The estimation of sugarcane yield with remote sensing can be done through information collected by sensors on satellites, aircrafts and recently in drones, which record the inte- raction between electromagnetic radiation and the sugarcane canopy in multiple spectral bands. With the spectral information collected from these bands different vegetation in- dices can be calculated and related to biophysical variables, trying to predict crop yield. For 20 years, the Centro de Investigación de la Caña de Aź̧ar de Colombia (Cenica) has worked with satellite images, being Landsat 5 (TM), Landsat 7 (ETM +), Landsat 8 (OLI) and Terra EOS AM-1 (MODIS), those satellites used for the monitoring and study of sugarcane in the sugar agroindustrial sector of the country, images captured by ultra-light aircraft and more recently to the capture of visible and multispectral images using RPAS, so that the evaluation of vegetation indices for the yield estimation can be focused at a plot level detail. This paper aims to present the results that have been obtained using different sensors on board different platforms in the early sugarcane yield estimation.