Terrestrial Backpack Laser Scanner Usage in Mobile Surveying: A Case Study for Cadastral Surveying

1. Introduction

Different methods are employed for topography measurements; and generally, aircrafts are preferred when the study site has a coverage field of kilometers. Historically, surveying studies have been performed with theodolites at the beginning, within the more limited distances of the study area. Then, points are enabled to be easily positioned within a global coordinate system with satellite navigation systems. Even though such methods ensure high accuracy and precision for the measurement of points, it requires more time to acquire data with adequate density for digital elevation models (DEMs). Additionally, dense datasets containing millions of point measurements for high-resolution DEMs are now routinely generated using ground-based terrestrial laser scanners (TLSs) with millimeter-centimeter accuracy. After TLSs became ground based, significant developments have occurred in the acquisition of topographic data with mobile laser scanners (MLSs).

MLS has recently been employed to compensate for the shortcomings of TLS. Finally, laser scanners that can be carried or worn by an operator have become popular. These types of scanners are called personal laser scanners (PLS). Depending on how the device is handled, PLS devices are divided into two categories: handheld personal laser scanners (HPLS) and backpack personal laser scanners (BPLSs) [1]. PLS Systems do not need a tripod or tool for use as in TLS, and its most important feature is its high mobility that allows rapid data collection. These systems are capable of digitalizing complex 3D scenarios on the move without a global navigation satellite system (GNSS), thanks to simultaneous localization and mapping (SLAM) algorithms [2, 3] that are based mostly on the robotic operative system [4] for point cloud registration and map extraction [5]. PLS, which are frequently studied in forestry activities, can be utilized in many studies including tree-level data acquisition for forest inventory, measurement of leaf area index of broad fields, and forest route determination for recreation. Laser scanning approaches such as TLS and PLS were employed to collect fuel density data, which bridges the gap between the topography and canopy giving rise to severe canopy fires [6]. On the other hand, data acquired with TLS can require time and include incomplete data and the use of PLS can reduce these drawbacks. For the prediction of tree position and diameter, it was found that the PLS method is faster than TLS about 4.7 times [7]. Also, it is stated that PLS is a very practical and appropriate method to generate the tree inventory and to record different tree characteristics in forested areas where the quality of PLS data matches the necessity of the implementation [8]. It is stated that HPLS is an economical and practical technological solution for data collection, assessment, and processing of individual tree characteristics for forest inventory, generating digital terrain model, harvesting volume computation, and parameterization of forest growth models [9]. Outdoor environment surfaces may be highly uneven, covered with vegetation, and rarely “surround” an observer, and thus represent more tough implementations for reconstruction [10]. It is expected that HPLS will soon become popular in the field of forestry because it could be employed easily in forestry surveys and data could be collected at the required standards [11]. Combining the reliability of laser technique, data density provided specific to scanning systems, the flexibility of walking with a handheld object scanner, and collecting topographic data in complex terrain represent significant advances in laser scanning technology. Generally, HPLS presents significant advantages over existing technologies for the rapid measurement of complex topography with poor field of view and geomorphological study areas. The HPLS system presents high performance considering the enclosed environments where static terrain properties surround the sensor and ensure well-distributed, consistent laser returns to facilitate convergence in the processing algorithms. BPLS combines a single operator while walking when obtaining data with laser scanning technology and inertial measurement units (IMU), a microcomputer, a tablet, and accessories (e.g., connection cables and a battery) in a portable equipment that can be handled. BPLS is a novel type of portable lidar for which the surveyor is the wearable platform, and thus it has a high capacity in terms of accessibility and route choice [12]. The backpack lidar system was designed to automatically collect and register under-copy lidar data [13]. The measurement accuracy and error sources have not been systematically explored for this system [12]. Today, BPLSs are currently used in various fields, from forestry studies [1, 5, 14] to building information model studies [15]. Compared with TLS, backpack lidar is generally lighter and more portable, and it can obtain much higher quality 3D point clouds in forest with different vegetation structures. In some areas where there is limited access, backpack laser scanning can be integrated with other measurement techniques, especially unmanned aerial vehicles (UAVs) [13].

It is possible to carry out cadastral renewal studies with classical methods. However, if cadastral studies are carried out by establishing a polygon network with classical methods, there are some limitations, such as the requirement that the points in the polygon network see the previous and next ones, the requirement to establish extra points during the work if needed, and the inability to receive signals from a sufficient number of satellites if a polygon is established with GNSS. It is also not possible to obtain the dense point cloud with 3D coordinates provided by the backpack lidar system with classical methods.

In this chapter, the use of a backpack personal laser scanner (BPLS), a PLS system, to update cadastral surveyings is analyzed. For this purpose, accuracy analysis was performed considering the geodetic and lidar measurements. The rest of the chapter is organized as follows: advantages and disadvantages of PLS are specified in section two and the case study section presents a detailed description of the study area and the methodology of the procedures performed. In the conclusion section, it was assessed that backpack laser scanners can be employed to renew property boundaries in places such as the study area.

