Dataset of trash objects for waste classification and detection

Attentional Learning of Trash Classification

Underwater Trash Detection

automatic trash separator using yolo, mtcnn

Classification of trash images using Convolutional Neural nets

CNN Image classification of trash of 6 categories in tensorflow and pytorch

Sorts recycling and trash using computer vision

On prototype, Modeling trash with CNN method

Self-collected trash dataset used in AlphaTrash project. Contains 5600+ images of trash gathered in Thailand, sorted into general, metal, organic, paper, and plastic wastes.

A computer vision application that helps classify waste items into appropriate disposal bins using a deep learning model.

Part of my Final Year Project (FYP), this web application allows platform administrators to manage users, shops, the point system, and feedback within an AI-powered waste management & recycling rewards system. It ensures smooth operation, security, and compliance for the Trash to Treasure platform.

Fine-tuning EfficientNet-B0 for trash classification on Google Cloud TPUs using PyTorch/XLA.

Code for Automatic Trash Classification using Convolutional Neural Network Machine Learning paper, published in 2019 IEEE Conference on Cybernetics and Intelligent Systems.

A group research project about AI Image Classification for Trash Sorting for the Research Methodology in Computer Science subject conducted by students at Bina Nusantara University

A CNN model that automatically classifies waste (plastic, paper, metal, glass, cardboard, trash) to boost recycling efficiency. Trained on the TrashNet dataset, it accurately identifies different materials, reducing sorting errors and simplifying waste management.

Klasifikasi gambar sampah dengan menggunakan Convolutional Neural Network

Web backend of machine learning trash classification system

Part of my Final Year Project (FYP), this web application enables partnered shops and cafes to verify and redeem user rewards earned through recycling. Integrated with an AI-powered trash classification system, it promotes sustainable waste management by making recycling a rewarding experience.

AutoTrashSort is a smart, 3D-printed trash can that uses AI and sensors to automatically sort waste into landfill, recycling, or compost. Powered by a Raspberry Pi, camera, and servo-driven tilt plate, it classifies trash in real time for efficient waste management at home or in public spaces.

This project compares MobileNetV2 and a custom CNN model for trash classification, using feature extraction techniques like edge detection and Local Binary Patterns (LBP). Results show that MobileNetV2 without explicit feature extraction achieves the highest accuracy, providing insights for improving waste management applications.