Object Detection Accuracy Enhancement in Color based Dynamic Sorting using Robotic Arm

Abstract

In this study, our aim is to develop a robust sorting algorithm to classify and organize objects in terms of their colors using a robotic arm. Developed code allows the robotic arm to retrieve items from a moving conveyor belt, and sort them into corresponding bins ac- cording to their color. The three main parts of the system are the conveyor belt, the robotic arm, and the camera that is used for ob- ject detection based on color. Using the developed system, red, green, and blue objects are detected and localized by the algorithm and the robotic arm successfully picks up objects with desired color from the moving conveyor belt, then place them in their respective bins based on their color.