技术领域Technical Field
本发明属于铁路基础设施监测技术领域，具体涉及一种基于无人机技术的铁路基础设施监测方法和系统。The present invention belongs to the technical field of railway infrastructure monitoring, and in particular relates to a railway infrastructure monitoring method and system based on unmanned aerial vehicle technology.
背景技术Background technique
矿区铁路作为矿山生产建设的重要交通设施，地下采煤对其造成的影响不可忽略。传统的矿区铁路变形监测手段有目视巡检、水准测量、导线测量和全球卫星导航系统（GNSS）等方法。As an important transportation facility for mine production and construction, the impact of underground coal mining on the railway in the mining area cannot be ignored. Traditional methods of monitoring the deformation of railways in mining areas include visual inspection, leveling, traverse measurement and global satellite navigation system (GNSS).
其中，目视巡检法人为因素影响过多，易导致监测结果不准确，尤其不适用于一些微小变形便会引发重大灾害的场景中；水准测量、导线测量同属于对点位坐标进行量测的方法，能够获取准确的监测数据，较之目视巡检更为可靠，但获取的信息量较少且费时费力，多数情况下需要通过点位坐标来对线状地物进行拟合，在一定程度上影响目标地物的监测精度；GNSS用于变形监测时仅能得到离散的点位信息。Among them, visual inspections are too influenced by human factors and can easily lead to inaccurate monitoring results. They are especially unsuitable for scenarios where even a small deformation can cause major disasters. Leveling and traverse measurement are both methods of measuring point coordinates. They can obtain accurate monitoring data and are more reliable than visual inspections. However, the amount of information obtained is small and time-consuming and labor-intensive. In most cases, point coordinates need to be used to fit linear features, which affects the monitoring accuracy of the target features to a certain extent. When GNSS is used for deformation monitoring, only discrete point information can be obtained.
近年来，光学雷达技术、地面式三维激光扫描技术等手段被逐渐应用于变形监测，该类技术能够获得大量的点云数，并且拥有高精度的三维空间坐标，但观测成本相对较高。因此，目前变形监测领域亟须一种监测精度高、面向点云监测且价格低廉的监测技术手段。In recent years, optical radar technology and ground-based 3D laser scanning technology have been gradually applied to deformation monitoring. Such technologies can obtain a large number of point clouds and have high-precision 3D spatial coordinates, but the observation cost is relatively high. Therefore, the current deformation monitoring field urgently needs a monitoring technology with high monitoring accuracy, point cloud monitoring and low price.
发明内容Summary of the invention
本发明为了解决现有技术中存在的上述至少一个技术问题，提供了一种基于无人机技术的铁路基础设施监测方法和系统。In order to solve at least one of the above-mentioned technical problems existing in the prior art, the present invention provides a railway infrastructure monitoring method and system based on drone technology.
本发明采用如下的技术方案实现：一种基于无人机技术的铁路基础设施监测方法，包括以下步骤：The present invention is implemented by the following technical solution: a railway infrastructure monitoring method based on drone technology, comprising the following steps:
S1：选取长度为L的铁路基础测线作为监测对象，在该铁路基础测线的轨道两侧布置两排共M个测点；S1: Select a railway basic survey line with a length of L as the monitoring object, and arrange two rows of M measuring points on both sides of the track of the railway basic survey line;
S2：将M个测点分为像控点和检查点，通过改变像控点与检查点的数量，设计N个包含不同像控点和检查点组合的试验方案；S2: Divide the M measurement points into image control points and check points, and design N test schemes containing different combinations of image control points and check points by changing the number of image control points and check points;
S3：基于无人机影像对N个试验方案下的铁路基础测线近地拍摄，并基于影像处理算法生成N个试验方案下的铁路基础测线的重构点云模型；S3: Based on the UAV images, the railway foundation survey lines under N test schemes are photographed close to the ground, and the reconstructed point cloud models of the railway foundation survey lines under N test schemes are generated based on the image processing algorithm;
S4：通过实测获取N个试验方案下检查点的平面坐标和高程，并与对应试验方案下的重构点云模型中检查点的平面坐标和高程对比，评估不同试验方案下重构点云模型的精度；S4: Obtain the plane coordinates and elevations of the checkpoints under N test schemes through actual measurement, and compare them with the plane coordinates and elevations of the checkpoints in the reconstructed point cloud model under the corresponding test scheme to evaluate the accuracy of the reconstructed point cloud model under different test schemes;
S5：选取平面坐标和高程的组合精度最高的试验方案作为该铁路基础测线的最优监测方案。S5: Select the test plan with the highest combined accuracy of plane coordinates and elevation as the optimal monitoring plan for the railway foundation survey line.
