process of preprocessing the data includes denoising the image data based on super pixels and intelligent optimization methods, performing data fusion processing based on the denoised image data and the three-dimensional point cloud data, and obtaining The three-dimensional monitoring data.
优选的，基于超像素和智能优化方法对所述图像数据进行去噪的过程包括，通过高斯滤波，采用掩模扫描图像中的每一个像素，用掩膜邻域内像素的加权平均灰度值替代掩膜中心像素点的值，获得去噪图像数据。Preferably, the process of denoising the image data based on super pixels and intelligent optimization methods includes using a mask to scan each pixel in the image through Gaussian filtering, and replacing it with the weighted average gray value of the pixels in the mask neighborhood. The value of the pixel in the center of the mask is used to obtain the denoised image data.
优选的，所述数据融合处理的过程包括，基于LVI-SAM算法对所述图像数据和所述三维点云数据进行处理，生成经过两相校准的尾矿库坝体的各个时刻的三维点云。Preferably, the data fusion process includes processing the image data and the three-dimensional point cloud data based on the LVI-SAM algorithm to generate a two-phase calibrated three-dimensional point cloud at each time of the tailings dam body. .
优选的，基于所述三维监测数据进行建模，获得三维监测模型的过程包括，基于所述三维监测数据获得监测图像，使用Sobel算子对所述监测图像进行预处理；对预处理后的图像进行代价计算；对进行代价计算后的图像进行动态规划，通过立体标定、立体匹配获得视差图；基于所述视差图，得到深度图；基于所述深度图绘制点云图；基于所述点云图，构建所述三维监测模型。Preferably, modeling is performed based on the three-dimensional monitoring data, and the process of obtaining the three-dimensional monitoring model includes: obtaining monitoring images based on the three-dimensional monitoring data, preprocessing the monitoring images using the Sobel operator; and preprocessing the preprocessed images. Perform cost calculation; perform dynamic programming on the image after cost calculation, and obtain a disparity map through stereo calibration and stereo matching; obtain a depth map based on the disparity map; draw a point cloud map based on the depth map; based on the point cloud map, Construct the three-dimensional monitoring model.
优选的，对所述三维监测模型进行处理分析的过程包括，将不同时刻的三维点云导入三维监测模型进行点云对齐，对齐后计算获得前 后点云距离；通过预设阈值进行所述前后点云距离差异的可视化，获得可视化结果；计算所述三维监测模型的面积和体积，基于所述面积和体积以及所述可视化结果获得所述坝体形变情况。Preferably, the process of processing and analyzing the three-dimensional monitoring model includes importing the three-dimensional point clouds at different times into the three-dimensional monitoring model to align the point clouds, and calculate and obtain the former after the alignment. rear point cloud distance; visualize the difference in distance between the front and rear point clouds through a preset threshold to obtain a visualization result; calculate the area and volume of the three-dimensional monitoring model, and obtain the dam based on the area and volume and the visualization result Body shape deformation.
优选的，将不同时刻的三维点云导入三维监测模型进行点云对齐的过程包括，计算三维点云间的倒角距离，基于所述倒角距离对生成点云与原点云进行比较对齐。Preferably, the process of importing three-dimensional point clouds at different times into the three-dimensional monitoring model for point cloud alignment includes calculating the chamfer distance between the three-dimensional point clouds, and comparing and aligning the generated point cloud and the original point cloud based on the chamfer distance.
优选的，基于所述倒角距离对生成点云与原点云进行比较对齐通过ICP点云配准算法完成；Preferably, the comparison and alignment of the generated point cloud and the original point cloud based on the chamfer distance is completed through the ICP point cloud registration algorithm;
所述ICP点云配准算法的过程包括，对所述三维点云进行预处理，获得原始变换；对所述原始变换进行匹配，获得最近点；通过加权，调整对应点对的权重，剔除不合理的对应点对；通过计算loss，获得最小化loss；基于最小化loss，获得最优变换。The process of the ICP point cloud registration algorithm includes preprocessing the three-dimensional point cloud to obtain the original transformation; matching the original transformation to obtain the closest point; adjusting the weight of the corresponding point pairs through weighting, and eliminating incorrect points. Reasonable corresponding point pairs; by calculating the loss, the minimized loss is obtained; based on the minimized loss, the optimal transformation is obtained.
本发明公开了以下技术效果：The invention discloses the following technical effects:
