Building low cost advanced driver assistance systems – vehicle tracking
Date of Issue2017-05-11
School of Electrical and Electronic Engineering
With the rise of the economy growth, more people are having their own vehicles and benefited from the convenience of traveling. However, roads are getting crowed and the car accidents are increasing rapidly yearly as the result. One small accident happened on the road during the peak hour may cause traffic paralysis and thus, lead to loss of money. Through counting, most of the car accidents are caused because of the reckless of the driver. Autonomous car is then introduced to the public because machines will never feel tiredness and they will always obey the set rules. Nowadays, some major corporations have announced plan to manufacture and develop self-driving car such as Google, Apple and Tesla. Such car will begin to sell in these few years, and some countries have passed legislation to allow autonomous cars. However, auto-driving system is still not completed yet and there are some serious accidents reported due to system failed to detect the object and cars on the road. Therefore, I am going to build a low cost advanced driver assistance system. Not just helping drivers to prevent car accidents, also becoming part of the autonomous driving system as a secondary protection/assistance. In this project, researching and studying on the Kernelized Correlation Filters (KCF) Algorithm is very important. KCF is used to track an object when it is moving and this algorithm is the fastest with highest precision so far. The programing software MATLAB is used to implement a user interface integrated with two algorithms. The final algorithm should be able to analyze the car plate information and track its movement.
Final Year Project (FYP)
Nanyang Technological University