2. Advantages and disadvantages of personal laser scanners

Collecting the three-dimensional information of objects on earth with high accuracy, and keeping them in the digital environment is a vital task in surveying, object scanning, and mobile mapping. There are multiple ways to obtain three-dimensional information about objects, such as lidar or radar scanning, stereo vision, and imaging sonar. Which of these techniques will be appropriate to employ depends on some parameters such as the size of the study area, expected accuracy, cost, resolution, practicality, and data processing time.

Laser scanning is a surveying method used for mapping topography, vegetation, urban areas and infrastructure, ice, and other targets of interest [16]. According to the platform on which the system is built, laser scanning technology is categorized into airborne, unmanned aircraft systems, terrestrial, mobile, and personal systems [17]. In light of this information, TLS is a ground-based type of lidar built on a tripod. The first studies on the possibilities of applying TLS in laser scanning started around the year 2000. The MLS system, which is an alternative to TLS, is an efficient method since it could decrease occlusion problems and data acquisition time. In addition, appropriate measurement and rapid data collection methods are always preferred [9]. On the other side, MLS systems generally comprise of three primary components: laser scanner (LiDAR sensor), Global Navigation Satellite System (GNSS) receiver, and inertial measurement unit (IMU). The precision and accuracy of all three components and their synchronization affect the accuracy of the MLS data [18]. StreetMapper, which is the first commercial MLS system for surveying implementations, appeared in the market in 2006. Moreover, MLS systems used in surveying and robotics have different perspectives and highlights. The data acquired using the MLS technique is less accurate in comparison to TLS technique due to positioning error propagation in the point cloud produced by MLS. On the other hand, the topographic condition can be not suitable for vehicle motion [9].

To acquire the whole scene with a static scanner, it is necessary to increase the number of scanning stations. One advantage of the PLS compared to the static scanner is that the environment is acquired more easily. Someone can move around the objects of the scene. On the other hand, the PLS data has a lower density of points than the static scanner. Besides a lower point density, the PLS also present a higher measurement noise. The results are very satisfying and promising since the PLS data, which are acquired much faster than TLS data, are close to the reference data. HPLS and BPLS systems provide similar results [15].

The operator’s path directly affects the performance of the simultaneous localization and mapping (SLAM) algorithm and the quality of the collected data. Therefore, it should be well-planned and executed. When planning, the time to acquire a point cloud with PLS and cost are factors to consider in measurement planning. PLS point clouds are often georeferenced, as in TLS, by scanning targets of known diameter and location, such as an easy-to-detect and precisely measured sphere, within the point cloud.

The flexibility of the SLAM approach is added to the high accuracy and precision of the TLS systems, making it possible to enrich the final digital model with details undetectable by stationary tools due to the excessive intricacy of the paths [19]. The SLAM algorithm processes IMU and laser data from PLS to locate the scanner in an unknown environment and to register the whole 3D point cloud without the need for GNSS [20]. While rapid changes in sensor technology have led to developments, the most important development in PLS systems is the use of the SLAM algorithm instead of GNSS. The SLAM algorithm processes the IMU and laser data to position the scanner, making it suitable for use in forest-like areas where GNSS signals are absent or weak and distorted. However, current SLAM techniques include remarkable errors. The data acquired by PLS is less precise than TLS point cloud data due to the propagation of positioning errors within PLS point cloud data [21].

3. The case study

This study aims to process laser point data, which is one of the products of backpack laser scanning technology, and to produce the cadastral map using the point cloud. The study area shown in Figure 1 covers approximately 90 hectares within the borders of Karaağaç District, Edirne province, Türkiye. Karaağaç District has an open space and detached buildings in terms of construction type. In the study area, there is a low-rise construction type, the surroundings of the buildings are relatively open space, and the vegetation is protected (trees, bushes, etc.).

The backpack laser scanner used to obtain point cloud data is LiBackpack DGC50 shown in Figure 2. The laser scanning device is a bagged LiBackpack DGC50 device that can produce coordinated and colored point clouds, thanks to the Inertial Navigation System embedded inside. LiFuser-BP Data Processing Software is used.

LiBackpack DGC50 is an operator-only wearable LIDAR 3D scanning system that is suitable for both indoor and outdoor scanning and can produce three-dimensional coordinates with high accuracy. LiBackpack DGC50, which has a panoramic imaging system, produces point cloud data with high accuracy, thanks to GNSS and SLAM technologies. Coordinate information can be obtained even if the GNSS signal is insufficient in some parts of the scanning route. This equipment can be widely employed in indoor and outdoor measurement, obtaining underground information, tunnel engineering, digital mining, counting forestry resource, building information systems, and other domains. The technical features and parameters of the hardware are shown in Table 1.

Before starting data collection, a reconnaissance was made in the study area. According to this, sessions were determined on a 15-minute session-time basis. The total number of sessions in the study area was 63 and the study area scanning was carried out in 7 days. Figures 3 and 4 show an example of a colored and coordinated point cloud in the study area for a session.

