A visual based target detection and tracking system (II)

• Main Author: Tan, Augustine Peng Soon.
• Other Authors: Xie Lihua
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering
• Online Access: http://hdl.handle.net/10356/40906
• Institution: Nanyang Technological University

The main objective of this project is to design and create a visual sensor system and it’s Graphics User Interface (GUI). 4 cameras were used for this visual sensor system to detect and identify the different robots from the images captured. There are 2 major portions in this project, the calibration process and the extraction process. The calibration process includes the color calibration, position calibration and the robot configuration. Previously the positions and orientations of the robot were calculated using Stargazer. In this project the Stargazer has been removed and the position and orientations of the robots will now be calculated solely from the image captured by the cameras. Color calibration is done by setting the values of the Hue, Saturation and Value (HSV) of the color to get the required color. Position calibration uses the known field position’s image point to calculate the homography matrix. This matrix is then used for the calculation of all the field coordinates. Robot configuration is to set the color combination of the robot to enable the identification of the robot from the images captured. The extraction process has 2 parts to it. The first part is the processing of the images captured to determine the robot’s ID and the calculation of its position and orientation. The second part is to send this calculated data to another computer via wireless communications. TCP/IP is the form of wireless communication chosen for this project.