对比目前市面上采用的很多针对很多户外大型地质图像信息都是采用无人机的方法进行数据采集的，单次的操作流程复杂，且耗时耗力，每次都需要对无人机进行操纵。本申请构建轨道机器人轨道的方法具有一劳永逸的优越性，一次设置，即可长期自动运行。Compared with many large-scale outdoor geological image information currently on the market, drones are used to collect data. The single operation process is complex, time-consuming and labor-intensive, and the drone needs to be controlled every time. . The method of constructing an orbital robot track in this application has the advantage of being once and for all. Once set, it can run automatically for a long time.
本发明提供的一种基于挂轨轨道机器人的尾矿坝表面形变巡检方法，监测的主体是整个尾矿坝的坝体，构建的轨道环境在室外，通过在机器人上搭载双目相机与激光雷达来进行对整个坝体的扫描从而生成三维点云，进而捕捉尾矿坝表面的形变及其属性，包括形变的具体体积、位置等信息，实现对尾矿坝体表面形变数据的三维监测与 重构。The present invention provides a tailings dam surface deformation inspection method based on a rail-mounted orbital robot. The main body of monitoring is the entire tailings dam body. The constructed track environment is outdoors. By equipping the robot with a binocular camera and a laser The radar is used to scan the entire dam body to generate a three-dimensional point cloud, and then capture the deformation and attributes of the tailings dam surface, including the specific volume, position and other information of the deformation, to achieve three-dimensional monitoring and control of the tailings dam surface deformation data. Refactor.
附图说明Description of the drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some of the drawings of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without exerting creative efforts.
图1为本发明实施例的方法流程图；Figure 1 is a method flow chart according to an embodiment of the present invention;
图2为本发明实施例的设备模型图；Figure 2 is a device model diagram of an embodiment of the present invention;
图3为本发明实施例的轨道架构示意图；Figure 3 is a schematic diagram of the track architecture according to the embodiment of the present invention;
图4为本发明实施例的设备信息捕捉区域例图；Figure 4 is an example diagram of a device information capture area according to an embodiment of the present invention;
图5为本发明实施例不同阶级在同一水平位置所拍摄的照片的重合率示意图。FIG. 5 is a schematic diagram of the coincidence rate of photos taken at the same horizontal position at different levels according to the embodiment of the present invention.
具体实施方式Detailed ways
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实施例，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他实施例，都属于本发明保护的范围。The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.
为使本发明的上述目的、特征和优点能够更加明显易懂，下面结合附图和具体实施方式对本发明作进一步详细的说明。 In order to make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.
如图1所示，本发明提供一种基于挂轨轨道机器人的尾矿坝表面形变巡检方法，包括：As shown in Figure 1, the present invention provides a tailings dam surface deformation inspection method based on a rail-mounted orbital robot, which includes:
(1)三维监测数据的获取(1) Acquisition of three-dimensional monitoring data
针对尾矿库坝体GNSS和测斜仪监测覆盖区域有限、无人机巡检精度不足的问题，拟选取三个已经装载双目相机和激光雷达的轨道滑轨机器人来对尾矿库坝体进行全区的在线监测(如图2所示)，将双目相机获取的视觉数据和激光雷达的获取的激光点云等多源数据进行融合。即收集尾矿库坝体表面的图像数据和激光扫描所获得的三维点云。In view of the problems of limited GNSS and inclinometer monitoring coverage area of the tailings dam body and insufficient accuracy of drone inspections, it is planned to select three orbital slide robots equipped with binocular cameras and lidar to inspect the tailings dam body. Carry out online monitoring of the entire area (as shown in Figure 2), and fuse multi-source data such as visual data obtained by binocular cameras and laser point clouds obtained by lidar. That is, the image data of the tailings dam surface and the three-dimensional point cloud obtained by laser scanning are collected.
在对尾矿库坝体进行全区的在线监测时，将数据采集和轨道构建进行高效结合(如图3所示)，在监测目标物体的外围构建挂轨机器人的阶梯式运行轨道，轨道内置供能回路，可使挂轨机器人正常运行。通过阶梯式的架设可以使设备采集的信息的区域提升。采集设备正常不改变工作朝向，但是针对一些特殊的需求可以通过改变轨道运行方向来进行工作朝向的改变(如图4所示)。When conducting online monitoring of the tailings reservoir dam body in the entire area, data collection and track construction are efficiently combined (as shown in Figure 3). A stepped operating track of the rail-mounted robot is constructed around the monitoring target object. The track is built-in The energy supply circuit allows the rail-mounted robot to operate normally. Through the stepped installation, the area of information collected by the equipment can be improved. The collection equipment normally does not change the working orientation, but for some special needs, the working orientation can be changed by changing the orbital running direction (as shown in Figure 4).