优选地，铁路基础测线的两端各设置有一组并排的像控点，试验方案中的其余组并排设置的像控点等间隔设置在端点处的两组像控点之间。Preferably, a group of parallel image control points are respectively arranged at both ends of the railway foundation survey line, and the remaining groups of parallel image control points in the test scheme are arranged at equal intervals between the two groups of image control points at the end points.
优选地，影像处理算法为：Preferably, the image processing algorithm is:
使用Pix4DMapper进行影像数据的处理，基于影像之间的特征点进行匹配重构三维模型；Use Pix4DMapper to process image data and reconstruct 3D models based on matching feature points between images;
生成数字表面模型、数字正射影像图以及三维重构点云模型。Generate digital surface models, digital orthophotos and 3D reconstructed point cloud models.
优选地，通过数字表面模型和数字正射影像图重合区域处的影像覆盖数量判断两者间影像的重合度。Preferably, the degree of image overlap between the digital surface model and the digital orthophoto is determined by the amount of image coverage at the overlapped area of the digital surface model and the digital orthophoto.
优选地，N个试验方案下重构点云模型的精度通过均方根误差进行评估，包括以下4个评估指标：Preferably, the accuracy of the reconstructed point cloud model under N test schemes is evaluated by the root mean square error, including the following four evaluation indicators:
式中，为每个试验方案中检查点的个数；分别为外业实测检查点对应的轴、轴、轴的坐标；分别为三维重构点云模型中检查点对应的轴、轴、轴的坐标；、、、分别为方向、方向、方向以及平面的均方根误差；其中用于评估平面坐标的精度，用于评估高程的精度。In the formula, is the number of checkpoints in each test plan; They correspond to the field measurement checkpoints. axis, axis, The coordinates of the axes; They are the corresponding check points in the 3D reconstructed point cloud model. axis, axis, The coordinates of the axes; , , , They are direction, direction, Directions and The root mean square error of the plane; where Used to evaluate the accuracy of plane coordinates, Used to assess the accuracy of elevation.
本发明还提供了一种基于无人机技术的铁路基础设施监测系统，包括数据采集层、数据传输与存储层、数据处理与分析层；The present invention also provides a railway infrastructure monitoring system based on UAV technology, including a data acquisition layer, a data transmission and storage layer, and a data processing and analysis layer;
其中数据采集层用于进行铁路基础测线的图像采集，数据采集层包括无人机和搭载于无人机上的影像采集设备和气象传感器；The data acquisition layer is used to collect images of railway foundation survey lines. The data acquisition layer includes drones and image acquisition equipment and meteorological sensors mounted on the drones.
数据传输与存储层用于将采集到的数据实时传输至地面站或云端，并存储历史数据；The data transmission and storage layer is used to transmit the collected data to the ground station or the cloud in real time and store historical data;
数据处理与分析层用于通过Pix4DMapper进行数据处理，基于影像之间的特征点进行匹配重构三维模型，通过对不同试验方案下重构点云模型的精度评估，得到铁路基础测线的最优监测方案，通过最优监测方案实现对铁路基础设施的异常和潜在问题的检测。The data processing and analysis layer is used to process data through Pix4DMapper, reconstruct the three-dimensional model based on the matching of feature points between images, and obtain the optimal monitoring plan for the railway foundation survey line by evaluating the accuracy of the reconstructed point cloud model under different test schemes. The optimal monitoring plan can be used to detect anomalies and potential problems in railway infrastructure.
优选地，影像采集设备包括外接式光学摄像头、红外热成像摄像头和激光测距雷达；气象传感器用于实现对铁路基础测线周围温度、湿度和风速的监测，以纠正影像数据并评估环境对铁路基础设施的影响。Preferably, the image acquisition equipment includes an external optical camera, an infrared thermal imaging camera and a laser ranging radar; the meteorological sensor is used to monitor the temperature, humidity and wind speed around the railway infrastructure survey line to correct the image data and evaluate the impact of the environment on the railway infrastructure.
