Betula

2.5-metre Super-resolution for Advanced Analytics

Our Betula technology applies artificial intelligence techniques to Sentinel-2 satellite images, boosting the spatial resolution of the R, G, B, and NIR bands fourfold. From a 10-metre image, we generate a product with a spatial resolution of 2.5 metres, preserving the characteristics of the original bands.

Betula is designed to minimise the generation of artefacts or “hallucinations” through supervised and deterministic training, without using Generative Adversarial Networks (GANs). In this way, the product we generate preserves the radiometry and geometry of the original bands, ensuring its suitability for geospatial analytics applications, such as the calculation of vegetation indices, change detection, and recurrent monitoring, among others.

Betula expands the potential of Sentinel-2 images to a wider range of applications requiring higher spatial resolution, while retaining the main advantages of the European Space Agency’s (ESA) Copernicus programme, such as its global coverage and high revisit frequency.

With Betula, we can, for example, enhance the photointerpretation of forest fires, delineate agricultural plots more accurately, compile inventories of urban areas without the need for fieldwork, continuous monitoring of critical infrastructure, etc. All at a low cost and applicable anywhere in the world.

Betula use cases

Olive groves in Sfax (Tunisia)

Around the Tunisian city of Sfax, vast and historic olive groves stretch out, creating one of the Mediterranean’s most distinctive agricultural landscapes.

On the left, agricultural fields in a Sentinel-2 image captured on 4 August 2026. On the right, Betula’s super-resolution image reveals a world of detail, where each plot of land is visible and every olive tree can be counted.

Sentinel-2 (10 m) / 04-08-2026 Betula (2.5 m) / 04-08-2026
Fire in Villa del Prado (Spain)

Images of the Madrid town of Villa del Prado, to which the fire that started on 22 July 2026 in Almorox (Toledo) spread.

BEFORE. In the comparison above, one can see false-colour images from 23 July, before the fire had even touched Villa del Prado.

AFTER. In the comparison below, using false-colour images from 24 July, we can see how the fire spread towards built-up areas, forests, and farmland.

BEFORE / Sentinel-2 (10 m) / 23-07-2026 BEFORE / Betula (2.5 m) / 23-07-2026
AFTER / Sentinel-2 (10 m) / 24-07-2026 AFTER / Betula (2.5 m) / 24-07-2026
Earthquake in La Guaira (Venezuela)

The area of La Guaira (Venezuela), on the Caraballeda coast, was shaken by the earthquakes that struck on 24 June 2026.

BEFORE. The comparison above shows images of the area captured on 22 June, before the earthquake struck.

AFTER. The comparison below shows images taken on 29 June, after the quake. The collapsed buildings are visible in the image super-resolved by Betula.

BEFORE / Sentinel-2 (10 m) / 22-06-2026 BEFORE / Betula (2.5 m) / 22-06-2026
AFTER / Sentinel-2 (10 m) / 29-06-2026 AFTER / Betula (2.5 m) / 29-06-2026
Landslide in Blatten (Switzerland)

The Swiss town was buried on 28 May 2025 when an Alpine glacier collapsed.

BEFORE. The comparison above shows images from 18 May, before the event.

AFTER. In the comparison below, images from 30 May reveal the town almost entirely engulfed by the glacier’s collapse.

BEFORE / Sentinel-2 (10 m) / 18-05-2025 BEFORE / Betula (2.5 m) / 18-05-2025
AFTER / Sentinel-2 (10 m) / 30-05-2025 AFTER / Betula (2.5 m) / 30-05-2025
For more information about Betula, to request a demo of our technology, or to ask for a quote, you can contact the Tracasa Global Business Development team at the following email address: commercial.support@tracasa.es.