Image processing based lane and kerb detection

• Main Author: Tan, Kuan Hong
• Other Authors: Wang Han
• Format: Final Year Project
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
• Online Access: http://hdl.handle.net/10356/77825
• Institution: Nanyang Technological University

Autonomous Unmanned Ground Vehicles (UGVs) operate without inputs from a human operator. This is possible due to a suite of sensors that observe the surrounding environment and make decisions on its next course of action. One of such sensors is the digital video camera. The objective of this project is to study existing research and attempt to implement a real time software in C++ for single lane detection using image processing techniques. Through the use of image processing techniques such as Canny edge detection and Hough line transform, the results show that it is possible to identify key lane features with high accuracy and fast processing time within the order of tens of milliseconds. The data can then be used to provide lane departure warning and avoidance for UGVs and also to augment other sensors such as radar, sonar and LIDAR for fully autonomous driving.