阅读说明：本技术 一种在道路测设中的无人机影像匹配点云的测量方法 (Method for measuring unmanned aerial vehicle image matching point cloud in road survey ) 是由 李涛 于 2019-04-26 设计创作，主要内容包括：本发明公开了一种在道路测设中的无人机影像匹配点云的测量方法,包括以下步骤：(1)进行外业航测；(2)进行基站坐标采集,以及对既有硬化路面边线坐标采集；(3)点云与检测点精度分析；(4)进行任意断面提取、数据冗余处理。本发明与现有技术相比的优点是：本发明的无人机影像匹配点云技术代替传统断面测量,实现任意断面提取和土方量计算的技术与方法,确定不同阶段具有最优性价比的测绘产品和测绘作业方法。(The invention discloses a method for measuring unmanned aerial vehicle image matching point cloud in road survey, which comprises the following steps: (1) carrying out field operation aerial survey; (2) acquiring coordinates of a base station and coordinates of side lines of the existing hardened pavement; (3) analyzing the precision of the point cloud and the detection points; (4) and carrying out any section extraction and data redundancy processing. Compared with the prior art, the invention has the advantages that: the unmanned aerial vehicle image matching point cloud technology replaces the traditional section measurement, realizes the technology and the method of arbitrary section extraction and earth volume calculation, and determines surveying and mapping products and surveying and mapping operation methods with optimal cost performance at different stages.)

1. A measuring method of unmanned aerial vehicle image matching point cloud in road survey is characterized in that: the method comprises the following steps:

(1) Performing field aerial survey, matching point clouds through an AgiSoftPhotoshop image by adopting an MAVinci Desktop image rapid air-to-three encryption method, and then performing point cloud noise filtering and removing; the field aerial survey equipment adopts a sirius image-control-free unmanned aerial vehicle, a survey area range line is led into MAVinci Desktop flight plan software, the GSD (ground sampling interval) is set to be 3cm, a banded flight plan and a self-adaptive terrain relief mode are used, and the relative height is 120 m; the course overlapping degree is 80 percent, and the side overlapping degree is 65 percent; the average flight length of each flight is 3km, the bandwidth is 200m, and the endurance time is 30 min;

(2) acquiring coordinates of a base station and coordinates of side lines of the existing hardened pavement;

(3) analyzing the precision of the point cloud and the detection points;

(4) and carrying out any section extraction and data redundancy processing.

Technical Field

The invention relates to a measuring method, in particular to a measuring method of unmanned aerial vehicle image matching point cloud in road survey.

Background

Cross-sectional measurements refer to the operation of measuring the relief profile of the pile perpendicular to the midline. Then drawing a cross section diagram for the design of roadbed, side slope and special structure, calculation of earth and stone space and construction lofting, and the cross section measurement must be carried out in the line engineering measurement of highway, railway and the like. Because large-scale line engineering roadbed is wide, design accuracy requires highly, the length and breadth section density is big, especially in hills or mountain area surface difference of elevation change is big, current measuring area of inconvenience, it is limited to use traditional method to measure and gather the quantity of elevation point, the cross section work efficiency is lower, in case change the line still need to advance the field again and measure, not only the manufacturing cost is high, and seriously influence the design time limit, can not adapt to current social development.

Road survey and design are important bases for calculating earth and stone volume and carrying out line design, cost budget and engineering construction, but the traditional surveying and mapping method adopted under the condition of complex terrain has a plurality of problems and defects: 1. the traditional surveying and mapping operation period is long, the working labor intensity is high, the new approach measurement is needed once the line is changed, and the design period is seriously influenced; 2. the current inconvenience usually under the complicated topography condition, the operation personnel are difficult to get into the operation scene, and the security can not be guaranteed, and the density of gathering the point location is not enough simultaneously, and the cubic meter precision of earth and stone is relatively poor, also seriously influences engineering cost.

In recent years, along with the high-speed development of economy in China, the unmanned aerial vehicle shows a thriving scene in various industries, and particularly in the geographic information industry, the aerial photograph of the unmanned aerial vehicle brings a leap-over leap to the traditional surveying and mapping means. As a professional measurement type unmanned aerial vehicle, the PPK, the image-free control technology and the oblique photography technology have greatly improved image positioning accuracy and elevation accuracy, and meanwhile mainstream unmanned aerial vehicle processing software such as AgiSoft PhotoSacan, PIX4D Mapper and Smart 3DCapture support the image matching point cloud function. The point cloud data is used as a brand-new geographic information product, and the huge application value of the point cloud data is gradually shown in the field of engineering measurement.

Therefore, developing a method for measuring the point cloud of the image matching of the unmanned aerial vehicle in the road survey becomes a problem to be solved by those skilled in the art.

Disclosure of Invention

The invention provides a method for measuring the unmanned aerial vehicle image matching point cloud in road surveying and setting, aiming at solving the defects in the prior art.

