Leveraging on the Transfer Learning with ResNet-50 for Efficient Classification of Waste Categories

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Automated waste classification achieves 97.8% accuracy in Nigerian urban settings, highlighting deep learning potential for sustainability.

The automated classification system achieved a validation accuracy of 97.8%, showcasing the model's strength and reliability.
Deep learning techniques were employed, utilizing a dataset of 2,527 images and applying data augmentation to enhance performance.
A two-stage training approach was used with ResNet-50, initially freezing layers and later fine-tuning them for waste-specific features.
The results underscore the potential for reducing reliance on manual sorting in waste management through efficient automation.

Suleiman et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Automated waste classification achieves 97.8% accuracy in Nigerian urban settings, highlighting deep learning potential for sustainability.