Real Time Object Tracking Using Color and Shape Features

  • Fathy Abd. Ellatif Elmisery, Ahmed Ezzat Ahmed, Ramadan Mahmoud Mustafa, Aymen Mohamed Breisha


Object tracking is one of the most advanced area of research in artificial intelligence and computer science. This Paper describes the design and implementation of object tracking system based on color and shape of the object. The proposed system consists of three stages: preprocessing stage, image processing stage, and tracking stage. in preprocessing stage, we transform the camera output from RBG to HSV. In image processing stage we subtract the object from the background and remove noise plops by erosion. For identifying the shape of the object, the number of points of its contour is counted and connected to a boundary line. After that we used the output to control robot moving to track the object. The proposed algorithm written in python 3.7.3 and run in a PC. The results show that the proposed system is simple, fast and can track the objects in Realtime.