The above object of the present invention is achieved by the following technical means: a method for measuring unmanned aerial vehicle image matching point cloud in road survey comprises the following steps:

(1) performing field aerial survey, matching point clouds through an AgiSoftPhotoshop image by adopting an MAVinci Desktop image rapid air-to-three encryption method, and then performing point cloud noise filtering and removing; the field aerial survey equipment adopts a sirius image-control-free unmanned aerial vehicle, a survey area range line is led into MAVinci Desktop flight plan software, the GSD (ground sampling interval) is set to be 3cm, a banded flight plan and a self-adaptive terrain relief mode are used, and the relative height is 120 m; the course overlapping degree is 80 percent, and the side overlapping degree is 65 percent; the average flight length of each flight is 3km, the bandwidth is 200m, and the endurance time is 30 min.

(2) Acquiring coordinates of a base station and coordinates of side lines of the existing hardened pavement;

(3) Analyzing the precision of the point cloud and the detection points;

(4) extracting any section and processing data redundancy.

Compared with the prior art, the invention has the advantages that: the unmanned aerial vehicle image matching point cloud technology replaces the traditional section measurement, realizes the technology and the method of arbitrary section extraction and earth volume calculation, and determines surveying and mapping products and surveying and mapping operation methods with optimal cost performance at different stages. And an unmanned aerial vehicle image matching point cloud technology is utilized to provide a brand-new solution for measuring the cross section of a highway or measuring the road marking and characteristic line of an enclosed area of a highway in a specific working environment.

Drawings

Fig. 1 is a flow chart of the unmanned aerial vehicle image matching point cloud work in the invention.

FIG. 2 is a schematic diagram of a conventional sectional measurement ground elevation point according to the present invention.

Fig. 3 is a schematic diagram of the ultra-high density unmanned aerial vehicle image matching point cloud in the invention.

Fig. 4 is a schematic diagram of the unmanned aerial vehicle image matching point cloud (after noise point removal) in the invention.

FIG. 5 is a schematic diagram of arbitrary cross-sectional design (cross-hatching) under a three-dimensional environment in space according to the present invention.

Fig. 6 is a schematic diagram of the ground elevation point obtained by matching the point cloud with the unmanned aerial vehicle in the present invention.

FIG. 7 is a schematic view of the overlay analysis of two cross-sectional lines according to the present invention.

Fig. 8 is a schematic diagram of DTM grid established by actually measured traverse lines in the present invention.

Fig. 9 is a schematic diagram of the DTM mesh established by using point clouds in the present invention.

FIG. 10 is a schematic diagram of the target point arrangement and detection in the present invention.

FIG. 11 is a schematic diagram of feature point inspection in the present invention.

FIG. 12 is a diagram illustrating the registration check of vector data in the present invention.

Detailed Description

The present invention will be described in more detail with reference to examples.

As shown in fig. 1, the method for measuring the unmanned aerial vehicle image matching point cloud in the road survey comprises the following steps:

(1) performing field aerial survey, matching point clouds through an AgiSoftPhotoshop image by adopting an MAVinci Desktop image rapid air-to-three encryption method, and then performing point cloud noise filtering and removing; the field aerial survey equipment adopts a sirius image-control-free unmanned aerial vehicle, a survey area range line is led into MAVinci Desktop flight plan software, the GSD (ground sampling interval) is set to be 3cm, a banded flight plan and a self-adaptive terrain relief mode are used, and the relative height is 120 m; the course overlapping degree is 80 percent, and the side overlapping degree is 65 percent; the average flight length of each flight is 3km, the bandwidth is 200m, and the endurance time is 30 min.

(2) Acquiring coordinates of a base station and coordinates of side lines of the existing hardened pavement;

(3) Analyzing the precision of the point cloud and the detection points;

(4) extracting any section and processing data redundancy.

The method takes a certain highway surveying project as an example, adopts the sirius image control-free unmanned aerial vehicle aerial photography equipment to obtain a high-resolution image, utilizes the image matching point cloud technology and the Lidar point cloud processing technology to manufacture and generate dense ground elevation data, and further replaces the section measuring work which is time-consuming and labor-consuming in the highway surveying process, so that the operation efficiency is greatly improved, the production cost is reduced, and great convenience is brought to road rerouting, any section design and BIM system application. The details are as follows:

the engineering is a cross section measurement project carried out at a certain expressway determination stage in the Laiwu city, and the line design mileage is 85km, the average altitude is 210m, the lowest altitude is 170m, and the highest position is 280 m. Most of the measuring area is positioned in a micro-hill area and a heavy-hill area, and the landform is relatively broken; the earth surface of the survey area is mainly sparse orchard and short crops, most of the earth surface is exposed, and the elimination of cloud noise points of the flying spots of the unmanned aerial vehicle is facilitated. The technical requirements are as follows: and acquiring transverse coordinates according to the corresponding pile number of the middle pile, and acquiring elevation coordinates above the ridge and below the ridge at the terrain slope transformation position.

