Waste image classification into organic or recyclable ones with CNN algorithm.

A NodeMCU-ML based project which performs extensive waste classification by leveraging ResNet50's precision and ESP8266's extensibility.

AI-powered waste classification system using deep learning, Combines a custom CNN and EfficientNet (transfer learning). Achieves 99% training and 95% validation accuracy. Classifies images into cardboard, glass, metal, paper, plastic, and trash. Includes prediction, evaluation, and visualization tools.

This project automates trash sorting using a Raspberry Pi-controlled robotic arm, leveraging TensorFlow Lite and OpenCV for real-time classification of paper, plastic, and metal waste.

This repo contains all the source code and obtained data for the waste classification

an object detection model to find waste on the fly

Waste classification system using MobileNetV2 transfer learning. Flask web app with upload, camera capture, and batch processing for 7 waste categories

Waste Classification into biodegradable, non-recyclable, recyclable and reusable.

Synthetic Municipal Solid Waste Generator for AI-powered Waste Recognition System

A deep learning project that classifies garbage into six categories (cardboard, glass, metal, paper, plastic, and trash) using a Convolutional Neural Network (CNN). Includes a complete frontend-backend system built with Flask for real-time image classification.

Waste image classification using CNN (MobileNetV2 & DenseNet121) on the TrashNet dataset with augmentation and class weighting.

This project implements a deep learning–based garbage classification system using a custom Convolutional Neural Network (CNN). It automatically classifies waste images into recyclable categories, supporting efficient and smart waste segregation through AI.

Exploring the use of Vision Transformers (ViT) for waste classification

EcoGuardian is a mobile app that uses AI-driven image recognition to classify waste into recyclables, compost, and landfill categories.

End-to-end waste product classification system using transfer learning (VGG16) and Flask for real-time image inference.

Web app basata su intelligenza artificiale per identificare, catalogare e classificare correttamente i rifiuti.

Edge AI Prototype enhances sustainability via AI. Task 1 classifies recyclables (Organic and Inorganic) using MobileNetV2. Task 2 predicts crop yields with Random Forest. Task 3 analyzes AI ethics in medicine. Uses TensorFlow, Scikit-learn, Pandas, and TFLite.

HR-ViT: A hybrid ResNet50–Vision Transformer model for six-class municipal waste classification (plastic, paper, metal, glass, organic, batteries), achieving 98.27% accuracy and designed for real-world recycling systems.

EcoWaste AI uses MobileNetV2 to classify waste as organic or recyclable and a RandomForest model to estimate CO₂ savings based on item weight. It helps users make better disposal choices by providing predictions, confidence scores, carbon-impact estimates, and simple eco-tips through an easy interactive interface.