特别地，针对较大范围内的采集任务，轨道的阶级构建要遵循图像采集的需求。即上下不同阶级在同一水平位置所拍摄的照片的重合率要达到50％(如图5所示)。In particular, for larger-scale acquisition tasks, the hierarchical construction of orbits should follow the needs of image acquisition. That is, the coincidence rate of photos taken at the same horizontal position from different levels above and below should reach 50% (as shown in Figure 5).
在轨道滑轨机器人的控制方面，拟搭建上位机监控层、通讯层、控制层、数据采集层的系统框架，开发巡检远程交互控制系统。其中上位机监控层用于实时查看机器人的具体位置和相关设备的运行状态，并通过通讯层实时反馈到上位机；通讯层作为设备与上位机进行 数据传输；控制层由上位机远程控制，用于控制设备的电机运转；数据采集层通过相机采集图像数据，通过激光雷达采集三维点云数据。特别地，用户可以根据预设采集区域灵活控制轨道滑轨机器人的运行速度。In terms of the control of the rail slide robot, it is planned to build a system framework of the host computer monitoring layer, communication layer, control layer, and data acquisition layer, and develop an inspection remote interactive control system. Among them, the host computer monitoring layer is used to view the specific position of the robot and the operating status of related equipment in real time, and feedback to the host computer in real time through the communication layer; the communication layer acts as a device to communicate with the host computer. Data transmission; the control layer is remotely controlled by the host computer and used to control the motor operation of the equipment; the data collection layer collects image data through cameras and three-dimensional point cloud data through lidar. In particular, users can flexibly control the running speed of the track slide robot according to the preset collection area.
自主研发基于高速伺服电机驱动的尾矿库坝体双环形轨道巡检机器人来辅助完成监测数据的初步采集。特别的，在轨道滑轨机器人上加载补光灯，用来弥补夜间视觉模式失效。Independently developed a double-circular track inspection robot for tailings dams driven by high-speed servo motors to assist in the initial collection of monitoring data. In particular, a fill light is installed on the track slide robot to compensate for the failure of the night vision mode.
(2)三维监测模型的生成(2) Generation of three-dimensional monitoring model
视觉方面首先通过超像素和智能优化等方法对采集到的图像进行图像去噪来得到更准确的图像数据。In terms of vision, we first perform image denoising on the collected images through methods such as super pixels and intelligent optimization to obtain more accurate image data.
利用高斯滤波进行去噪处理，具体操作是：用一个模板(或称卷积、掩模)扫描图像中的每一个像素，用模板确定的邻域内像素的加权平均灰度值去替代模板中心像素点的值。Use Gaussian filtering for denoising. The specific operation is: use a template (or convolution, mask) to scan each pixel in the image, and use the weighted average gray value of the pixels in the neighborhood determined by the template to replace the central pixel of the template. point value.
然后通过SGBM算法对图像数据进行预处理，建立尾矿库三维监测模型，并将其三维监测数据储存到数据库中。Then the image data is preprocessed through the SGBM algorithm, a three-dimensional monitoring model of the tailings pond is established, and the three-dimensional monitoring data is stored in the database.
激光雷达方面将采集到的雷达数据和双目相机所采集到的图像数据相融合，具体做法是用LVI-SAM算法对图像数据和雷达数据进行处理生成一个经过两相校准的整个尾矿库坝体的三维点云数据。LiDAR fuses the collected radar data with the image data collected by the binocular camera. The specific method is to use the LVI-SAM algorithm to process the image data and radar data to generate a two-phase calibrated entire tailings dam. 3D point cloud data of the volume.
尾矿库三维监测模型的具体构建方法是，使用Sobel算子对图像进行预处理；对预处理后的图像进行代价计算；对进行代价计算后的图像进行动态规划，通过立体标定、立体匹配获得视差图；根据所述 视差图，得到深度图；根据所述深度图绘制点云图；根据所述点云图，构建所述三维监测模型。The specific construction method of the three-dimensional monitoring model of the tailings pond is to use the Sobel operator to preprocess the image; perform cost calculation on the preprocessed image; perform dynamic programming on the cost-calculated image, and obtain it through stereo calibration and stereo matching. Disparity map; as described The disparity map is used to obtain a depth map; a point cloud map is drawn based on the depth map; and the three-dimensional monitoring model is constructed based on the point cloud map.
