利用無人飛機系統航拍輔助土地複丈

Abstract

無人飛機系統(Unmanned Aircraft System，UAS) ，已經可以整合全球衛星導航系統(Global Navigation Satellite System，GNSS)、慣性導航系統(Inertial Navigation System，INS)，以得到外方位參數之空間坐標(X,Y,Z)外，INS又可以得到外方位之姿態參數(ω,φ,κ)，目前數位相機與攝影機在解析度以及取樣之精細度部分上都已經大幅提升， 再加上無人飛機系統具有即時傳輸訊號、影像及自動定位、飛行與起降其遇到障礙物亦具有避讓之功能，讓無人飛機載具有更安全之控制機制。 就地籍測量而言其宗地坐落位置經界物之取樣，如果能利用UAS並配合航線規劃利用其自動駕駛 (automatic pilot)功能進行航拍作業，除了可提高資料蒐集的自動化程度外，對於經界物之取樣亦具有著高效率、省人力及經費之益處，對於土地複丈作業其助力更甚。 目前的UAS系統大部分搭載非測量等級相機，其飛行高度約在起飛地點相對之50至200公尺左右高度，就傳統航測而言UAS是以近景攝影測量的方式獲取目標物資訊，要率定非量測型像機的內方位參數則依據近景攝影測量圖資以及商業軟體解算，故本研究將以四旋翼 UAS 系統之飛行載具DJI Phantom 4 PRO PLUS，搭載非測量等級相機，於研究區進行地籍經界物航拍路徑規劃、正射影像之產製及影像檢測點之間與真實世界物點之間相對距離之精度分析、經界物與地籍圖之套繪、執行土地複丈以及獲取現地圖資使用傳統地測與無人機空拍兩種測量模式之成本工時分析，並探討加入控制點(GCP)與連結點時機，進而使得經正射影像處理過後之經界物與地籍圖套繪成果之精度得以提升並可達到地籍測量實施規則之規範要求。Unmanned Aircraft System (UAS) can be integrated with Global Navigation Satellite System (GNSS) and Inertial Navigation System (INS) to obtain the spatial coordinates (X,Y,Z) of exterior orientation parameters. INS can also provide pose parameters (ω,φ,κ). In nowadays, digital cameras and video recorders have become more advanced in terms of image resolution and sampling rate. With the functions of real-time signal and video transmission, auto positioning, and auto collision avoidance during landing and takeoff, unmanned aerial vehicles are equipped with safer control mechanisms. Cadastral surveying involves sampling of passing calls of the surveyed land. If UAS can be used in conjunction of route planning and automatic pilot to perform aerial photography, the data collection procedure can be more automated, and the sampling of passing calls can be more efficient, labor saving, and cost saving. The benefit for land resurveying will be even greater. Most UAS is equipped with a non-metric digital camera and limited to a flight attitude of 50~200 meters from the takeoff point. In terms of traditional aerial survey, UAS captures data of the target object based on close-range photogrammetry. The interior orientation parameters of the non-metric digital camera are determined according to the close-range photogrammetry data and analysis with commercial software. This study uses a four-rotor UAS, DJI Phantom 4 PRO PLUS built with a non-metric camera, as the instrument and plans the route for aerial photography of passing calls in the area to survey. This study produces orthophotos of the area, analyzes the precision of relative distances between the survey points on the image and between the object points in the real world, overlays the passing calls and the cadastral map, and compares the cost and time of land resurveying and acquiring map data using the traditional land survey method versus the UAS-assisted aerial photography. This study also discusses the timing for incorporating ground control points (GCP) and tie points to further improve the precision of the overlay of passing calls obtained from orthophotos and the cadastral map and meet the requirements of the Cadastral Survey Implementation Rules.