Home > Portfolio > CARBON SAT-TRACKER: SPACE-TECH AI
AI Satellite Space-Tech Python

CARBON SAT-TRACKER: SPACE-TECH AI

Detects large plastic waste on water and trash dumps using satellite imagery.

CARBON SAT-TRACKER: SPACE-TECH AI Hero Image

A satellite imagery analysis system built to detect large-scale garbage dumps and marine plastic accumulation using computer vision.

"By processing multispectral satellite bands, the tool highlights pollution hotspots, supporting environmental clean-up operations."

The Tech Stack

We built this planetary pipeline using Python, processing OpenStreetMap and Sentinel-2 satellite imagery. Convolutional Neural Networks (CNNs) classify anomalies on land and water. The pipeline computes spectral indices to distinguish plastics from natural vegetation, rendering the coordinates onto an interactive global map dashboard.

Impact

Enables environmental agencies to monitor remote marine channels and track illegal dump growths without requiring expensive manual surveys.

System Architecture

[GEOSPATIAL CANVAS INTERFACE]
Leaflet.js Map Dashboard / Spectral Layer Toggle Overlays
▼
[ANALYSIS WORKER & CNN PIPELINE]
Sentinel API Data Scraper / PyTorch ResNet U-Net Segmenter / GDAL Library
▼
[SPATIAL COORDINATES DATABASE]
PostGIS-Enabled PostgreSQL (Anomaly Coordinates & Area Metrics)