MobileNet-Based Garbage Classification: Enhancing Recycling with Machine Learning

Effective waste management depends critically on garbage classification, which tries to improve recycling procedures and lessen environmental effect. Automation garbage sorting systems now operate with far higher precision and efficiency thanks to recent developments in machine learning. The performance of two widely used convolutional neural network (CNN) architectures, MobileNet and ResNet50, in categorizing different kinds of waste is assessed in this paper. We contrast the more intricate and deeper ResNet50 model with MobileNet, which is renowned for its light weight structure and high computing efficiency. Results from experiments show that MobileNet outperforms ResNet50 in terms of accuracy and computing efficiency by classifying garbage into predetermined categories with an accuracy of 80%. The effective model architecture of MobileNet is responsible for its better performance since it enables it to keep high accuracy while using less processing power. This qualifies MobileNet especially for use in practical applications where processing speed and resource limitations are important factors. The results open the door to more successful waste management solutions by demonstrating the feasibility of leveraging MobileNet for scalable and effective Garbage classification.