Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

ReDivia Repositorio Digital del Instituto Valenciano de Investigaciones Agrarias
Publicación Contribución de congreso (ConferenceOutput)

Crop recognition at orchard level in Mediterranean conditions using time series of Sentinel-2 spectral indices

  • Moltó, Enrique
  • Izquierdo-Sanz, Héctor
The Valencian agricultural landscape is made up of a multitude of small-sized plots. Most remote sensing research has been focused on the discrimination of the large-scale land covers or uses, however, in Spain, crop recognition at plot level is crucial to develop management policies at different scales. This work describes preliminary work for constructing and validating a classification model for automatic identification of crops in an agricultural zone where different species of fruit trees are cultivated. The model is based on the analysis of time series of two spectral indices obtained from Sentinel-2 images and using a Fourier decomposition approach. A two-step classification procedure is proposed: a random forest classifier at pixel level, followed by a plot classification. The overall accuracy achieved was 84.18% and the Kappa index was 0.71.
ReDivia Repositorio Digital del Instituto Valenciano de Investigaciones Agrarias
Publicación Contribución de congreso (ConferenceOutput)

Estimación de los niveles nutricionales de hojas de cítricos mediante análisis no destructivo aplicando técnicas de aprendizaje automático

  • Miralles, Guillem
  • Rodríguez-Carretero, Isabel
  • Cubero, Sergio
  • Martínez, Marcelino
  • Mateo, Fernando
  • Albert, Francisco
  • Quinones, Ana
  • Blasco, José
  • Gómez-Sanchis, Juan
One of the characteristics of all living beings is that adequate nutrition has a positive impact on health. In the case of plants, and specifically in fruit trees, adequate nutrition is also essential for them to grow healthy and produce fruits in the highest quantity and quality possible. Therefore, optimal nutrition is key for any farmer. However, excessive use of fertilisers can harm the environment and be a waste of resources for farmers. One of the keys to achieving adequate fertilisation is an accurate diagnosis of the nutritional status of the tree. Traditionally, this diagnosis is made by destructive ionomics analysis, which represents a high economic cost and a delay in obtaining the results. This work proposes Vis-NIR hyperspectral imaging and machine learning regression models to estimate the concentrations of macronutrients (N, P, K, and Ca) and micronutrients (Mn and Fe) in citrus leaves. The methodology involved the application of several machine learning regression methods (linear regression, partial least squares, random forest, support vector regression, and Ada Boost). Data were normalised with standard normal variable (SNV), and principal component analysis (PCA) was used to reduce dimensionality. The results were promising in estimating nutrients with R2 greater than 0,50 in all cases, especially nitrogen (R2 of 0.77).
ReDivia Repositorio Digital del Instituto Valenciano de Investigaciones Agrarias
Publicación Contribución de congreso (ConferenceOutput)

Allometric relationships for biomass estimation of persimmon trees using a field robot, LiDAR and photogrammetry

  • Rodríguez, Alejandro
  • Morales, J.
  • rey, B.
  • Cubero, Sergio
  • Aleixos, Nuria
  • Blasco, José
Chemical analyses of plants for nutritional diagnoses are destructive, complex, and timeconsuming. Fast and non-destructive alternatives based on optical sensors can perform the nutritional status of plants in the field [1]. While data captured by spectral sensors can be used to determine the nutritional status, the actual needs are influenced by other factors, such as the size of the plant or growth evolution. Allometric measurements, such as height, diameter, or canopy volume, along with the study of their temporal evolution, can provide valuable information. The use of 3D sensors can optimise this process, with one of the most widely used sensors being LiDAR [2]. Another technique is photogrammetry, which generates a 3D model from photographs. The aim was to develop non-destructive tools for determining allometric measurements of a persimmon crop at various phenological stages using LiDAR and photogrammetry on a robotic platform.