IOT-BASED WASTE MANAGEMENT SYSTEM TO SEGREGATE AND RECOGNIZE FERROUS METALS

Abstract

Ferrous material contains Iron which is one of the valuable elements. Waste management
systems are mainly detecting and sorting three different types of materials including: glass, plastic, and
metal. However, there have been some systems capable of detecting ferrous materials but they are only
differencing the ferrous materials from non-ferrous materials.
The proposed system is capable of differentiating between the mild and stainless-steel using rust detection
method based on Supervised SVM classifier. Supervised SVM classifier includes two layers as: 1) object
moving detection, 2) rust detection method.
The findings indicate that the accuracy of detection is 100% for actual object placed on the conveyor.
However, the negative detection percentage is 93.7% where a moving object was detected while there was
no actual object over the conveyor. Accuracy of rust detection and SVM classifier as one of the main
components of this system was also tested. The accuracy percentage yields that the accuracy of system in
detection of moving object, rusty, and non-rusty materials are as 97.1%, 93.75%, and 91.43%.