(3)三维监测模型的处理分析(3) Processing and analysis of three-dimensional monitoring model
选取前后不同时刻生成的三维监测模型或三维点云导入cloudcompare中进行点云对齐，对齐后计算前后点云距离，通过选取合适的阈值来进行前后点云差异的可视化，再通过进一步的计算三维模型的面积和体积可以进一步的深入了解到坝体的形变情况，并根据这个设立报警阈值Select the 3D monitoring model or 3D point cloud generated at different times before and after and import it into cloudcompare for point cloud alignment. After alignment, calculate the distance between the front and rear point clouds. Select an appropriate threshold to visualize the difference between the front and rear point clouds, and then further calculate the 3D model. The area and volume can further understand the deformation of the dam body, and set an alarm threshold based on this.
进行点云对齐的过程包括，计算三维点云间的倒角距离，即计算生成点云和groundtruth点云之间，平均的最短点距离。通过倒角距离的大小来判定与作比较的原点云的差别。The process of point cloud alignment includes calculating the chamfer distance between three-dimensional point clouds, that is, calculating the average shortest point distance between the generated point cloud and the groundtruth point cloud. The difference with the origin cloud for comparison is determined by the size of the chamfer distance.
倒角距离对生成点云与原点云进行比较对齐通过点云配准完成。The chamfer distance compares the generated point cloud with the original point cloud and the alignment is completed through point cloud registration.
点云配准(Point Cloud Registration)指的是输入两幅点云(source)和(target)，输出一个变换使得(source)和(target)的重合程度尽可能高。本发明只考虑刚性(rigid)变换，即变换只包括旋转、平移。点云配准可以分为粗配准(Coarse Registration)和精配准(Fine Registration)两步。粗配准指的是在两幅点云之间的变换完全未知的情况下进行较为粗糙的配准，目的主要是为精配准提供较好的变换初值；精配准则是给定一个初始变换，进一步优化得到更精确的变换。Point Cloud Registration refers to inputting two point clouds (source) and (target), and outputting a transformation to make the degree of overlap between (source) and (target) as high as possible. This invention only considers rigid transformation, that is, the transformation only includes rotation and translation. Point cloud registration can be divided into two steps: Coarse Registration and Fine Registration. Coarse registration refers to a relatively rough registration when the transformation between two point clouds is completely unknown. The main purpose is to provide a better initial transformation value for fine registration; the fine registration criterion is to give an initial value. Transform and further optimize to obtain a more accurate transformation.
目前应用最广泛的点云精配准算法是迭代最近点算法(Iterative Closest Point,ICP)及各种变种ICP算法，本发明基 于ICP算法进行点云对齐。The most widely used point cloud precision registration algorithm at present is the iterative closest point algorithm (ICP) and various variants of the ICP algorithm. This invention is based on Point cloud alignment is performed using the ICP algorithm.
对于T是刚性变换的情形，点云配准问题可以描述为：
For the case where T is a rigid transformation, the point cloud registration problem can be described as:
这里ps和pt是源点云和目标点云中的对应点。Here p s and p t are the corresponding points in the source point cloud and target point cloud.
ICP一般算法流程为：对所述三维点云进行预处理，获得原始变换；对所述原始变换进行匹配，获得最近点；通过加权，调整对应点对的权重，剔除不合理的对应点对；通过计算loss，获得最小化loss；基于最小化loss，获得最优变换；对以上步骤进行迭代，直到收敛。The general algorithm flow of ICP is: preprocess the three-dimensional point cloud to obtain the original transformation; match the original transformation to obtain the closest point; adjust the weight of the corresponding point pairs through weighting, and eliminate unreasonable corresponding point pairs; By calculating the loss, the minimized loss is obtained; based on the minimized loss, the optimal transformation is obtained; the above steps are iterated until convergence.
本说明书实例所述的内容仅仅是对发明构思的实现形式的列举，本发明的保护范围不应当被视为仅限于实例所陈述的具体形式，本发明的保护范围也及于本领域技术人员根据本发明构思所能够想到的等同技术手段。 The content described in the examples in this specification is only an enumeration of the implementation forms of the inventive concept. The protection scope of the present invention should not be considered to be limited to the specific forms stated in the examples. The protection scope of the present invention also extends to those skilled in the art according to the Equivalent technical means that can be thought of according to the concept of the present invention.