The field aerial survey equipment adopts a sirius image-control-free unmanned aerial vehicle, a survey area range line is led into MAVinci Desktop flight plan software, GSD (ground sampling interval) is set to be 3cm, a banded flight plan and a self-adaptive terrain relief mode are used, and the relative height is 120 m; the course overlapping degree is 80 percent, and the side overlapping degree is 65 percent; the average flight length of each flight is 3km, the bandwidth is 200m, and the endurance time is 30 min.

Unmanned aerial vehicle image matching point cloud work flow chart is shown in fig. 1.

The cross section measurement is verified by adopting a traditional measurement mode, the line length is manually acquired by 5km, and the number of the cross section points is about 14800. In the present invention, only three sections are selected for convenience of display, and more sections are similar to the three sections.

The conventional profile measures the ground elevation point as shown in fig. 2.

The ultra-high density unmanned aerial vehicle image matches the point cloud, as shown in fig. 3.

The unmanned aerial vehicle image is matched with the point cloud (after noise is removed), as shown in fig. 4.

And designing an arbitrary section (drawing a section line) in a three-dimensional environment in space, as shown in figure 5.

And (5) acquiring a ground elevation point by using the unmanned aerial vehicle matching point cloud, as shown in figure 6.

As shown in FIG. 7, the two section lines are basically matched according to the superposition analysis of the two section lines, but a slight deviation exists, the point cloud point location of the main reason is not completely consistent with the point location of the artificial actual measurement, in addition, the manual collection of the section points also has a trade-off, and the incomplete matching of the section lines generated by the two modes also belongs to a normal phenomenon.

Calculation and precision analysis of earth volume:

and (3) earth volume calculation: as shown in fig. 8 and 9, the DTM triangulation files are generated by using the two different data acquisition modes, and compared by using the earth calculation function of the CASS software DTM method, the calculation results of the three sections from K44+400 to K44+460 are excavation amount 366.6m3, filling amount 265.7m3, filling and excavating balance 99.9m3 and ground surface area 3600m2, and the overall influence on the ground surface height is about 3 cm; the detection range is expanded, thirty actual measurement sections from K44+220 to K44+500 are compared with the unmanned aerial vehicle matching point cloud, and the calculation results are that the total excavation amount is 2723.8m3, the total filling amount is 2490.4m3, the excavation balance amount difference is about 233.4m3, the ground surface area is 21355m2, and the overall influence on the ground surface height is about 11 cm.

Comparing section measurement schemes:

because the elevation point acquisition means of the two modes are completely different, the traditional section measurement has higher precision on a specified section line, but the DTM model constructed by lacking data between two section lines has low precision, and the measurement precision has larger relation with the section piling at the position of the geodesic line and the working experience of operators; the unmanned aerial vehicle matching point cloud mode is more inclined to the square grid method, the point cloud density of the unmanned aerial vehicle matching point cloud mode basically reaches 2m x 2m (the point cloud density on the bare surface can reach 0.5m x 0.5 m), and a very accurate earth volume result can be obtained under the condition that the point cloud precision is reliable (see table 1: comparison of cross section measurement schemes).

Table 1: comparison of Cross-sectional measurement protocols

Precision inspection and quality control measures:

a) target point inspection

In order to test the accuracy of the aerial survey result of the unmanned aerial vehicle, ground detection target marks are drawn at the positions of hardened pavements in a test area at intervals of about 500m, the size of each target is 30cm multiplied by 30cm, and independent GPS static network observation and four-equal leveling measurement are carried out. As shown in fig. 10.

The maximum difference value between the point cloud elevation and the actually measured elevation of the target point is 0.118m, and the error in the average elevation is +/-0.076 m.

Unit for poor elevation precision of target detection: m is

b) And (5) checking the characteristic points, as shown in figure 11.

Selecting 2280 points of hardened road sideline elevation points on two sides of the road, calculating the error in elevation to be +/-0.107 m (eliminating four points with gross errors of bridges and culverts), wherein 1363 points with the height difference less than or equal to 10cm account for 59.78%; the height difference is at 638 points of 10-20 cm in total, and accounts for 27.98%. Considering the influence of the self measurement precision of the network RTK, the comprehensive evaluation of the unmanned aerial vehicle aerial photography elevation precision is about 3 times GSD, the elevation precision of +/-10 cm can be basically achieved, and the precision requirement of section measurement can be met.

c) And (4) checking the vector data of the existing large-scale topographic map in a registration way, as shown in figure 12.

Vector data registration inspection is the most common inspection means for surveying and mapping production units and user units, and is also the most acceptable and intuitive inspection mode for users.

The foregoing detailed description is given by way of example only, and is provided to better enable one skilled in the art to understand the patent, and is not intended to limit the scope of the patent; any equivalent alterations or modifications made according to the spirit of the disclosure of this patent are intended to be included in the scope of this patent.