Development of a Color-Based Image Recognition System for Robotic Sorting and Picking

Abstract

This research presents the development of an automated object sorting and picking system using a robotic arm controlled by colour-based image recognition. The system is designed to enhance efficiency and accuracy in manufacturing processes by eliminating the need for manual sorting. A Dobot Magician robotic arm, an Arduino microcontroller, a conveyor belt, a photoelectric sensor, and a camera are integrated to achieve this goal. Colour segmentation is implemented using the HSV colour space, enabling the system to accurately classify objects based on colour. Experimental results demonstrate the system's ability to successfully sort objects of three colours in a random sequence with 100% accuracy over ten trials.