Claims (10)

  1. 基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot is characterized by including:
    采集尾矿库坝体表面的数据，对所述数据进行预处理，获得三维监测数据；Collect data on the surface of the tailings reservoir dam, preprocess the data, and obtain three-dimensional monitoring data;
    基于所述三维监测数据进行建模，获得三维监测模型；Perform modeling based on the three-dimensional monitoring data to obtain a three-dimensional monitoring model;
    对所述三维监测模型进行处理分析，获得坝体形变情况。The three-dimensional monitoring model is processed and analyzed to obtain the deformation condition of the dam body.
  2. 根据权利要求1所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 1, characterized in that it includes:
    采集所述尾矿库坝体表面的数据的过程包括，基于轨道滑轨机器人对所述尾矿库坝体进行数据采集，通过所述轨道滑轨机器人的双目相机采集获得图像数据，通过所述轨道滑轨机器人的激光雷达采集获得三维点云数据。The process of collecting data on the surface of the tailings dam body includes collecting data on the tailings dam body based on an orbital slide robot, acquiring image data through the binocular camera collection of the orbital slide robot, and collecting the image data through the binocular camera of the orbital slide robot. The laser radar of the orbital slide robot is used to collect three-dimensional point cloud data.
  3. 根据权利要求2所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 2, characterized in that it includes:
    所述轨道滑轨机器人通过搭建数据采集层，上位机监控层，通讯层，控制层获得；The rail slide robot is obtained by building a data collection layer, a host computer monitoring layer, a communication layer, and a control layer;
    所述数据采集层用于采集所述图像数据和所述三维点云数据；The data collection layer is used to collect the image data and the three-dimensional point cloud data;
    所述上位机监控层用于实时查看轨道滑轨机器人的位置信息和设备的运行状态；The host computer monitoring layer is used to view the position information of the track slide robot and the operating status of the equipment in real time;
    所述通讯层用于与上位机进行数据传输；The communication layer is used for data transmission with the host computer;
    所述控制层由上位机远程控制，用于控制设备的电机运转。 The control layer is remotely controlled by the host computer and is used to control the motor operation of the equipment.
  4. 根据权利要求2所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 2, characterized in that it includes:
    对所述数据进行预处理的过程包括，基于超像素和智能优化方法对所述图像数据进行去噪，基于去噪后的图像数据和所述三维点云数据进行数据融合处理，获得所述三维监测数据。The process of preprocessing the data includes denoising the image data based on super pixels and intelligent optimization methods, performing data fusion processing based on the denoised image data and the three-dimensional point cloud data, and obtaining the three-dimensional point cloud data. Monitoring data.
  5. 根据权利要求4所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 4, characterized in that it includes:
    基于超像素和智能优化方法对所述图像数据进行去噪的过程包括，通过高斯滤波，采用掩模扫描图像中的每一个像素，用掩膜邻域内像素的加权平均灰度值替代掩膜中心像素点的值，获得去噪图像数据。The process of denoising the image data based on superpixel and intelligent optimization methods includes scanning each pixel in the image with a mask through Gaussian filtering, and replacing the mask center with the weighted average gray value of the pixels in the mask neighborhood. The value of the pixel point is used to obtain the denoised image data.
  6. 根据权利要求4所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 4, characterized in that it includes:
    所述数据融合处理的过程包括，基于LVI-SAM算法对所述图像数据和所述三维点云数据进行处理，生成经过两相校准的尾矿库坝体的各个时刻的三维点云。The process of data fusion processing includes processing the image data and the three-dimensional point cloud data based on the LVI-SAM algorithm to generate a three-dimensional point cloud at each moment of the two-phase calibrated tailings dam body.
  7. 根据权利要求1所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 1, characterized in that it includes:
    基于所述三维监测数据进行建模，获得三维监测模型的过程包括，Modeling is performed based on the three-dimensional monitoring data. The process of obtaining the three-dimensional monitoring model includes:
    基于所述三维监测数据获得监测图像，使用Sobel算子对所述监测图像进行预处理；对预处理后的图像进行代价计算；对进行代价计 算后的图像进行动态规划，通过立体标定、立体匹配获得视差图；基于所述视差图，得到深度图；基于所述深度图绘制点云图；基于所述点云图，构建所述三维监测模型。Obtain a monitoring image based on the three-dimensional monitoring data, use the Sobel operator to preprocess the monitoring image; perform cost calculation on the preprocessed image; perform cost calculation on the preprocessed image. The calculated image is dynamically programmed, and a disparity map is obtained through stereo calibration and stereo matching; a depth map is obtained based on the disparity map; a point cloud map is drawn based on the depth map; and the three-dimensional monitoring model is constructed based on the point cloud map.
  8. 根据权利要求1所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 1, characterized in that it includes:
    对所述三维监测模型进行处理分析的过程包括，将不同时刻的三维点云导入三维监测模型进行点云对齐，对齐后计算获得前后点云距离；通过预设阈值进行所述前后点云距离差异的可视化，获得可视化结果；计算所述三维监测模型的面积和体积，基于所述面积和体积以及所述可视化结果获得所述坝体形变情况。The process of processing and analyzing the three-dimensional monitoring model includes: importing three-dimensional point clouds at different times into the three-dimensional monitoring model to align the point clouds, and calculate and obtain the distance between the front and rear point clouds after the alignment; and calculate the distance difference between the front and rear point clouds through a preset threshold Visualize to obtain visualization results; calculate the area and volume of the three-dimensional monitoring model, and obtain the deformation of the dam body based on the area and volume and the visualization results.
  9. 根据权利要求8所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 8, characterized in that it includes:
    将不同时刻的三维点云导入三维监测模型进行点云对齐的过程包括，计算三维点云间的倒角距离，基于所述倒角距离对生成点云与原点云进行比较对齐。The process of importing three-dimensional point clouds at different times into the three-dimensional monitoring model for point cloud alignment includes calculating the chamfer distance between the three-dimensional point clouds, and comparing and aligning the generated point cloud with the original point cloud based on the chamfer distance.
  10. 根据权利要求9所述的基于挂轨轨道机器人的尾矿坝表面形变巡检方法，其特征在于，包括：The tailings dam surface deformation inspection method based on a rail-mounted orbital robot according to claim 9, characterized in that it includes:
    基于所述倒角距离对生成点云与原点云进行比较对齐通过ICP点云配准算法完成；Based on the chamfer distance, the generated point cloud and the original point cloud are compared and aligned through the ICP point cloud registration algorithm;
    所述ICP点云配准算法的过程包括，对所述三维点云进行预处理，获得原始变换；对所述原始变换进行匹配，获得最近点；通过加权，调整对应点对的权重，剔除不合理的对应点对；通过计算loss， 获得最小化loss；基于最小化loss，获得最优变换。 The process of the ICP point cloud registration algorithm includes preprocessing the three-dimensional point cloud to obtain the original transformation; matching the original transformation to obtain the closest point; adjusting the weight of the corresponding point pairs through weighting, and eliminating incorrect points. Reasonable corresponding point pairs; by calculating loss, Obtain the minimized loss; based on the minimized loss, obtain the optimal transformation.
PCT/CN2023/075438 2022-04-25 2023-02-10 Tailing dam surface deformation patrol method based on rail-mounted rail robot WO2023207257A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210439870.3A CN114841944B (en) 2022-04-25 2022-04-25 Tailing dam surface deformation inspection method based on rail-mounted robot
CN202210439870.3 2022-04-25