与现有技术相比，本发明的有益效果是：Compared with the prior art, the present invention has the following beneficial effects:
本发明通过无人机和搭载于无人机上的影像采集设备和气象传感器对铁路基础测线的图像进行采集，基于Pix4DMapper对采集的数据进行处理，基于影像之间的特征点进行匹配重构三维模型，通过对不同试验方案下重构点云模型的精度评估，得到铁路基础测线的最优监测方案，通过最优监测方案实现对铁路基础设施的异常和潜在问题的检测。The present invention collects images of railway infrastructure survey lines through unmanned aerial vehicles and image acquisition equipment and meteorological sensors carried on the unmanned aerial vehicles, processes the collected data based on Pix4DMapper, reconstructs a three-dimensional model based on matching feature points between images, and obtains the optimal monitoring scheme for the railway infrastructure survey lines by evaluating the accuracy of the reconstructed point cloud model under different test schemes. The optimal monitoring scheme is used to detect abnormalities and potential problems of railway infrastructure.
在不同试验方案中，平面误差与高程误差均会随着单位模型长度减小而减小，但误差减小到一定程度之后会趋于稳定，且平面误差远小于高程误差，通过该方法能够使无人机探测的数据精度实现增加。In different test schemes, both the plane error and the elevation error will decrease as the unit model length decreases, but the error will tend to stabilize after being reduced to a certain extent, and the plane error is much smaller than the elevation error. This method can increase the data accuracy of UAV detection.
在本监测方法和系统中，无人机近地摄影测量精度可达毫米级，能够为外业监测矿区铁路等类似地面建筑物变形提供一种快速、灵活、高性价比的技术手段。In the present monitoring method and system, the near-ground photogrammetry accuracy of the UAV can reach the millimeter level, which can provide a fast, flexible and cost-effective technical means for field monitoring of deformation of similar ground buildings such as mining railways.
附图说明BRIEF 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的介绍，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 more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
图1是本发明的10个试验方案的设计示意图；Fig. 1 is a schematic diagram of the design of 10 experimental schemes of the present invention;
图2是本发明的10个试验方案的像控点布设位置示意图；FIG2 is a schematic diagram of the layout positions of image control points of 10 test schemes of the present invention;
图3是本发明的数据处理流程图；FIG3 is a data processing flow chart of the present invention;
图4是本发明中的三维重构点云（稀疏）图；FIG4 is a 3D reconstructed point cloud (sparse) diagram in the present invention;
图5是本发明的10个试验方案的RMSE计算结果图；FIG5 is a diagram showing the RMSE calculation results of 10 test schemes of the present invention;
图6是本发明的平面坐标精度和高程精度模拟实验结果；FIG6 is a simulation experiment result of plane coordinate accuracy and elevation accuracy of the present invention;
图7是本发明的铁路基础设施变形测量精度图。FIG. 7 is a diagram showing the railway infrastructure deformation measurement accuracy of the present invention.
具体实施方式Detailed ways
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，而不是全部实施例。基于本发明的实施例，本领域的普通技术人员在没有做出创造性劳动的前提下所得到的所有其他实施方式，都属于本发明所保护的范围。In conjunction with the drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
须知，本说明书附图所绘示的结构、比例、大小等，均仅用以配合说明书所揭示的内容，以供熟悉此技术的人士了解与阅读，并非用以限定本发明可实施的限定条件，故不具技术上的实质意义，任何结构的修饰、比例关系的改变或大小的调整，在不影响本发明所能产生的功效及所能达成的目的下，均应落在本发明所揭示的技术内容能涵盖的范围内，需要说明的是，在本说明书中，诸如第一和第二之类的关系术语仅仅用来将一个实体与另外几个实体区分开来，而不一定要求或者暗示这些实体之间存在任何实际的关系或者顺序。It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so that people familiar with this technology can understand and read them. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportional relationship or adjustment of size should fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention. It should be noted that in this specification, relational terms such as first and second are only used to distinguish one entity from several other entities, and do not necessarily require or imply any actual relationship or order between these entities.
