NDBI
Normalized Difference Built-up Index. Highlights urban and built-up areas by exploiting the spectral response of built surfaces in the shortwave infrared band. Used for urban mapping, sprawl monitoring, and land use classification.
Formula
NDBI = (SWIR - NIR) / (SWIR + NIR)Overview
The Normalized Difference Built-up Index (NDBI) highlights urban and built-up areas by exploiting the higher SWIR reflectance of impervious surfaces compared to their NIR reflectance. Developed by Zha, Gao, and Ni in 2003 for automated urban area mapping.
How It Works
NDBI = (SWIR − NIR) / (SWIR + NIR). On Sentinel-2: (B11 − B8) / (B11 + B8). Built-up materials yield positive values. Dense vegetation produces negative values. A common workflow subtracts NDVI from NDBI to isolate built-up pixels from bare soil.
Key Facts
- Formula: (SWIR − NIR) / (SWIR + NIR) — the algebraic inverse of NDMI.
- Introduced by Zha, Gao, and Ni (2003).
- Commonly combined with NDVI to separate built-up areas from bare soil.
- Sentinel-2 bands: B11 (SWIR-1) and B8 (NIR).
Applications
Urban Sprawl Monitoring
Tracking expansion of built-up areas over time.
Land Use Classification
Input feature for distinguishing urban from non-urban classes.
Urban Heat Island Studies
Correlating NDBI with land surface temperature.
Limitations & Considerations
Primary limitation is confusion between built-up surfaces and bare soil/rock. Shadows in dense urban cores suppress SWIR and cause underestimation. Spectral variability of building materials affects results.
History & Background
Published in 2003, achieving 92% accuracy in the original study. Has become a foundational index in urban remote sensing, supported in ArcGIS, QGIS, and Google Earth Engine.
Analyze NDBI data with LYRASENSE
Use our agentic notebook environment to work with satellite data and compute indices like NDBI — no setup required.