Identification and Sorting of Objects based on Shape and Colour using robotic arm
Lennon Fernandes, B. R. Shivakumar
Abstract
Robotic Automation is favored since human eyes are unable to constantly distinguish colors which will eventually reduce the work efficiency. In this work, a system that sorts objects based on their shape and color is proposed and implemented using a robotic arm. The system captures a real-time image from a web camera and is preprocessed by performing RGB to HSV conversion and noise removal by using a median filter. Colour detection of the objects is then carried out using the lower and upper HSV values. The system then performs an object’s shape detection using the contour detection technique. In this technique the contours are identified first using the modified boundary fill approach and the shapes are then detected from the contours using the Douglas-Peucker algorithm. A Robotic arm is constructed to sort objects based on their identified shape and color. The system is tested for three object shapes: square, triangle, and rectangle.