Building a low cost advanced driver assistance system : vehicle detection

• Main Author: Wong, Cheng Hao
• Other Authors: Wang Gang
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/68008
• Institution: Nanyang Technological University

While software computing has gained its popularities among numerous industrials and real-life applications, safe driving has always been a top priority in the automotive industry. Without a doubt advanced driving assistance systems are part of the enhancement that automobile manufacturers can improve on to keep their competitive edge in the market. While a typical driving assistance system includes detection of humans, road lanes to provide early warnings to driver, avoiding drifting out of road or even fatal collision. This project comprises the aspect of Piotr Dollár’s Matlab Toolbox and techniques of integral channel features applied in object detection to develop into vehicle detection. And base on two key factors, the feature representation and the learning algorithm, we would determine the performance of vehicle detection system. Additionally incorporating symmetric feature design along the vertical axis to further improve vehicle detection from the existing codes; measurement of efficiency is also illustrated in the closure of the report.