Color Based Object Sorting System using Deep Learning 
Pratik Roy1, Satakshi Roy2, Rahul Agrawal3, Sonal Sharma4
1Pratik Roy*, Junior, Department of Electronics and Communication Engineering, VIT University, Vellore, India.
2Satakshi Roy, Student, Department of Electronics and Communication, Vellore Institute of Technology, Vellore, India.
3Rahul Agrawal, Student, Department of Electronics and Communication, Vellore Institute of Technology, Vellore, India.
4Sonal Sharma, Student, Department of Electronics and Communication, Vellore Institute of Technology, Vellore, India.
Manuscript received on June 01, 2020. | Revised Manuscript received on June 08, 2020. | Manuscript published on June 30, 2020. | PP: 964-968 | Volume-9 Issue-5, June 2020. | Retrieval Number: E9896069520/2020©BEIESP | DOI: 10.35940/ijeat.E9896.069520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Object sorting is a very common industrial application but at the same time it is a tiresome process as handling so many objects is a menial task which is not so promising in maintaining consistency and thereby arising quality issues. Object sorting, if done manually, is not only time consuming but also it seems to be an uphill task pragmatically. Nowadays amid various technological advancements, industries have become fully automated so an automated sorting system is essentially required to replace this conventional system of manual sorting knowing that this process can be made completely autonomous by properly channeling the use of technology. The main objective of this paper is to propose a smarter, intelligent and cost-effective object sorting system which categorizes the objects based on their respective color and will place them at their designated locations to minimize the cost and optimize the productivity. We have implemented the sorting system using Raspberry pi (an open-sourced Linux based board) interfaced with a camera module along with some side electronic circuitry such as servo motors and sensors. The color recognition is done using the IBM Watson visual recognition model where we have uploaded the dataset of captured images. For picking and sorting the objects, we have made use of a robotic arm that will rotate with the help of servo motor up to certain angles.
Keywords: IBM Watson Studio, Raspberry Pi, Robotic Arm, Transfer Learning.