<![CDATA[Waste management presents a significant challenge in ensuring environmental sustainability, requiring an automated classification system to improve efficiency. This study designs a waste classification system (biological, electronic, glass, plastic) using a deep learning approach based on computer vision. The proposed method implements a custom Convolutional Neural Network (CNN) with MobileNet efficiency principles, consisting of Mobile Inverted Bottleneck Convolution (MBConv) and Squeeze-and-Excitation (SE) blocks. The model is developed from scratch using a four-class dataset and optimized with GPU processing and a batch size of 16. After fine-tuning the regularization and hyperparameters, the model achieved the highest accuracy of 75.59%.]]>
Copyrights © 2026
