Application of Three-dimensional Laser Scanning Technology in the Teaching Practice of Surveying and Mapping of Ancient Buildings

Yapeng DUAN, Junli LIU, Zhendong LIU, Dingtao CAI, Shaomin NI

College of City Construction, Jiangxi Normal University, Nanchang 330022, China

Abstract Based on the study of the application of three-dimensional laser scanning technology in ancient building surveying and mapping, this paper briefly describes the working principle and flow of three-dimensional laser scanning technology. Based on the practical application, this paper puts forward the discussion of related problems and matters needing attention. This has a certain reference significance for the study of new technology in surveying and mapping of ancient buildings.

Key words Three-dimensional laser, Scanning technology, Surveying and mapping of ancient buildings

1 Introduction

At present, there are practical courses of ancient architecture surveying and mapping in the teaching system of architecture department in many architecture colleges. The purpose of its teaching is to enable students to understand the relevant knowledge of ancient buildings through the actual surveying and mapping of each part of ancient buildings, to master the methods of ancient architecture design, and teach the relevant technical contents of ancient architecture protection. So far, the manual surveying of ancient buildings has some limitations. Generally, we can only measure the length, width and heightof buildings, and need to contact with the main body of the survey, which is not conducive to the protection of ancient buildings. The three-dimensional laser scanning technology can obtain the three-dimensional point cloud information and form a three-dimensional model without touching the surface of the object. This will bring little impact on ancient buildings, so this technology has a significant advantage. In recent years, many schools have begun to use the combination of three-dimensional laser scanning technology and manual surveying and mapping to survey and map ancient buildings, so that students can master this new technology and method in practice, which can be said to be an innovative course. The development of three-dimensional laser scanner provides a new technical means for obtaining spatial information. In the application of ancient building survey, the traditional single point data acquisition is changed into continuous automatic data acquisition, which greatly improves the efficiency and accuracy of measurement[1]. This paper makes a preliminary discussion and research on the application of three-dimensional scanner in ancient building surveying and mapping.

2 Overview of three-dimensional laser scanning technology

As a three-dimensional measurement technology in the late last century, three-dimensional laser scanning technology undoubtedly brings together a variety of modern technologies. And it has become an important tool for obtaining spatial information data. Through the method of high-speed laser scanning measurement, it obtains the array geometric image data of the object or toposurface in the form of the collection of sampling points of the measured object, which is called "point cloud". A large number of spatial point information can be collected quickly, which provides a new technical means for the rapid establishment of three-dimensional image model of objects. It can go deep into any complex field environment for scanning operation, and directly collect the three-dimensional data of various entities into the computer, so as to quickly reconstruct the target line, surface, object, space and three-dimensional model and other mapping data. At the same time, the three-dimensional laser point cloud data collected by it can also be used to carry out a lot of post-processing work, such as analysis, simulation, display, monitoring, and virtualization[2]. All the collected three-dimensional point cloud data and three-dimensional modeling data can be converted for a variety of different engineering software directly through the standard interface format. Three-dimensional laser scanner is small and convenient, accurate and efficient, safe and stable, and has strong operability. It can establish a detailed and accurate three-dimensional image of the area of interest in a few minutes and provide accurate quantitative analysis. It is widely used in a variety of industries, such as rapid establishment of urban model, mapping and restoration of ancient buildings, protection of cultural relics, construction of complex buildings, geological research, building deformation monitoring and safety applications.

2.1 Scanning principle of three-dimensional laser scannerThe three-dimensional laser scanning system is mainly composed of host computer, power supply, control and data storage computer, reference point target and triangular bracket. The host device of the scanner has a horizontal 360° and vertical 270° scanning field of view in the range of 1-100 m. The surface accuracy of the scanning model is±2 mm and the scanning speed is 1 800 points/s in the range of 50 m. If conditions permit, it can also be equipped with electronic tachometer and means of transport to make it more effective[3]. The three-dimensional laser scanning measurement system is mainly composed of a linear laser light source, a color camera, a rotatable filter controlled by a motor, a control circuit board, a computer and the corresponding software. The stepmotor drives the line structure laser light source to do the stepping rotation, so that the surface of the object is scanned by the light, thus realizing the three-dimensional measurement of the whole measured surface. By using the mapping technique to match the color information of the object to the three-dimensional data of each measured point, the color three-dimensional information of the whole measured object can be obtained[4-5].

2.2 Measuring principle of three-dimensional laser scanner

Three-dimensional laser scanning measurement generally uses the internal coordinate system of the instrument. The X axis is in the transverse scanning plane, the Y axis is perpendicular to the X axis in the transverse scanning plane, and the Z axis is perpendicular to the transverse scanning plane. As a result, the coordinates calculation formula of three-dimensional laser measuring point, namely "point cloud", can be obtained. That is, the spatial position information is determined by the distance (d) between the target point and the instrument, the observed value (α) at the transverse scanning angle of the laser pulse and the observed value (β) at the longitudinal scanning angle. All the points in the "point cloud" have measured values, which have a one-to-one corresponding relationship with the XYZ coordinates of the internal coordinate system of the instrument[3]. In the computer screen display, the three-dimensional shape of the scanned object is composed of a certain density of points, namely "point cloud".