本发明提供了一种实施例：The present invention provides an embodiment:
一种基于无人机技术的铁路基础设施监测方法，包括以下步骤：A railway infrastructure monitoring method based on drone technology includes the following steps:
S1：选取长度为L的铁路基础测线作为监测对象，在该铁路基础测线的轨道两侧布置两排共M个测点；S1: Select a railway basic survey line with a length of L as the monitoring object, and arrange two rows of M measuring points on both sides of the track of the railway basic survey line;
S2：将M个测点分为像控点和检查点，通过改变像控点与检查点的数量，设计N个包含不同像控点和检查点组合的试验方案；S2: Divide the M measurement points into image control points and check points, and design N test schemes containing different combinations of image control points and check points by changing the number of image control points and check points;
S3：基于无人机影像对N个试验方案下的铁路基础测线近地拍摄，并基于影像处理算法生成N个试验方案下的铁路基础测线的重构点云模型；S3: Based on the UAV images, the railway foundation survey lines under N test schemes are photographed close to the ground, and the reconstructed point cloud models of the railway foundation survey lines under N test schemes are generated based on the image processing algorithm;
S4：通过实测获取N个试验方案下检查点的平面坐标和高程，并与对应试验方案下的重构点云模型中检查点的平面坐标和高程对比，评估不同试验方案下重构点云模型的精度；S4: Obtain the plane coordinates and elevations of the checkpoints under N test schemes through actual measurement, and compare them with the plane coordinates and elevations of the checkpoints in the reconstructed point cloud model under the corresponding test scheme to evaluate the accuracy of the reconstructed point cloud model under different test schemes;
S5：选取平面坐标和高程的组合精度最高的试验方案作为该铁路基础测线的最优监测方案，对铁路基础测线进行监测。S5: Select the test plan with the highest combined accuracy of plane coordinates and elevation as the optimal monitoring plan for the railway foundation survey line, and monitor the railway foundation survey line.
如图1、图2所示，本实施例中，铁路基础测线的长度L为60m，测点的数量M为100，试验方案的数量N为10；铁路基础测线的两端各设置有一组并排的像控点，试验方案中的其余组并排设置的像控点等间隔设置在端点处的两组像控点之间。其中，试验方案1所对应的像控点数量为4个，检查点数量为96个，单位模型长度为60m；方案2对应的像控点数量为6个，检查点数量为94个，单位模型长度为30m；方案3对应的像控点数量为8个，检查点数量为92个，单位模型长度为20m；方案4对应的像控点数量为10个，检查点数量为90个，单位模型长度为15m；方案5对应的像控点数量为12个，检查点数量为88个，单位模型长度为12m；方案6对应的像控点数量为14个，检查点数量为86个，单位模型长度为10m；方案7对应的像控点数量为16个，检查点数量为84个，单位模型长度为8.57m；方案8对应的像控点数量为18个，检查点数量为82个，单位模型长度为7.5m；方案9对应的像控点数量为20个，检查点数量为80个，单位模型长度为6.67m；方案10对应的像控点数量为22个，检查点数量为78个，单位模型长度为6m.其中单位模型长度=铁路基础测线总长度/测段数。As shown in Figures 1 and 2, in this embodiment, the length L of the railway foundation survey line is 60m, the number of survey points M is 100, and the number of test schemes N is 10; a group of image control points arranged side by side are respectively set at both ends of the railway foundation survey line, and the remaining groups of image control points arranged side by side in the test scheme are arranged at equal intervals between the two groups of image control points at the end points. Among them, the number of image control points corresponding to test scheme 1 is 4, the number of checkpoints is 96, and the unit model length is 60m; the number of image control points corresponding to scheme 2 is 6, the number of checkpoints is 94, and the unit model length is 30m; the number of image control points corresponding to scheme 3 is 8, the number of checkpoints is 92, and the unit model length is 20m; the number of image control points corresponding to scheme 4 is 10, the number of checkpoints is 90, and the unit model length is 15m; the number of image control points corresponding to scheme 5 is 12, the number of checkpoints is 88, and the unit model length is 12m; the number of image control points corresponding to scheme 6 is The number of image control points for scheme 7 is 16, the number of checkpoints is 84, and the unit model length is 8.57m; the number of image control points for scheme 8 is 18, the number of checkpoints is 82, and the unit model length is 7.5m; the number of image control points for scheme 9 is 20, the number of checkpoints is 80, and the unit model length is 6.67m; the number of image control points for scheme 10 is 22, the number of checkpoints is 78, and the unit model length is 6m. The unit model length = the total length of the railway foundation survey line/the number of survey sections.
本实施例中，像控点是指摄影测量控制加密和测图的基础，像控点全称像片控制点，是摄影测量学，中的一个重要概念，它指的是在摄影测量控制加密和测图过程中所依据的基础控制点。这些点通常在实地布设并进行测定，直接为摄影测量的控制点加密或测图需要服务。像控点的分布、数量和联测精度根据航空摄影资料、后续工序工艺流程及成图精度的要求来确定。像控点的类型包括仅具有平面坐标的像片平面控制点，仅具高程的像片高程控制点及同时具有平面坐标与高程的像片平高控制点；In this embodiment, image control points refer to the basis of photogrammetric control encryption and mapping. The full name of image control points is image control points, which is an important concept in photogrammetry. It refers to the basic control points based on the photogrammetric control encryption and mapping process. These points are usually laid out and measured in the field, directly serving the needs of photogrammetric control point encryption or mapping. The distribution, quantity and joint measurement accuracy of image control points are determined according to the requirements of aerial photography data, subsequent process flow and mapping accuracy. The types of image control points include image plane control points with only plane coordinates, image elevation control points with only elevation, and image plane-height control points with both plane coordinates and elevation;
像控点（GCP）用于纠正因各种因素导致拍摄的目标三维空间误差，检查点（CP）为计算拍摄成果误差的测量点，检查点的布设位置为现有技术，在此不做赘述。The image control points (GCP) are used to correct the three-dimensional spatial errors of the captured target caused by various factors. The check points (CP) are measurement points for calculating the errors of the captured results. The layout of the check points is existing technology and will not be described here.
