# mapbox/robosat

Semantic segmentation on aerial and satellite imagery. Extracts features such as: buildings, parking lots, roads, water, clouds

Repository: https://github.com/mapbox/robosat
Canonical: https://ross.abutalabs.com/products/robosat
Language: Python
License: MIT
License Family: permissive
Topics: satellite-imagery, aerial-imagery, machine-learning, segmentation, openstreetmap
Last push: 2026-06-29T15:11:59+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 90, release rhythm 8, longevity 100
- inputs: {"age_days": 3009, "days_push": 65, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2064, forks 385 (observed 2026-08-28T04:06:09.996509+00:00)

## What it is
RoboSat is an end-to-end Python pipeline for semantic segmentation and feature extraction from aerial and satellite imagery, identifying features like buildings, roads, parking lots, and water. It provides CLI tools covering the full workflow from data preparation and training to prediction and GeoJSON polygon export.

## Use cases
- extract building footprints from satellite imagery
- segment roads from aerial imagery
- train a segmentation model on drone imagery
- convert satellite imagery masks to GeoJSON polygons
- detect water or parking lots in aerial photos
- generate OpenStreetMap features from imagery

## When to choose
- you need an end-to-end pipeline for extracting features like buildings or roads from aerial or satellite imagery
- you want to train custom semantic segmentation models on geospatial raster data
- you are contributing to OpenStreetMap with imagery-derived features

## When to avoid
- you need actively maintained software - Mapbox no longer maintains RoboSat
- you need general-purpose image segmentation outside aerial/satellite imagery
- you want a modern deep-learning framework integration with recent model architectures

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, computer-vision, image-processing, cli
- domain: machine-learning, computer-vision
- platform: python, cli
- tags: semantic-segmentation, satellite-imagery, aerial-imagery, openstreetmap, feature-extraction, deep-learning, geospatial, maps, linux, macos

## Member repositories
- mapbox/robosat (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:09.996509+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:57:09.020556+00:00, confidence not recorded.
  - readme: https://github.com/mapbox/robosat (fetched 2026-08-28T04:06:09.996509+00:00, sha f5dc7207aff1)
- Data as of 2026-08-30T08:39:29.467469+00:00.