2.3 Point cloud splicingIf the measured object is large, the object is generally scanned in the form of block and multi-site. Then the object is measured or the point cloud is spliced in modeling. In order to ensure the accuracy of splicing, the images scanned by adjacent sites should have a certain degree of overlap. The experimental results show that when the shape of the object is complex, the degree of overlap should be at least more than 10%. Because of the scanning angle and position, if the feature points are directly used for splicing, the splicing accuracy of the image can only be about 10 mm, which directly affects the reliability of object measurement and modeling. The splicing accuracy will be improved by using the method of fitting the feature plane and then adding the point cloud constraint on the basis of the feature plane.

2.4 Establishment of three-dimensional modelThe scanned point cloud is transformed into a three-dimensional model in the usual sense. System software should have at least the following conditions: the commonly used three-dimensional model components, such as cylinder, sphere, tube, box and other solid geometric shapes; the point cloud matching algorithm corresponding to the model component; TIN polygon algorithm on geometric surface. When three-dimensional modeling is carried out, the automatic segmentation processing tool provided by the system software can be used to extract part of the scanned point cloud image and form an object or a part of the object for automatic matching processing. However, this automatic matching method is only suitable for those target solid components which are consistent with the common geometric objects contained in the software. It is invalid for target solid components that cannot be decomposed into common geometric solid. At this time, it is necessary to construct TIN polygons in the corresponding point set to simulate irregular surfaces. Finally, using the texture function of Ployworks software, the mapping technology is used to match the color information of the object captured by the CCD camera to the three-dimensional data of each measured point, so as to get the real three-dimensional image of the object[6]. You can also output three-dimensional models built in Ployworks or Cyclone software to AOTUCAD for editing, rendering, or printout.

3 Application in surveying and mapping of ancient buildings

3.1 Characteristics of ancient buildingsAncient Chinese architecture embodies the air of ancient culture and art. And with the continuous changes of the times, ancient architecture has different characteristics in structure and aesthetic performance. In the ancient architecture of our country, its external form is mainly composed of roof, major structure and foundation. And in the ancient buildings, the structure of the column, foundation and roof is very complex. On the other hand, there are a variety of styles in ceilings, doors and windows,etc., and the composition of the patterns is extremely complex and vivid, and the connotation expressed is also very rich. Therefore, the existence of the above factors will undoubtedly increase the difficulty of surveying and mapping of ancient buildings. At the same time, with the application of three-dimensional laser scanning technology, the scanning accuracy of ancient buildings can be greatly improved. It can even achieve the single point scanning accuracy of millimeter level, and the scanning spacing can reach the submillimeter level. So the accurate scanning measurement of ancient buildings can be realized, and the scanning information can be directly input into the computer. In addition, the three-dimensional laser scanning technology will not have direct contact with the ancient buildings in the process of practical application, so it will not damage the ancient buildings, so that the ancient buildings will be further protected. Therefore, its application in ancient building surveying and mapping can achieve twice the result with half the effort.

3.2 Application examples

3.2.1Preparation. Before surveying and mapping with three-dimensional laser scanner, we need to survey the area and collect data to grasp the key points of the survey so as to ensure that the survey work can be carried out smoothly. At the same time, according to the characteristics of the survey area and the accuracy requirements of the project, the exact position of the scanner is defined in detail. In principle, it is necessary to ensure that the number of scanning sites is the least, the scanning content and range is the largest, the redundant data is the smallest, the target setting is the most reasonable, and the image is the clearest.

3.2.2Surveying and mapping methods. Generally speaking, when using the ground three-dimensional laser scanning technology to scan the ancient buildings, it is necessary to set up scanners in different sites to obtain the point cloud data inside and outside the ancient buildings. Then, the point cloud data after preprocessing is obtained by splicing, denoising and other steps. On this basis, the plan view, elevation view, section view and three-dimensional model of the ancient buildings are drawn.