本实施例中，影像处理算法为：In this embodiment, the image processing algorithm is:
如图3所示，使用Pix4DMapper进行影像数据的处理，基于影像之间的特征点进行匹配重构三维模型；As shown in Figure 3, Pix4DMapper is used to process the image data and reconstruct the 3D model based on matching feature points between images;
生成数字表面模型、数字正射影像图以及三维重构点云模型；其中通过数字表面模型和数字正射影像图重合区域处的影像覆盖数量判断两者间影像的重合度，绝大部分区域的影像覆盖数量均不少于5幅。其中数字正射影像图是指将无人机拍摄的一张张图片合成一张整图，数字表面模型是指包含铁路基础测线的地表建筑物、桥梁和树木等高度的地面高程模型，用于提取地表高程，此处为现有技术，在此不做赘述。Generate digital surface model, digital orthophoto map and 3D reconstructed point cloud model; the overlap between the digital surface model and the digital orthophoto map is determined by the number of images covered in the overlapping area of the two images, and the number of images covered in most areas is not less than 5. The digital orthophoto map refers to a whole picture composed of pictures taken by drones, and the digital surface model refers to a ground elevation model including the heights of surface buildings, bridges and trees of the railway foundation survey line, which is used to extract the surface elevation. This is the existing technology and will not be described in detail.
如图4至图6所示，10个试验方案下重构点云模型的精度通过均方根误差进行评估，通过比对外业实测检查点的坐标与内业三维重建点云模型中的坐标，计算出以下4个评估指标来评估重建点云模型的精度：As shown in Figures 4 to 6, the accuracy of the reconstructed point cloud model under the 10 test schemes is evaluated by the root mean square error. By comparing the coordinates of the field-measured inspection points with the coordinates in the 3D reconstructed point cloud model, the following four evaluation indicators are calculated to evaluate the accuracy of the reconstructed point cloud model:
式中，为每个试验方案中检查点的个数；分别为外业实测检查点对应的轴、轴、轴的坐标；分别为三维重构点云模型中检查点对应的轴、轴、轴的坐标；、、、分别为方向、方向、方向以及平面的均方根误差；其中用于评估平面坐标的精度，用于评估高程的精度。In the formula, is the number of checkpoints in each test plan; They correspond to the field measurement checkpoints. axis, axis, The coordinates of the axes; They are the corresponding check points in the 3D reconstructed point cloud model. axis, axis, The coordinates of the axes; , , , They are direction, direction, Directions and The root mean square error of the plane; where Used to evaluate the accuracy of plane coordinates, Used to assess the accuracy of elevation.
根据图5和图6所示的平面精度分析模拟试验结果可知（平面精度分析模拟试验可以通过平面精度分析软件对采集的图像进行分析）：According to the plane accuracy analysis simulation test results shown in Figures 5 and 6 (the plane accuracy analysis simulation test can analyze the collected images through the plane accuracy analysis software):
1. 随着像控点数量增加，单位模型长度相应减小（单位模型长度=测线总长度/测段数），检查点的也随之减小；1. As the number of image control points increases, the unit model length decreases accordingly (unit model length = total length of the survey line/number of survey sections). It also decreases;
2. 单位模型长度从试验方案1中的60m减小到了试验方案4中的15m，由11.1mm减小至2.3mm；2. The unit model length was reduced from 60m in Test Scheme 1 to 15m in Test Scheme 4. Reduced from 11.1mm to 2.3mm;
3. 从试验方案5到试验方案10，单位模型长度变化几乎不再对产生影响，最终约2mm，即在试验方案5中单位模型长度为12m时，可获得较优的平面坐标精度。3. From test plan 5 to test plan 10, the unit model length change is almost no longer Make an impact, Finally, about 2 mm, that is, when the unit model length is 12 m in test scheme 5, a better plane coordinate accuracy can be obtained.
