4.3.2Artificial neural networks Neural network is another supervised classification method that has been adopted by many researchers[65,75–77], due to its nonparametric nature, arbitrary decision boundary, etc. Multilayer perceptron is the most popular type of neural network in image classification[65...
(2004) combined ground-based hyperspectral imaging with self-organizing maps (unsupervised neural networks) to detect R. reniformis in cotton, and obtained a classification accuracy between 83% and 97%. This approach of classifying infested plants was further developed by Doshi (2007). Rupe et al...
Other approaches have been proposed but only applied theoretically. Convolutional neural networks have shown promising results for classification of hyperspectral images. Chaczko et al. [76] manipulated an existing collection of hyperspectral images of textiles to mimic MP. A neutral network was applied ...
plant traits from the plant’s spectral signature are regularized least squares such as lasso and ridge regression, partial least squares regression, decision-tree-based methods such as random forest, extreme gradient boosting or cubist models, Gaussian process regression or also artificial neural ...
In contrast, the number of hyperspectral infrared profiles assimilated is much higher since the retrieved profiles can be assimilated in some partly cloudy scenes due to profile coupling other data, such as microwave or neural networks, as first guesses to the retrieval process. As the operational ...
intrinsic capability of neural networks to parallelize the whole hyperspectral unmixing process (Pérez et al., 1999). The results shown in this work indicate that neural network models are able to find clusters of closely related hyperspectral signatures, and thus can be used as a powerful tool ...
structure, respectively. SIF, WDI, and RVI indicate solar-induced fluorescence, water deficit index, and radar vegetation index, respectively. MLR, ANN, and RF refer to the ensemble models multiple linear regression, artificial neural network, and random forest. GPC indicates grain protein ...
They range in complexity level from simple models such as linear discriminant analysis to complex machine learning techniques and neural networks. Although statistical models are very accurate locally, they can only be applied to the areas they were developed for and are not readily applicable to ...
Convolutional neural networks (CNNs) are the go-to model for hyperspectral image (HSI) classification because of the excellent locally contextual modeling ability that is beneficial to spatial and spectral feature extraction. However, CNNs with a limited receptive field pose challenges for modeling lon...
Some techniques are based on Convolutional Neural Networks (CNN). In [31], the authors proposed a Spatial-Spectral Reconstruction Network (SSR-Net) trained by optimizing both spatial and spectral edge losses. In [32], a new loss function called RMSE, angle and Laplacian (RAP) to reduce the...