Road edge detection algorithm for autonomous UAV based on canny edge detection

Name: Chaleampon Photeam

Organization : Kasetsart University. Faculty of Engineering

Email : chaleampon.p@ku.th

Name: Chaiwat Klampol

Organization : Kasetsart University. Faculty of Engineering

keyword: Road edge detection

ThaSH: Drone aircraft

; Canny edge detection

ThaSH: Detectors

; Unmanned aerial vehicles

ThaSH: Roads -- Surveying

; Hue histogram

Abstract: The road surface condition is important information both in economics and social. This information can be used to help the road user for the good travel planning which affects to reduce both the travel time and expenses. Moreover, the responsible agency for road maintenance also employs the information to increase the performance of maintenance management for the maximum benefit. There have been several technologies that help to collect the road surface condition data, but they must be done on the road area which is the causes of the traffic jam and accident. This research is a part of the system development for the road surface surveying by using the autonomous unmanned aerial vehicle (UAV). The research focuses on the capability enhancement of autonomous UAV in the effectiveness of the road surveying and precision of the road surface condition data collection. The image processing on Raspberry Pi computer broad is worked with the flight controller to estimate the greater precise location than the only one GPS system. In this paper, we propose the real-time road edge detection algorithm for autonomous UAV. The algorithm is processed on the Raspberry Pi by using the real-time video. The image resizing and blur are done in the preprocessing. The hue histogram is applied in the image segmentation process. After that, the Canny edge detection is employed to determine the road edge location. Finally, linear regression is used to increase the accuracy of the road edge position. The experimental results show that our algorithm provides more than 90% accurate detection of road edge line in several environmental conditions. In future work, the road edge information will be used to design the control command of the UAV flight controller.

King Mongkut's University of Technology North Bangkok. Central Library

Address: BANGKOK

Email: library@kmutnb.ac.th

Created: 2019

Modified: 2020-07-29

Issued: 2020-07-29

บทความ/Article

application/pdf

BibliograpyCitation : ใน มหาวิทยาลัยมหาสารคาม คณะวิศวกรรมศาสตร์ ร่วมกับ สมาคมวิศวกรเครื่องกลไทย. การประชุมวิชาการเครือข่ายวิศวกรรมเครื่องกลแห่งประเทศไทย ครั้งที่ 33 (ME-NETT 2019) (p.171-178). มหาสารคาม : มหาวิทยาลัยมหาสารคาม

eng

©copyrights King Mongkut's University of Technology North Bangkok

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	ME-NETT 2019pp.171-178.pdf	459.29 KB	4	2023-12-31 13:38:54

ใช้เวลา

0.035826 วินาที