Applying Unmanned Aerial Vehicle Photo Image to Traffic Analysis System

• Main Authors: Li,Chun-Te, 李俊德
• Other Authors: Chung,Yi-Nung
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/51338352114202755364

碩士 === 國立彰化師範大學 === 電機工程學系 === 104 === This thesis proposes a traffic analysis system using aerial image and image processing techniques. With high degree of freedom of unmanned aerial vehicle (UAV), we can capture any road at any time without setting traditional cameras. It will play an important role in the intelligent transport system (ITS). We use the road lane as a scale to calculate the speed of vehicles. Temporal differencing method is used to separate moving objects and backgrounds, and moving objects are usually vehicles. We can find the centroid of vehicles and calculate the moving distance of it expressed in pixel. To get the position of lane and calculate the speed of vehicles, thresholding method is used in the backgrounds. The experiment was done in the height from 20 meter to 40 meter in real roads. The result shows that this method performs well in 20m and 30m, and the mean error of calculating vehicle speed is 5.45%.