Today, the county police authorities deal with road traffic accidents and more to
measuring wheel as the main measurement tool. The police officer in the field to measure
the relevant data and draw the traffic accident scene sketch is completed, and then return
to the unit to operate Microsoft Visio software or Easy Street Draw and other computer
graphics software to computerize the traffic accident scene sketch and upload, which
inevitably exists human error, and can only be presented in 2D way restrictions. The scene
sketch of the traffic accident is an extremely important part of the accident handling. The
factors such as the drawing ability, the search technique and the subjective judgment of the
police officer are directly related to the good and bad of the accident scene sketch, which
in turn affects the accident analyst to make a correct report. Affect the interests of the
parties to the traffic accident is huge.
In recent years, unmanned aerial vehicles (UAV) for aerial imaging applications
becomes more and more widely. The aerial photography of UAV which is equipped with
high-resolution cameras can provide Point Cloud, Digital Surface Model (DSM) and
Orthophoto image for the planning of traffic engineering improvement and analysis of
driving behavior simulation. This study is based on using multi-axis aircraft in unmanned
aerial vehicles for traffic accident dealing , which is discussing the performance of
multi-axis aircraft and simulation reconstruction function of Pix4Dmapper and Comparing
the benefits and the difference between the traditional scene of the accident. It is expected
to 3D simulation of traffic accident scene with shoot of the multi-axis aircraft will be more
realistic than 2D sketch. On the other hand, it is hoped that the multi-axis aircraft can be
used as a measurement tool to reduce the man-made error caused by the use of the
measuring wheel. Finally, it is hoped that the study will be used as a reference for the
current traffic accident investigation.
第一章 緒論1
1.1 研究背景 1
1.2 研究動機與目的 2
1.3 研究範圍與內容 3
1.4 研究方法 4
1.5 研究流程 4
第二章 文獻回顧與探討 7
2.1 無人飛行機之起源及沿革 7
2.2 航空測量相關理論11
2.3 交通事故處理之現況與缺失 14
2.4 無人飛行機法律規範16
第三章 多軸飛行器實驗設計與功能介紹25
3.1 多軸飛行器之功能與限制 25
3.2 相關使用的程式與APP29
3.3 實驗設計內容34
第四章 多軸飛行器實驗及建模分析 41
4.1. 實驗模擬場景與實驗工具41
4.2. Pix4DMapper 軟體影像後製 45
4.3. 其他測量工具之比較60
4.4. 各項工具時間效益與誤差比較76
第五章 結論與建議 83
5.1. 結論 83
5.2. 建議 84
參考文獻 87
附錄一 多軸飛行器Phantom 3 Standard 使用說明 89
附錄二 Pix4DMapper 軟體影像後製程序 93
Distinctive Image Features，2004，David G. Lowe.
Pix4D, UAV Mapping sofeware ，2015.
Vlastimil Kriz , Petr Gabrlik . 2015.UranusLink - Communication Protocol for UAV with Small Overhead and Encryption Ability.
申承翰(2013)，無人飛行載具影像數值地形模型建置及精度評估，國立台北科技大學，碩士論文。
李宏振、程玉傑、陳嘉隆(2001)，改良網格測距法以還原事故現場之研究，中央警察大學，碩士論文
周文生、陳雯龍等人(2002)，整合道路交通事故調查作業管理規範之研究，九十二年道路交通安全與執法研討會
林秋芬(2015)，無人飛行載具針實三維模型建模精度評估，國立台北科技大學，碩士論文。
施旻汶(2015)，自動化無人飛行載具輔助擷取橋梁影像之研究，國立中央大學，碩士論文。
簡榮興、王天佑、曹鼎志、辜炳寰、簡志凱 2010），應用無人飛行載具 UAV 技術進行土石災區之調查-以88 水災後之六龜鄉大智瀑布為例，Taiwan Rock
Engineering Symposium。
徐偉勛(2015)，運用多軸飛行器執行軍事任務之探討，陸軍學術雙月刊，第五十一卷541 期第131 頁至第151 頁。
莊大賢、蕭宇伸(2015)，利用無人飛行載具可見光與近紅外光影像分析神木村崩塌潛勢，國立中興大學，碩士論文
曾子峻(2010)，多軸飛行器於橋梁檢測之應用研究，國立高雄應用科技大學土木工程與防災科技研究所，碩士論文。
黃英婷(2001)，影像鑲嵌自動化之研究，國立中興大學，碩士論文。
詹丙源、陳英智 (2012)，道路交通事故現場管制與跡證蒐集之探討，101 年道路交通安全與執法研討會論文集，第199 至210 頁。
羅平慶、吳重緯、郭俊傑，淺談多軸飛行器於工程上之應用，大地技師期刊，第9期第40 頁到第47 頁。
蘇志強(2010)，交通事故偵查理論與實務(增訂二版)。桃園：中央警察大學
蘇柏軒、蕭宇伸(2015) ，以無人飛機系統(UAS)獲取國立中興大學校區三維地形之精度分析，水土保持學報第1349 頁至第1366 頁
蘇柏軒、蕭宇伸(2015)，利用無人飛行載具可見光與近紅外光影像分析神木村崩塌潛勢，碩士論文。
劉進金 (2002) ，應用工程地質學航照地質判釋，航照課程資料，v2.1 版