Segmentación de zonas de cultivo de caña panelera utilizando técnicas de Region Growing con imágenes satelitales Sentinel 2A

Remote sensing and satellite imagery are tools of great importance for agronomic and environmental sciences. One of the applications in which this type of techniques are used in agronomy is crop condition monitoring. For this purpose, the use of machine learning techniques to perform classification tasks is a very common option to meet this objective. However, a common problem that arises when trying to apply this type of solutions is the lack of a significant amount of data that allows to adequately teach the classifier the specific characteristics of the crop. Likewise, this problem is made even more difficult given the insufficient resolution of public satellite images that prevent manually delimiting the crop region with high precision. Therefore, the need to carry out field work to record the points that delimit the region constitutes a labor-intensive task with high costs that prevent scaling up the data collection process...

• Autores: González Rodríguez, Geovanny Andrés
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2020
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: spa
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/51479
• Acceso en línea: http://hdl.handle.net/1992/51479
• Palabra clave: Sensores remotos; Imágenes de detección a distancia; Aprendizaje automático (Inteligencia artificial); Caña de azúcar; Ingeniería
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/