LiBackpack DGC50 provides an output of the point cloud in the form of a raw data file with a “.bag” extension. When the “.mp4” extension video taken simultaneously during scanning is integrated into the point cloud, a point cloud is produced in the real RGB values of the scanned region. In the study, the point cloud obtained on a session basis was processed in the LiFuser-BP program and turned into a point cloud with the “.laz” extension. Then, it was converted into a file with the extension “.las” with the help of cloud compare software. Since the software does not decompress the .laz extension, the conversion process was performed using cloud compare software. The drawing was made using Trimble Business Center (TBC) software.

Point cloud data was produced almost completely and with high accuracy in the selected study area in Karaağaç District. In the scans made in the study area, almost every corner of all the buildings could be drawn. Although it is difficult to distinguish the structures such as coal sheds, sheds, etc. adjacent to the buildings from the main building, the Lidar device’s ability to produce point cloud data in color has made it easier to detect the separation of the buildings. Since GNSS data could be obtained continuously and uninterruptedly, the scanning process was carried out by producing real-time coordinates. Coordinated point cloud data was produced directly for each session. The point cloud merging process was not performed. The study was conducted on a session basis. After each session was drawn separately, the drawings were combined in the CAD environment. The combined view of the sessions in NetCAD software is given in Figure 5.

The study area chosen in Karaağaç District is a region with detached buildings. Since GNSS works at full efficiency, scanning can be done with LiBackpack DGC50 without establishing any polygons. Polygon points were established with GNSS to measure the coordinates of the detail points determined for accuracy control. The coordinates of 25 detail points were measured from the polygon points with a total station. Table 2 shows the differences between the classical measurement (with total station) and Lidar measurement for the study area. The maximum difference detected is below 7 cm. It has been concluded that the application carried out in the working area with a backpack laser scanner to determine property boundaries remains within the error limits of the regulations. Standard deviation values were calculated with Eq. (1) for X and Y as 0.031 m and 0.025 m, respectively.

In eq. (1), n is the number of points, xi and x̂ represent the measured point and arithmetic mean, respectively. Since the study was about determining property boundaries, the accuracy of the Z coordinate was not taken into account.

For further analysis, the Similarity transformation (Eq. (2)) was applied considering classical measurement reference to the coordinates of each point.

In the equation, T_x and T_y represent the shift in the direction of the X and Y axis, R_α represents rotation on the X and Y axis, and λ refers to the scale factor. If equality 1 is explicitly written, the following equation is obtained:

where x and y are the first system coordinates, X and Y are the second system coordinates, and a, b, c, and d are the similarity transformation parameters. Two-dimensional similarity transformation parameters between classical measurements and lidar measurements are given in Table 3, and residuals for 25 points are given in Table 4. The standard deviation values calculated for the residues are 0.0254 for Y and 0.02 m for X.

4. Conclusion

Backpack laser scanners represent a significant advance in spatial data collection, combining the reliability of laser techniques with the flexibility of measurement. The backpack laser scanning method is faster than the classical method and requires less labor. For the purpose of cadastral renewal studies, a backpack laser scanner, which can produce data at a density equivalent to other scanning systems, was used and evaluated. It was successful in identifying structures such as walls and wires built to determine parcel boundaries. However, it is a fact that the number of noisy data will increase in places with high-rise buildings and narrow streets. Additionally, UAVs can be used to obtain data in areas with limited physical access, such as backyards. In this way, data integration and complete data for the region will be provided.

Listed below are the advantages of backpack laser scanners.

1. In areas with detached and low-rise buildings, BPLS reduces processing steps as they provide coordinate data directly.

2. Scanners that produce colored point clouds provide great convenience to the operator making the drawing in separating the details of complex structures.

3. Most of the revisions can be made by examining the relevant session without fieldwork again if there is no distortion in the point cloud.

4. Collecting data from the field in the classical method requires more labor. In this respect, BPLS provides flexibility and practicality to the operator during work.

5. Although office work is longer than the classical method, BPLS is more productive when the entire process time is considered.

6. BPLS has some important advantages over TLS and MLS when making field measurements, such as not using a tripod and seeing the object from different directions.

Listed below are the disadvantages of backpack laser scanners.

1. A polygon network should be established before starting measurements in areas where GNSS cannot be used with the desired accuracy in the scanning process.

2. It is affected by limitations such as adjacent buildings and high-rise buildings that make data acquisition difficult.

3. BPLS devices are relatively expensive, so they are not preferred in small-scale fields.

It is clear that as a result of the development of sensor technology in the future, BPLS will find much wider use. Particularly, mobility and speed will make BPLS a practical measurement tool for acquiring three-dimensional data in challenging environments. It has been concluded that backpack laser scanners can be used in the production of cadastral maps in regions such as in this study area.

Acknowledgments

The application data of the study were provided by Halit Enes DEMIR, a Geomatics Engineer who is a master’s student of Gebze Technical University.

Conflict of interest

The authors declare no conflict of interest.