pyotb

A pythonic extension of Orfeo Toolbox

Nicolas Narçon¹, Rémi Cresson¹, Vincent Delbar²

(1) INRAE, (2) La TeleScop

Firenze, FOSS4G - 25 Aug. 2022

Satellite image processing in Python

A basic example: Pansharpening

Challenges of remote sensing

in Python

What a Python developer goes trough

Searching for pansharpening

What a Python developer goes trough

Trying the pansharpening

What a Python developer goes trough

What a Python developer goes trough

Satellite image processing does not fit in RAM

Hard to find high-level Python Open Source remote sensing solutions

Pansharpening using pyotb

pyotb: remote sensing made easy

Fits in RAM

Many already implemented remote sensing solutions in Orfeo Toolbox

Easy to write

pyotb usecase

An usecase of pyotb

Goal: labeling the bare soils

An usecase of pyotb

Goal: labeling the bare soils

Conclusion

An easier way to use Orfeo Toolbox in Python

Takes advantage of the algorithms and efficiency of Orfeo Toolbox

Easy pythonic syntax

Easy to interact with Python processing libraries
https://gitlab.orfeo-toolbox.org/nicolasnn/pyotb/


		import pyotb

		# Pansharpening
		pxs = pyotb.BundleToPerfectSensor(inp='pan.vrt', inxs='xs.tif') 

		# Computing NDVI
		ndvi = (pxs[:, :, -1] - pxs[:, :, 0]) / (pxs[:, :, -1] + pxs[:, :, 0])

		# Computing a boolean raster to identify pixels without vegetation
		bare_soils = (ndvi < 0.3)

		# Creating a boolean cloud mask from the GML vector
		cloud_mask = pyotb.Rasterization('cloud_mask.GML', im=bare_soils)

		# Masking clouds (i.e. assigning to 0) on the result
		bare_soils_masked = pyotb.where(cloud_mask == 1, 0, bare_soils)
		bare_soils_masked.write('bare_soils.tif', pixel_type='uint8')

Code sample