Publications (1)

Publication Number Publication Date
WO2023207257A1 true WO2023207257A1 (en) 2023-11-02

Family

ID=82566207

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/075438 WO2023207257A1 (en) 2022-04-25 2023-02-10 Tailing dam surface deformation patrol method based on rail-mounted rail robot

Country Status (2)

Country Link
CN (1) CN114841944B (en)
WO (1) WO2023207257A1 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114841944B (en) * 2022-04-25 2023-03-28 中钢集团马鞍山矿山研究总院股份有限公司 Tailing dam surface deformation inspection method based on rail-mounted robot
CN116879920B (en) * 2023-07-12 2024-02-09 中电建绿建有限公司 Dyke quality detection method and system based on laser point cloud technology

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106323243A (en) * 2016-08-18 2017-01-11 广州地理研究所 Dam deformation observation early-warning system, method and device based on unmanned aerial vehicle
CN109060025A (en) * 2018-08-13 2018-12-21 中国恩菲工程技术有限公司 Safety of Tailings Dam monitoring system and method
CN109269555A (en) * 2018-09-03 2019-01-25 空中梦想（北京）科技有限公司 System and method is monitored based on unmanned plane and the Safety of Tailings Dam of dimensional Modeling Technology
CN110672026A (en) * 2019-08-30 2020-01-10 泉州佰桀智能制鞋设备研究院有限公司 Roller coaster track detection method, intelligent detection robot and detection system
CN114841944A (en) * 2022-04-25 2022-08-02 中钢集团马鞍山矿山研究总院股份有限公司 Tailing dam surface deformation inspection method based on rail-mounted robot