根据图5和图6的高程精度分析模拟试验结果可知：According to the simulation test results of the elevation accuracy analysis in Figures 5 and 6, it can be seen that:
1. 随着像控点数量增加，单位模型长度相应减小，所有检查点的也随之减小；1. As the number of image control points increases, the unit model length decreases accordingly. It also decreases;
2. 单位模型长度从试验方案1中的60m减小到了试验方案5中的12m，由65.2mm减小至7.4mm；2. The unit model length was reduced from 60m in Test Scheme 1 to 12m in Test Scheme 5. Reduced from 65.2mm to 7.4mm;
3. 从试验方案6到试验方案10，单位模型长度变化对的影响减小，最终约7mm，即在试验方案5中单位模型长度为12m时，可获得较优的高程精度。3. From test plan 6 to test plan 10, the change of unit model length The impact of Finally, about 7 mm, that is, when the unit model length is 12 m in test scheme 5, a better elevation accuracy can be obtained.
结合平面精度分析模拟试验结果和高程精度分析模拟试验结果，可知试验方案5所对应的平面坐标误差和高程误差的组合误差最低，将试验方案5作为该铁路基础测线的最优监测方案，对铁路基础测线进行监测。Combining the results of the plane accuracy analysis simulation test and the elevation accuracy analysis simulation test, it can be seen that the combined error of the plane coordinate error and the elevation error corresponding to test plan 5 is the lowest. Test plan 5 is used as the optimal monitoring plan for the railway foundation survey line, and the railway foundation survey line is monitored.
通过上述试验，可知无人机近地监测矿区铁路的平面与高程精度均可达到毫米级。Through the above experiments, it can be seen that the plane and elevation accuracy of UAV's close-range monitoring of mining area railways can reach millimeter level.
后续监测中，选择一段长度60m的矿区专用铁路作为试验对象，测点布设在铁路两侧的枕木之上，分两排共布设了22个测点，点间距6m，两排间距1.6m，使用与前述试验相同的无人机影像获取参数以及平面坐标与高程测量方法获得影像及测点坐标数据。In the subsequent monitoring, a 60-meter-long mining railway was selected as the test object. The measuring points were arranged on the sleepers on both sides of the railway. A total of 22 measuring points were arranged in two rows, with a point spacing of 6m and a spacing of 1.6m between two rows. The same UAV image acquisition parameters as in the previous experiment and the plane coordinate and elevation measurement methods were used to obtain the image and measuring point coordinate data.
选择前述试验方案5进行验证，即像控点布设间隔为6m，选择12个测点作为像控点，其余10个点作为检查点（在后续实验中，检查点数目可对应减少，只需保证像控点数量与试验方案5中的一致即可），试验结果如图7所示。The aforementioned test plan 5 was selected for verification, that is, the image control points were arranged at an interval of 6m, 12 measuring points were selected as image control points, and the remaining 10 points were selected as check points (in subsequent experiments, the number of check points can be reduced accordingly, as long as the number of image control points is consistent with that in test plan 5). The test results are shown in Figure 7.
由图7可知：As shown in Figure 7:
1. 高程误差明显大于平面误差，且最小高程误差仍大于最大平面误差；1. The elevation error is significantly greater than the plane error, and the minimum elevation error is still greater than the maximum plane error;
2. 高程误差波动较为明显，平面误差起伏波动较为稳定；2. The fluctuation of elevation error is more obvious, while the fluctuation of plane error is more stable;
3. 高程误差平均为6.3mm，平面坐标误差平均为2.3mm，该结果与模拟试验结果基本一致，符合试验预期。3. The average elevation error is 6.3 mm, and the average plane coordinate error is 2.3 mm. The results are basically consistent with the simulation test results and meet the test expectations.
本实施例中，无人机近地摄影测量平面坐标误差最小约为2mm，该精度是在单位模型长度为12m时获得；高程误差最小约为7mm，该精度也是在单位模型长度为12m时获得。在实测研究中，选择单位模型长度为12m的方案，获得的平面坐标误差为2.3mm，高程误差为6.3mm，与模拟试验结果基本一致。平面误差与高程误差均会随着单位模型长度减小而减小，但误差减小到一定程度之后会趋于稳定，而且平面误差远小于高程误差，通过该方法使得无人机探测的数据精度实现增加。In this embodiment, the minimum plane coordinate error of the drone near-ground photogrammetry is about 2mm, and this accuracy is obtained when the unit model length is 12m; the minimum elevation error is about 7mm, and this accuracy is also obtained when the unit model length is 12m. In the actual measurement research, the scheme with a unit model length of 12m was selected, and the obtained plane coordinate error was 2.3mm and the elevation error was 6.3mm, which is basically consistent with the simulation test results. Both the plane error and the elevation error will decrease as the unit model length decreases, but the error will tend to stabilize after it is reduced to a certain extent, and the plane error is much smaller than the elevation error. This method increases the data accuracy of drone detection.
