environmental-ai

Leveraging Bayesian Neural Networks for multimodal AUV data fusion, enabling precise and uncertainty-aware mapping of underwater environments.

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

This MVP demonstrates a multi-indicator, high-reliability wildfire detection framework that surpasses conventional approaches. By combining Earth observation with intelligent vector analytics, it opens pathways to operational-scale environmental monitoring.

Predicting alkalinity, conductance and phosphorus in unseen river stations using satellite imagery, climate data and spatial generalization — EY AI & Data Challenge 2026.

AI-powered marine debris detection and drift prediction system using deep learning, multispectral satellite imagery, geospatial analysis, and physics-based ocean modeling for environmental monitoring.

LLL-based Disaster Detector Agentic AI Application : This project enables the detection and interpretation of environmental threats (e.g., floods, infrastructure risks) by leveraging large language models (LLMs) and multimodal inputs derived from CCTV-based river surveillance feeds.

GreenGPT is an AI-powered platform designed to address India's and the world's most pressing environmental challenges through advanced document analysis, real-time chat assistance, and actionable insights generation.

AI-powered wildfire detection and risk mapping using U-Net on satellite imagery with an interactive Streamlit dashboard.

Deterministic real-time environmental regulatory escalation engine with Pathway streaming, satellite verification, and policy-grounded enforcement

Explainable CNN-BiLSTM with SHAP-based seasonal & diurnal temporal attribution for AQI forecasting across Delhi, Mumbai, Kolkata, and Chennai. Trained on CPCB hourly data (2015–2020).

This notebook implements a deep learning-based image classification system for identifying different types of garbage (e.g., plastic, paper, metal). It includes custom image preprocessing functions and prepares the dataset for model training.

Open-source environmental intelligence and AI research for extreme heat, PM2.5, air quality, wildfire smoke, urban climate, forecasting, uncertainty, and resilient homes.

Global Fire Monitoring System v3.2 is an advanced satellite-based fire analysis platform that leverages ESA CEDA Fire_cci data for large-scale global fire pattern detection and clustering analysis. The system processes 12,500+ grid cells simultaneously and provides comprehensive insights into fire behavior patterns across 6 continents.

Ecosystem health analysis through sound. Computes ACI, ADI, BIO and NDSI acoustic indices from .wav recordings or live microphone input, with optional BirdNET bird species identification and an auto-generated dashboard.

Semantic segmentation of marine debris in underwater images using DeepLabV3+ · PyTorch · TrashCan dataset

AI-powered environmental sustainability analysis using satellite imagery and deep learning

AI-powered underwater microplastic detection and classification using EfficientNet-B2. Classifies 5 polymer types (alkyd, bead, cellophane, degraded, fiber) with 97.3% accuracy and a real-time Flask dashboard for live image analysis.

Large-scale fire detection analysis using NASA FIRMS data. feat: Add dynamic region support for South America case study in v1-4_area - Enabled flexible geospatial parameterization for wildfire analysis - Updated preprocessing pipeline to accommodate South American satellite data formats.

A deep learning project implementing YOLOv8 for multi-class waste detection and classification using the TACO dataset.

Real-time oil spill detection using YOLOv8 and 3D CNN — Streamlit interface with AWS EC2 and Docker deployment