Good day fellow researchers, am about to compose the downloaded satellite image of Landsat images for the purpose of LULC changes to evaluate deforestation and forest degradation. before the commencement, i knew of Geometric and Radiometric correction as pre-processing image techniques before any classification approach. I observed and noted from my past experienced that, it seems not to be any differences if i carry out the correction process before the classification. The work at hand is my PhD work and is of multiples images because of the area coverage e.g (Nigeria). The question is as follow, do i go straight to composition of bands or i need to first carry out image correction approaches (Geo and Radiometric correction of the image or bands) before i can subject it into LULC Classification. Thanks