SciELO - Scientific Electronic Library Online

vol.98 número1Análisis de la temperatura y humedad del suelo y rendimiento cultural en dos sistemas de manejo del cañaveral en Tucumán - ArgentinaPosibilidades de generación distribuida de energía eléctrica mediante gasificación de las principales biomasas residuales de Tucumán, Argentina índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




  • No hay articulos citadosCitado por SciELO

Links relacionados

  • No hay articulos similaresSimilares en SciELO


Revista industrial y agrícola de Tucumán

versión On-line ISSN 1851-3018


FANDOS, Carmina; GANDINI, Marcelo L.  y  SORIA, Federico J.. Efficacy of different methods of classification in images landsat 8 OLI and TIRS for inventories in the sugarcane planted area of the Lules department . Tucuman, R. Argentina. Rev. ind. agric. Tucumán [online]. 2021, vol.98, n.1, pp.21-31. ISSN 1851-3018.

Information about sugarcane planted area and production at the beginning of the harvest is fundamental for the planning of harvesting strategies, logistics and marketing. Their estimation, using technologies based only on field surveys, involves considerable effort, and high costs. Spatial remote sensing allows to reduce costs in land use/land cover determinations. The Operational Land Imager (OLI) and the Thermal Infrared Sensor (TIRS) on board of Landsat 8 satellite, with their bands in the range from visible to thermal infrared, constitute an alternative for the discrimination of vegetation covers, taking into account that vegetation has a reduced spectral response in the visible, higher in the near infrared and lower in the mid infrared; and that vegetation covers present lower temperatures than their environment during the day, which facilitates their differentiation from other covers. The objectives were to evaluate different multispectral classification techniques using the red, near infrared and middle infrared bands of the OLI sensor, and the thermal bands of TIRS sensor, and as a product of the analysis, estimate crop area and sugarcane production by GIS processing. The study of bands 4, 5 and 6 showed that band 5 has the greatest separability between coverages and between sugarcane production levels. The combination of mid-infrared and thermal spectral bands achieved improved classification accuracy. Precision reached in the classification of different production levels indicates that the maximum likelihood classification restricted by band 10 would be adequate for the identification and quantification of sugarcane crops and production levels, in the case of similar proportions of the three production levels, or with a predominance of the low production level. The classification by decision tree of band 5 restricted by band 10 would be the most appropriate for areas with predominantly medium and high levels of production. 

Palabras clave : multispectral classification; decision tree; GIS.

        · resumen en Español     · texto en Español     · Español ( pdf )


Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons