利用光達及無人機技術輔助3D地籍測量之研究

Abstract

近年來，光達(Light Detection And Ranging ,LiDAR)技術應用蓬勃發展，可以迅速獲得高精度的點雲資料及提供豐富且細緻3D資訊，相較一般測量方式逐點觀測物體角點，藉由點、線、面呈現土地權屬位置與形狀之2D空間資訊地籍圖，較無法直觀轉換平面地籍與垂直地上物相對關係；原2D測量地籍方式亦間接影響後續應用與拘限用途，例如:地籍圖與現地地上物相對關係立體感、地籍圖重測時地籍調查表所記載指界經界物(牆壁、巷道)因有疑義亦或土地開發銷毀原有經界物，而如何直覺、佐證與復原地籍圖重測指界位置與經界物等。本論文旨在探討如何應用光達及虛擬實境(Virtual Reality , VR)及擴增實境（Augmented Reality，AR）等技術輔助台灣地籍現況的改進。傳統的地籍調查常常面臨困難，包括地形複雜、環境危險及資料精度低等問題。本研究提出利用光達技術進行精準地形測量，並建立高度真實感的三維地籍資料庫。透過VR、AR等技術，使用者可以在360度環景中自由探索地籍資訊，提高地籍資料的可視化程度。透過VR、AR與近距離無線通訊技術(Near Field Communication ,NFC)等新興科技技術，分別應用於地籍圖重測、土地複丈、智慧圖根點等業務，經實際證實在輔助地籍測量成效及實際效益是可行的。本次研究成果如下: (一)LiDAR技術在地籍測量中的應用已經成熟且有效。透過對三維成果的平均值和標準差進行分析，縱坐標差異為0.004m±0.007m、橫坐標差異為0.005m±0.007m、高程差異為0.007m±0.008m。這顯示LiDAR技術的精度達到公分等級，符合地籍測量實施規則的要求，能夠提高地籍測量的可信度。(二)由於地面光達(Terrestrial LiDAR)受到FOV(Field of View)限制，導致點雲數據的不完整性，進而影響測量的精確度及後續應用。因此，利用無人機（Unmanned Aviation Vehicle ,UAV）進行高空掃描，可以補充地面光達點雲的缺漏，使得整體點雲資料更加完整。經過驗證分析，UAV點雲與地面光達點雲的三維坐標差異為平面距離差值為0.023m±0.025m，高程差值為0.037m±0.059m。這顯示UAV可以有效輔助地面光達及全站儀無法到達的測量區域，提供更完整的資料。(三)在牆上設置NFC圖根點，可以避免因道路翻修等原因導致圖根點損毀，進而影響土地複丈的辦理。利用智慧NFC圖根點結合VR、AR技術，可以提高精度、作業效率和土地複丈成果的品質。NFC、VR和AR技術的加值應用有助於新進人員快速尋覓圖根點點位資訊，並有效輔助地籍測量。(四)目前使用的地籍調查表僅記載重測時的經界物，並無相關影像紀錄和三維資訊，對後續的法院鑑測造成困擾。引入3D地籍與光達點雲平台可以免現地指界，加速地籍重測作業時間。然而，現行地籍調查表格式仍為舊式，無法直觀呈現3D地籍資訊。因此，建議引入4D地籍調查表，以彌補現有調查表的不足，並提供更完善的地籍資訊。In recent years, the application of LiDAR (Light Detection And Ranging) reverse engineering technology has flourished, enabling the rapid acquisition of high-precision point cloud data and providing rich and detailed 3D information. This is in contrast to traditional measurement methods, which observe objects point by point and use points, lines, and planes to represent the 2D spatial information of land ownership positions and shapes in cadastral maps. These traditional methods make it difficult to intuitively transform the relationships between planar cadastral maps and vertical objects above ground. The original 2D cadastral mapping method also indirectly affects subsequent applications and restricts their usage. For example, the 3D sense of the relationship between the cadastral map and the actual objects above ground, as well as during cadastral map re-measurement, boundary survey forms record boundary objects (walls, alleys) that may be ambiguous or destroyed due to land development. Therefore, it becomes necessary to find intuitive ways to verify and restore the boundary positions and objects during cadastral map re-measurement.This paper aims to explore the application of LiDAR and Virtual Reality (VR) technology to improve the current cadastral situation in Taiwan. Traditional cadastral surveys often face difficulties such as complex terrain, hazardous environments, and low data accuracy. This study proposes using LiDAR technology for precise terrain measurements and creating a highly realistic 3D cadastral database. With VR technology, users can freely explore cadastral information in a virtual environment, enhancing the visualization of cadastral data.Emerging technological tools, such as Virtual Reality (VR), Augmented Reality (AR), and Near Field Communication (NFC), have been utilized in the remeasurement of cadastral maps, land digitized revision, and in the business of intelligent mapping control points. Empirical evidence has demonstrated that these technologies effectively enhance the results and practical benefits of cadastral surveying. The following are the outcomes of this research:1. The application of LiDAR technology in cadastral surveying is mature and effective. By analyzing the average and standard deviation of the three-dimensional outcomes, the differences in longitudinal coordinates, transverse coordinates, and elevation are 0.004m±0.007m, 0.005m±0.007m, and 0.007m±0.008m, respectively. This demonstrates that the precision of LiDAR technology is at the centimeter level, fulfilling the requirements of cadastral surveying regulations and enhancing the credibility of cadastral surveys.2. Due to the limitations of FOV (Field of View) in Terrestrial LiDAR, the incompleteness of point cloud data ensues, thereby affecting the accuracy of measurement and subsequent applications. Therefore, by using Unmanned Aviation Vehicles (UAV) for high-altitude scanning, gaps in the terrestrial LiDAR point cloud can be supplemented, making the overall point cloud data more complete. After validation analysis, the three-dimensional coordinate difference between UAV point cloud and terrestrial LiDAR point cloud is 0.023m ± 0.025m in planar distance, and 0.037m ± 0.059m in elevation difference. This demonstrates that UAV can effectively assist areas unreachable by terrestrial LiDAR and total station instruments, providing more comprehensive data.3. Installing NFC mapping control point on walls can prevent damage due to road repairs, thus preventing disturbances to the process of land re-surveying. Combining intelligent wall mapping control point with VR and AR technology can improve precision, work efficiency, and the quality of land re-surveying results. The value-added application of NFC, VR, and AR technologies aids new personnel in quickly locating and understanding the intelligent mapping control point NFC and point location information and effectively assists in cadastral surveying.4. The current cadastral survey form only records boundary objects during remeasurement, and lacks relevant image records and three-dimensional information, causing difficulties for subsequent court evaluations. The introduction of a 3D cadastral and LiDAR point cloud platform can eliminate the need for onsite boundary pointing and expedite the time required for cadastral remeasurement. However, the current cadastral survey form format remains outdated, unable to intuitively present 3D cadastral information. Therefore, it is suggested that a 4D cadastral survey form be introduced to compensate for the deficiencies of the existing form and provide more comprehensive cadastral information.