Summary: | 碩士 === 國立東華大學 === 電子工程研究所 === 97 === This thesis proposes a fast Lane-Marking tracking algorithm based on Kalman filter. This algorithm is used to obtain lane-marking feature from the image. The lane-marking feature is popularly utilized for driving-assistance system.
This algorithm consists of initial predicting, lane-marking tracking, and lane-marking modeling. The objective of the initial predicting is to predict the good initial value on the each frame. Then the Kalman filter and Nearest Neighbor Peak algorithm (NNP) are used to track the trend of the lane-marking and found out the lane-marking location. After the lane-marking location is found out, the linear-parabolic model is used to process the model parameters of the lane-marking.
This algorithm is used to tracking the lane-marking via Kalman filter needs only partial region to detect lane-marking location in each frame. Thus, the developed system can reduce the complexity of vision data processing and meet the requirements of real-time embedded system. The experiments show that the proposed method is satisfactory to the various lane environments.
|