Development of Multispectral Indices for Organic Fertilization Monitoring in Tomato Plants at Early Stages

[EN] A crop monitoring system was developed for monitor organic fertilization status of tomato plants at early stages. An automatic and nondestructive approach was used to analyze tomato plants with different levels of water-soluble organic fertilizer (3+5 NK) and vermicompost. The evaluation system...

ver descrição completa

Detalhes bibliográficos
Autor: Cardim Ferreira Lima, Matheus
Formato: tesis de maestría
Fecha de publicación:2019
País:España
Recursos:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:riunet.upv.es:10251/129471
Acesso em linha:https://riunet.upv.es/handle/10251/129471
Access Level:acceso abierto
Palavra-chave:Imagen multiespectral
Visión computacional
Agricultura de precisión
Índices de vegetación.
Multispectral image
Computer vision
Precision agriculture
Vegetation indexes
PRODUCCION VEGETAL
Máster Universitario Erasmus Mundus en Sanidad Vegetal en Agricultura Sostenible/ European Master degree in Plant Health in Sustainable Cropping Systems-Màster Universitari Erasmus Mundus en Sanitat Vegetal en Agricultura Sostenible / European Master&apos
s Degree in Plant Health in Sustainable Cropping Systems
Descrição
Resumo:[EN] A crop monitoring system was developed for monitor organic fertilization status of tomato plants at early stages. An automatic and nondestructive approach was used to analyze tomato plants with different levels of water-soluble organic fertilizer (3+5 NK) and vermicompost. The evaluation system was composed by a multispectral camera with five lenses: green (550nm), red (660nm), red edge (735nm), near infrared (790nm) + 16MP RGB and a computational image processing system. The water-soluble fertilizer was applied weekly in four different treatments: (T0: 0 ml, T1: 6.25 ml, T2: 12.5 ml and T3: 25 ml) and the vermicompost was added in the 1st (T0: 0ml; T1: 75 ml; T2:150ml; T3: 300 ml) and in the 5th week (T0: 0ml; T1: 237,5 ml; T2:475ml; T3: 950 ml). The trial was conducted in a greenhouse and 192 images were taken with each lens. An plant segmentation algorithm was developed and several vegetation indexes were calculated: Normalized Difference Vegetation Index (NDVI), Green Normalized Difference Vegetation Index (GNDVI), Red-Edge Normalized Difference Vegetation Index (RENDVI), Nonlinear Vegetation Index (NLI), Optimized Soil Adjusted Vegetation Index (OSAVI), Green Ratio Vegetation Index (GRVI), Simple Ratio (SR), Modified Simple Ratio (MSR), Structure Intensive Pigment Index 2 (SPI2) and Leaf Chlorophyll Index (LCI). Multiple morphological features were obtained through image processing and the results were compared between treatments using Tukey¿s HSD test with 1% of probability. The morphological features such as, Area, Filled Area, Convex Area, Perimeter, Major Axis Length, Minor Axis Length and Equivalent Diameter revealed to be feasible to distinguish between the control and the organic fertilized plants despite the vegetation indexes. The system was developed in order to be assembled in a precision organic fertilization robotic platform.