Citation

Ramasamy, Manickam and Deivasigamani, Subbramania Pattar and Liyas, Mohammad Arif and Thankaraj, A. and Senthilpari, Chinnaiyan and Sakthivel, Sudha and Gowrishankar, K. (2025) DESIGN OF SMART WASTE SORTING SYSTEM USING DEEP LEARNING TECHNIQUES. Journal of Engineering Science and Technology., 20 (6). pp. 1877-1894. ISSN 1823-4690

Text Scopus - Document Details8.pdf - Published Version Restricted to Repository staff only Download (214kB)

Abstract

Due to the exponential growth in garbage production worldwide, garbage management has emerged as one of the planet’s most critical challenges. The effectiveness of waste classification and sorting systems, which are an integral part of waste management, is vital to recycling programs. However, designing an automated system for waste classification is a significant difficulty in waste management. The experiments were performed in five phases: dataset preparation, training the DCNN models for waste classification, evaluation of the trained framework, model performance comparison, and real-time performance analysis. This study assesses the efficacy of multiple Deep Convolutional Neural Network (DCNN) models in waste material classification. The goal will be identifying the optimal Deep Learning (DL) technique for solid waste classification. Following the training of three pre-trained DCNN models on a trash image dataset, the most accurate model was chosen for evaluation with our prototype garbage sorting device. The DCNN models were trained using a publicly available collection of waste image datasets, enabling automated waste classification into distinct categories such as plastic, glass, paper, and metal. Standard performance criteria, such as accuracy, precision, recall and F1-score, were employed to assess the trained DCNN models. Further, the trained model was tested in real-time using an in-house-developed prototype waste sorting system model. The experiential results show that the trained DCNN-based waste classification model scored an average of 98% waste classification accuracy for real-time evaluation.

Item Type:	Article
Uncontrolled Keywords:	Convolutional neural network
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	22 Dec 2025 08:18
Last Modified:	26 Dec 2025 03:45
URII:	http://shdl.mmu.edu.my/id/eprint/15129

Downloads

Edit (login required)