An automated algorithm of GRASS GIS to retrieve the data on land cover types in Guinea, West Africa, from Landsat-8 OLI/TIRS images

The objective of this research is to evaluate Landsat OLI/TIRS multi-spectral images, and fusion of GRASS GIS, GMT and QGIS software for land cover mapping in Guinea, West Africa. The scenes of Landsat imagery were acquired on February 2014, 2018 and 2023. Land cover data were used from FAO for validation of classes. The images were classified into 18 classes and upscaled to 10 classes for generalisation towards the study area regional setting. The method included K-means clustering using scripting approach with programming codes included in Appendix. The results demonstrated that the script-based computer vision approach to image processing, classification and analysis is effective in extracting land cover classes for environmental mapping of tropical region of West Africa.