Data Augmentation Based Waste Classification Using CNN

This study presents the development of a machine learning model that uses Convolutional Neural Networks (CNNs) to automatically classify waste into two categories: Organic and Recyclable. The model is built with three convolutional layers, followed by max-pooling and fully connected layers. It uses ReLU activation functions to extract important features from images and a sigmoid function to make final classification decisions. To improve the model’s ability to generalize, we trained and validated it on image datasets with data augmentation techniques. The proposed model achieved 96.78% accuracy in classifying waste with less computational complexity. To make the solution more accessible and efficient for real-world use, especially on portable devices, the model was also converted into TensorFlow Lite (TFLite) format. Overall, this system offers a practical and scalable approach to automating waste classification.