Bibliographic citations
Rivadeneyra, B., (2022). Identificación de anomalías en hojas de palto Hass usando procesamiento de imágenes térmicas y RGB obtenidas a nivel foliar y de parcela [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/26882
Rivadeneyra, B., Identificación de anomalías en hojas de palto Hass usando procesamiento de imágenes térmicas y RGB obtenidas a nivel foliar y de parcela [Tesis]. PE: Universidad Nacional de Ingeniería; 2022. http://hdl.handle.net/20.500.14076/26882
@misc{renati/712836,
title = "Identificación de anomalías en hojas de palto Hass usando procesamiento de imágenes térmicas y RGB obtenidas a nivel foliar y de parcela",
author = "Rivadeneyra Bustamante, Bruno Daniel",
publisher = "Universidad Nacional de Ingeniería",
year = "2022"
}
Among the main agricultural export products of Peru is Persea americana (avocado). However, the occurrence of pests and deficiencies cause a decrease in crop yield. Therefore, the detection of anomalous visual characteristics has been studied for thermal and multispectral images of avocado trees at the foliar and plot level. In the present work, a method for the detection of such anomalies through the processing of thermal and RGB images acquired by using a thermographic camera is proposed. In the results obtained at the leaf level, an SVM classification model is designed with an average accuracy of 82.67% in ten experiments performed with a reduced set of images. At the plot level, an SVM-OVA model is designed, from which a low accuracy of 25.34% is obtained, however, a better estimation of the color index Ipca is proposed in addition to considering fixed but arbitrary heights for the acquisition of the images. Finally, a guide of procedures for the acquisition of thermal and RGB images of the Hass avocado leaf and canopy is elaborated, taking into account the main characteristics of the measuring instrument (thermographic camera), as well as the experiences of its use and handling in the field throughout the working days.
This item is licensed under a Creative Commons License