本发明还提供了一种基于无人机技术的铁路基础设施监测系统，包括数据采集层、数据传输与存储层、数据处理与分析层；The present invention also provides a railway infrastructure monitoring system based on UAV technology, including a data acquisition layer, a data transmission and storage layer, and a data processing and analysis layer;
其中数据采集层用于进行铁路基础测线的图像采集，数据采集层包括无人机和搭载于无人机上的影像采集设备和气象传感器；The data acquisition layer is used to collect images of railway foundation survey lines. The data acquisition layer includes drones and image acquisition equipment and meteorological sensors mounted on the drones.
数据传输与存储层用于将采集到的数据实时传输至地面站或云端，并存储历史数据，支持大数据分析和长期趋势分析的基础支撑；The data transmission and storage layer is used to transmit the collected data to the ground station or the cloud in real time and store historical data, providing basic support for big data analysis and long-term trend analysis;
数据处理与分析层用于通过Pix4DMapper进行数据处理，基于影像之间的特征点进行匹配重构三维模型，通过对不同试验方案下重构点云模型的精度评估，得到铁路基础测线的最优监测方案，通过最优监测方案实现对铁路基础设施的异常和潜在问题的检测。The data processing and analysis layer is used to process data through Pix4DMapper, reconstruct the three-dimensional model based on the matching of feature points between images, and obtain the optimal monitoring plan for the railway foundation survey line by evaluating the accuracy of the reconstructed point cloud model under different test schemes. The optimal monitoring plan can be used to detect anomalies and potential problems in railway infrastructure.
本实施例中，影像采集设备包括外接式光学摄像头、红外热成像摄像头和激光测距雷达；气象传感器用于实现对铁路基础测线周围温度、湿度和风速的监测，以纠正影像数据并评估环境对铁路基础设施的影响。在无人机不同的飞行速度下，受到环境的风速、温度以及湿度的影响，摄像头采集的照片也会受到影响的，而通过气象传感器提前获取拍摄区域的温度、湿度以及风速，能够控制摄像头的拍摄参数、拍摄的间隔时间以及频率等等，从而避免排出失帧或模糊不清楚的图像，达到纠正影响数据的功能。In this embodiment, the image acquisition equipment includes an external optical camera, an infrared thermal imaging camera and a laser ranging radar; the meteorological sensor is used to monitor the temperature, humidity and wind speed around the railway infrastructure survey line to correct the image data and evaluate the impact of the environment on the railway infrastructure. At different flight speeds of the drone, the photos collected by the camera will also be affected by the wind speed, temperature and humidity of the environment. By obtaining the temperature, humidity and wind speed of the shooting area in advance through the meteorological sensor, the camera's shooting parameters, shooting interval time and frequency, etc. can be controlled to avoid the discharge of lost frames or blurred images, and achieve the function of correcting the affected data.
数据采集层还包括有实时监控与远程操作层，用于实时监控无人机的飞行状态、电池电量，支持远程操控，使操作者能够及时调整监测计划和路径。数据处理与分析层还包括有实时报警与通知层，用于监测到异常情况时，系统实时生成报警并通知相关维护人员，通过设置通信系统，使若干人员之间实现及时沟通和采取紧急措施。The data collection layer also includes a real-time monitoring and remote operation layer, which is used to monitor the flight status and battery power of the drone in real time, and supports remote control, so that the operator can adjust the monitoring plan and path in time. The data processing and analysis layer also includes a real-time alarm and notification layer, which is used to monitor abnormal conditions, and the system generates alarms in real time and notifies relevant maintenance personnel. By setting up a communication system, timely communication between several personnel can be achieved and emergency measures can be taken.
