Diseño y Construcción de un Sistema Multiespectral para Determinar el Índice de Vegetación de Diferencia Normalizada de las Plantas en un Cultivo de Área menor a 100 m^2.

Abstract

In this Curricular Integration Work, the methodology for implementing a portable prototype capable of identifying the NDVI of plants is detailed, which consists of an optical system, a raspberry pi, a battery charging system, and a touch screen. The developed prototype captures spectral photographs, thanks to an optical system, and processes them for the obtaining of the NDVI vegetation index. The optical system of the prototype is made up of a camera which has been modified by removing its IR filter, replacing it with a specialized filter placed in front of the camera's sensor with the purpose of acquiring a spectral image. Subsequently, with the purpose of processing the spectral image and obtaining the NDVI index, a programming code has been developed in the Python language, which is executed on a raspberry pi. The NDVI index obtained with the programming code is validated through the use of the ArcGIS software. Additionally, the implemented prototype has a user interface (GUI) on a touch screen that allows the user to acquire spectral images, also obtain the vegetation index with its respective NDVI image. On the other hand, in this document, the technical characteristics of the commercial electronic modules used for the system to work in a portable way and that is rechargeable are detailed. Similarly, the process for creating a graphical interface that is capable of working with programming in the Python language is presented. Finally, the study is complemented with theoretical research on plant reflectance and the theory of digital image processing.