March 2021
·
515 Reads
·
6 Citations
Journal of Physics Conference Series
The proposed work presents efficient objection detection and tracking algorithm for autonomous vehicles, which is developed in MATLAB with Image Processing Computer Version Toolbox and Automated Driving Toolbox. The developed algorithm track and detects moving and stationary objects such as other vehicles, pedestrians, and traffic lanes. Accurate and efficient tracking are important to analyze object behavior. For this work various build in detectors from MATLAB tool box were tested and compared. The evaluation of algorithm was carried on for 19 short videos from 8 seconds to 23 seconds, and then it applied to K-dataset and full road experiments by the Automated Driving Lab. The full road tests are between one to five minutes, including CAR to CAR WEST, CAR WEST to CAR, and campus marked roads.