以上所述，仅为本发明较佳的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到的变化或替换，都应该涵盖在本发明的保护范围之内。因此，本发明的保护范围应以所述权利要求的保护范围为准。The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410897673.5ACN118429902A (en) | 2024-07-05 | 2024-07-05 | Railway infrastructure monitoring method and system based on unmanned aerial vehicle technology |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410897673.5ACN118429902A (en) | 2024-07-05 | 2024-07-05 | Railway infrastructure monitoring method and system based on unmanned aerial vehicle technology |
| Publication Number | Publication Date |
|---|---|
| CN118429902Atrue CN118429902A (en) | 2024-08-02 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410897673.5APendingCN118429902A (en) | 2024-07-05 | 2024-07-05 | Railway infrastructure monitoring method and system based on unmanned aerial vehicle technology |
| Country | Link |
|---|---|
| CN (1) | CN118429902A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105783878A (en)* | 2016-03-11 | 2016-07-20 | 三峡大学 | Small unmanned aerial vehicle remote sensing-based slope deformation detection and calculation method |
| CN112113542A (en)* | 2020-09-14 | 2020-12-22 | 浙江省自然资源征收中心 | Method for checking and accepting land special data for aerial photography construction of unmanned aerial vehicle |
| CN114998536A (en)* | 2022-05-31 | 2022-09-02 | 广州市城市规划勘测设计研究院 | Model generation method and device based on novel basic mapping and storage medium |
| CN116883604A (en)* | 2023-08-02 | 2023-10-13 | 中色蓝图科技股份有限公司 | Three-dimensional modeling technical method based on space, air and ground images |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105783878A (en)* | 2016-03-11 | 2016-07-20 | 三峡大学 | Small unmanned aerial vehicle remote sensing-based slope deformation detection and calculation method |
| CN112113542A (en)* | 2020-09-14 | 2020-12-22 | 浙江省自然资源征收中心 | Method for checking and accepting land special data for aerial photography construction of unmanned aerial vehicle |
| CN114998536A (en)* | 2022-05-31 | 2022-09-02 | 广州市城市规划勘测设计研究院 | Model generation method and device based on novel basic mapping and storage medium |
| CN116883604A (en)* | 2023-08-02 | 2023-10-13 | 中色蓝图科技股份有限公司 | Three-dimensional modeling technical method based on space, air and ground images |
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | Structural health monitoring and inspection of dams based on UAV photogrammetry with image 3D reconstruction | |
| CN110453731B (en) | A dam slope deformation monitoring system and method | |
| CN113611082B (en) | Unmanned aerial vehicle railway slope monitoring and early warning system and method | |
| US20140336928A1 (en) | System and Method of Automated Civil Infrastructure Metrology for Inspection, Analysis, and Information Modeling | |
| CN106292717B (en) | A kind of full-automatic information acquisition aircraft | |
| Yeh et al. | Modeling slope topography using unmanned aerial vehicle image technique | |
| CN113744393A (en) | Multi-level slope landslide change monitoring method | |
| CN113989670B (en) | Method for quickly extracting height of forest obstacle of power grid power transmission line | |
| CN114755693B (en) | Infrastructure facility measuring system and method based on multi-rotor unmanned aerial vehicle | |
| CN117722969A (en) | Hydropower station terrain deformation monitoring method based on four-dimensional hydropower station model | |
| CN114415203A (en) | Slope section monitoring and early warning system combining Beidou positioning and unmanned aerial vehicle radar | |
| CN112802004A (en) | Portable intelligent video detection device for health of transmission line and tower | |
| CN115762067B (en) | Landslide monitoring system based on laser point cloud and video data fusion | |
| CN116642536A (en) | Breakwater structure safety monitoring and early warning system based on multi-source data | |
| CN119555034A (en) | A building surveying method based on drone | |
| Thuse et al. | Accuracy assessment of vertical and horizontal coordinates derived from Unmanned Aerial Vehicles over District Six in Cape Town | |
| CN114440769B (en) | Multi-measuring-point three-dimensional displacement measuring method and measuring system | |
| CN114593713A (en) | A method and system for terrain inversion in tidal flat vegetation area | |
| CN119374566A (en) | A method and system for engineering surveying and mapping based on unmanned aerial vehicle remote sensing | |
| CN118429902A (en) | Railway infrastructure monitoring method and system based on unmanned aerial vehicle technology | |
| CN112228289A (en) | Apparatus and method for non-destructive in situ testing of windmill blades using penetrant dyes | |
| CN116758442A (en) | A transmission line data processing method based on laser scanning technology | |
| Adi et al. | Measurement of railway ballast deficiency using UAV drone and total station by graphical, statistical, and volume comparison | |
| Markovic et al. | Application of modern technologies in assessing facade condition of building structures | |
| Yusoff et al. | Beach volume measurement on variation of UAV altitude mapping |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20240802 | |
| RJ01 | Rejection of invention patent application after publication |