Author(s)

K Vijay, K Meenaloshini, S Madhumitha, B Kanimozhi, V Yuvatharshini

  • Manuscript ID: 120268
  • Volume 2, Issue 4, Apr 2026
  • Pages: 296–302

Subject Area: Computer Science

DOI: https://doi.org/10.5281/zenodo.19551052
Abstract

Rapid urbanization and industrial growth have significantly increased global solid waste generation, leading to severe environmental and public health challenges. Traditional waste segregation methods rely on manual sorting or centralized processing, which are labor-intensive, inefficient, and prone to human error. This paper proposes ECO VISION AI, an intelligent waste segregation and recycling system that integrates Artificial Intelligence (AI), Convolutional Neural Networks (CNN), and Internet of Things (IoT) technologies for automated waste classification at the source. The system captures images of waste items using a camera module, processes them using a lightweight CNN model deployed on a Raspberry Pi, and directs waste into appropriate bins using servo actuators. IoT integration enables real-time monitoring of bin levels, classification data, and system status through a cloud dashboard. Experimental evaluation demonstrates improved segregation accuracy, reduced human intervention, and enhanced recycling efficiency. The proposed system contributes toward sustainable waste management and supports smart city initiatives.

Keywords
Waste SegregationIoTConvolutional Neural NetworkRaspberry PiImage ClassificationSmart CitySustainable Waste Management.