Análisis e implementación de un sistema de detección de la plaga monilia en cultivos de cacao basado en procesamiento de imágenes multiespectrales en la ciudad del Sacha.

Abstract

The objective of this degree work was the analysis and implementation of a system of detection for the monilia plague, in cocoa crops in Sacha, based on the multispectral image processing. Enough information was collected and, documentation for the analysis of the factors and parameters required in the treatment of information, so it was necessary to develop it in two stages. The first consisted of acquisition of images of cocoa plantations, through the Micasense Camera RedEdge-MX, which was coupled to a Dji Phantom 4 model drone, this platform unmanned area served to carry out the respective overflight on the cocoa plots at a height of 25 meters, with the purpose of collecting data and information. The second stage was directed in the treatment of multispectral images using the COCOA application PLANTATION ANALYZER, where each of the samples were taken and were applied correction, calibration, filtering processes and segmentation, in order to calculate the indices of vegetation NDVI, NDRE and SAVI to obtain as a result representative image in scale of colors, of easy understanding, that indicate the state of health of the crops previously mentioned. For the statistical analysis, the Pearson chi-square statistical test was used, in order to validate the data obtained with the QGIS program, yielding homogeneous results among them. It is concluded that the detection of the monilia plague can be detected through the vegetation indices, being a guideline for the assessment of the health status of crops in the place. It is recommended to continue the study of other diseases and pests that affect cocoa in future research projects.