Automated Conveyor Belt Sorting System Using Machine Learning

Abstract

This project presents the design and implementation of an Automated Conveyor Belt Sorting System using Machine Learning for industrial automation applications. The system integrates a conveyor belt mechanism with a Convolutional Neural Network (CNN)-based object classification model to perform real-time sorting of objects. A camera module captures images of objects moving on the conveyor belt, and the trained machine learning model processes the images to identify the object category. The classification result is transmitted to an Arduino Uno microcontroller through GPIO communication. Based on the received signal, servo motors are activated to divert objects into their respective bins. A 16×2 LCD display provides real-time status updates of the sorting process. The system reduces manual effort, improves sorting accuracy, and enables efficient automated material handling. Experimental results show that the trained model achieves high classification accuracy and reliable real-time performance, making the system suitable for industrial automation applications.

Introduction

The Automated Conveyor Belt Sorting System is designed to improve industrial sorting processes by replacing manual and sensor-based methods with a machine learning–driven solution. Traditional systems are often inefficient and error-prone, while this proposed system uses a Convolutional Neural Network (CNN) to classify objects based on images captured by a camera on a conveyor belt.

The system architecture integrates a camera for image acquisition, a CNN model for real-time classification, and an Arduino Uno microcontroller for controlling servo motors that physically sort objects into different bins. A DC motor drives the conveyor belt, while an LCD displays object categories and system status. The entire process is powered by a regulated supply for stable operation.

The working principle involves capturing images of moving objects, classifying them using the trained CNN model, sending the result to the Arduino, and activating servo motors to divert objects accordingly. This cycle repeats continuously for automated sorting.

Experimental results show high performance, with around 95.8% classification accuracy and about 97% overall sorting accuracy. The system demonstrates reliable real-time operation, strong precision and recall, and effective coordination between image processing and hardware control.

Conclusion

The Automated Conveyor Belt Sorting System using Machine Learning successfully demonstrates the integration of artificial intelligence and embedded systems for industrial automation. The proposed system provides efficient real-time object classification and automated sorting with minimal human intervention. The CNN-based machine learning model achieved high classification accuracy, while the Arduino-controlled servo mechanism ensured reliable sorting operation. The system improves efficiency, reduces human error, and offers a scalable solution for modern industrial material handling applications. Overall, the project demonstrates the practical implementation of machine learning in industrial automation and serves as a foundation for future intelligent sorting systems.

References

[1] Krizhevsky, A., Sutskever, I., & Hinton, G. E., “ImageNet Classification with Deep Convolutional Neural Networks,” Advances in Neural Information Processing Systems, 2012. [2] Sharma, D., & Patel, K., “Vision-Based Automated Sorting Using OpenCV,” International Journal of Advanced Research in Computer Engineering and Technology, 2018. [3] Prabhu, S., Rajan, V., & Mohanraj, T., “Arduino-Based Automated Sorting System with Servo Control,” Journal of Embedded Systems and Applications, 2020. [4] Ghosh, A., Mondal, S., & Das, P., “ML-Integrated Sorting System Using Raspberry Pi and Arduino,” International Conference on Intelligent Systems and Computing, 2021.

Copyright

Copyright © 2026 Kabilan S, Joshua L, Lokeshwaran K, Mrs. R. Gracelin Sheeba. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.