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH09113239A (en) * 1995-10-19 1997-05-02 Ishikawajima Harima Heavy Ind Co Ltd Instrument for measuring structure having three-dimensional shape
CN110232736B (en) * 2019-06-18 2023-05-26 中国矿业大学 Method for quickly constructing three-dimensional scene of underground fully-mechanized mining face
CN111547084B (en) * 2020-04-22 2021-07-30 中国铁路设计集团有限公司 Data processing method based on self-moving type rail transit moving three-dimensional scanning system
CN113763583A (en) * 2020-05-19 2021-12-07 深圳市路桥建设集团有限公司 Tunnel inspection equipment and tunnel inspection method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106323243A (en) * 2016-08-18 2017-01-11 广州地理研究所 Dam deformation observation early-warning system, method and device based on unmanned aerial vehicle
CN109060025A (en) * 2018-08-13 2018-12-21 中国恩菲工程技术有限公司 Safety of Tailings Dam monitoring system and method
CN109269555A (en) * 2018-09-03 2019-01-25 空中梦想（北京）科技有限公司 System and method is monitored based on unmanned plane and the Safety of Tailings Dam of dimensional Modeling Technology
CN110672026A (en) * 2019-08-30 2020-01-10 泉州佰桀智能制鞋设备研究院有限公司 Roller coaster track detection method, intelligent detection robot and detection system
CN114841944A (en) * 2022-04-25 2022-08-02 中钢集团马鞍山矿山研究总院股份有限公司 Tailing dam surface deformation inspection method based on rail-mounted robot

Also Published As

Publication number Publication date
CN114841944A (en) 2022-08-02
CN114841944B (en) 2023-03-28

Similar Documents

Publication Publication Date Title
WO2023207257A1 (en) Tailing dam surface deformation patrol method based on rail-mounted rail robot
WO2022188379A1 (en) Artificial intelligence system and method serving electric power robot
CN110108258B (en) Monocular vision odometer positioning method
CN111679291B (en) Inspection robot target positioning configuration method based on three-dimensional laser radar
CN110189399B (en) Indoor three-dimensional layout reconstruction method and system
CN104537707A (en) Image space type stereo vision on-line movement real-time measurement system
CN109242959A (en) Method for reconstructing three-dimensional scene and system
CN115082254A (en) Lean control digital twin system of transformer substation
WO2024007485A1 (en) Aerial-ground multi-vehicle map fusion method based on visual feature
CN115222884A (en) Space object analysis and modeling optimization method based on artificial intelligence
CN111860651A (en) Monocular vision-based semi-dense map construction method for mobile robot
CN116030208A (en) Method and system for building scene of virtual simulation power transmission line of real unmanned aerial vehicle
CN114419259B (en) Visual positioning method and system based on physical model imaging simulation
CN113920254B (en) Monocular RGB (Red Green blue) -based indoor three-dimensional reconstruction method and system thereof
CN113129348B (en) Monocular vision-based three-dimensional reconstruction method for vehicle target in road scene
CN111354028B (en) Binocular vision-based power transmission channel hidden danger identification and tracking method
CN114273826A (en) Automatic identification method for welding position of large-sized workpiece to be welded
CN111191513A (en) Method for estimating position of mobile robot based on scene size analysis
Gharib et al. A Review of Computer Vision-Based Techniques for Construction Progress Monitoring
Kaiser et al. Co-registration of video-grammetric point clouds with BIM–first conceptual results
CN109360174A (en) Method for reconstructing three-dimensional scene and system based on camera pose
Chen et al. Improving completeness and accuracy of 3D point clouds by using deep learning for applications of digital twins to civil structures
Dong et al. ClusterFusion: Real-time Relative Positioning and Dense Reconstruction for UAV Cluster
Zhang et al. Multi-Source and Multi-Dimensional Data Fusion of Magnetic Levitation Track Transportation Based on Digital Twin
Yue et al. Transmission Line Component Inspection Method Based on Deep Learning under Visual Navigation

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: 23794693

Country of ref document: EP

Kind code of ref document: A1