(i) Three-dimensional laser scanning. Control of survey: In the later stage, the method of rear view orientation is used to splice the point cloud. Therefore, it is necessary to have a measurement control over the scanning sites. The surveying is carried out in accordance with the requirements of the mapping control points of 1: 500 topographic map. Point cloud acquisition: The laser scanner and the rearview target ball are set up on the measuring site and the rearview point, respectively. The point cloud data inside and outside the ancient buildings are obtained in turn. Image collection: The RIEGLVZ-300 laser scanner is used in this project, it can obtain images automatically after a single-site point cloud scanning is completed. In addition, it is also necessary to take pictures of the plane, vertical face, and section of the building in different directions and parts in order to provide texture information for the three-dimensional model of the ancient building. (ii) Data preprocessing. The preprocessing of point cloud data includes the removal of noise points and redundant data, point cloud splicing and point cloud coloring. Due to the reasons of the scanner itself, it will produce certain noise points, and it is also necessary to eliminate the point cloud data which is not related to the ancient building surveying and mapping area. Through registration, the single-site data obtained by field scanning is classified into the same coordinate system, and is combined into the whole point cloud result of one survey area. Then, the preprocessing of the point cloud data is completed by coloring the point cloud with the image registered with the point cloud. (iii) Data production. At present, the data that need to be applied to the protection of ancient buildings are plan view, elevation view, section view, three-dimensional model and so on. And plan view, elevation view, section view are completed by PointCloud software through the collection and editing of feature points, lines and faces. The three-dimensional model mainly uses the relevant tools provided by Geomagic software, and the obtained point cloud data are spliced automatically. Image matching is also automatically completed by the software. By analyzing and making the main geometric structure of the building, the geometric model of the building is established, and then the model is imported into 3-D Max software to complete the texture mapping, so that the fine three-dimensional model of the ancient building can be obtained. Among them, there are point cloud holes in some ancient building areas, which are filled by the function of Geomagic software. The main principle is to fill the hole by linear or curved surface interpolation with the points around the hole.

4 Discussions on related problems

As a technical means, three-dimensional laser scanner has its own technical characteristics in the measurement of ancient buildings. According to the practical application of three-dimensional laser scanner, the following problems need to be paid attention to and discussed.

4.1 Careful surveying and reasonable distribution of sites before scanningThe laser of a three-dimensional laser scanner works in the range of visible light and obscures part of the space. It needs to be solved by changing the working position of the instrument, namely the so-called "changing site". In general, to get complete data on a building, you need 3 to 5 sites. In complex cases, more sites are required, otherwise there will be problems such as blind spots in scanning, and black holes in data splicing. However, if too many sites are selected, the scanning time will be increased and the measurement efficiency will be reduced. At the same time, in the data post-processing, the accumulation error of splicing is increased. Therefore, when the conditions permit, the measurement should be carried out at the survey site as far as possible before the survey, and the scanning site[3]should be determined according to the location, shape and size of the building, the site conditions and the spatial information to be obtained.

4.2 Selection and optimization of scanning parametersAfter selecting test site and completing the erection and adjustment of the device, the scanning parameters should be set correctly before scanning in order to complete the scanning operation efficiently. The determination of scanning resolution and scanning distance directly affects the scanning accuracy and efficiency. The resolution is generally determined according to the scanned object and the spatial information that needs to be obtained. For example, it is generally 25 mm for the common wall of a building, while it is generally not larger than 8 mm for the characteristic parts such as eaves, boundaries and some components. Similarly, the determination of the scanning range, that is, the average distance between the scanning object and the instrument, will also directly affect the scanning efficiency. The scanning range can be determined by different methods, such as instrument detection, and visual inspection. If field conditions permit, images of scanned objects can be taken through a digital camera integrated with the device itself. The scanning area is selected on the image, and the scanner scans automatically according to the parameters set in the software environment.

4.3 Layered denoisingAfter obtaining the scanned "point cloud" map of each site, it is generally necessary to classify the "point cloud" layer by layer. In particular, it is necessary to set up a "noise" layer so that trees, vehicles, pedestrians and other unrelated data do not participate in the calculation, and we can call the "point cloud" data through the management of the layer. By layering the data, the data denoising can achieve the best effect.

4.4 Coordinates systemThe three-dimensional laser image scanner provides the internal coordinates of the instrument, and its origin is the laser emission center point. At present, for airborne laser scanning system, the inertial navigation system and satellite positioning system can be used to determine the spatial coordinates of its launch point, so as to solve the problem of geodetic coordinates of ground objects. However, for the three-dimensional laser image scanner, it can only be solved by coordinate conversion. That is, the geodetic control point is established on the surface of the scanned object, and the instrument coordinate data is transformed into geodetic coordinate data by using the control point. How to improve the resolution of control points on the point cloud model and improve the measurement accuracy of "point cloud" needs to be further studied[7].

5 Conclusions

The three-dimensional laser scanning technology can be used to complete the surveying and mapping of ancient buildings efficiently and accurately. This breaks through the traditional surveying and mapping methods of technicians and provides great convenience for the surveying and mapping and restoration of ancient buildings. In this paper, the working principle of three-dimensional laser scanning technology is summarized, its application in ancient building surveying and mapping is expounded, and some existing problems are analyzed. Through the research, it is known that the three-dimensional laser scanning technology will be widely popularized and applied in the future.