Matteo Ciotola
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Matteo Ciotola has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Giuseppe Scarpa
Department of Engineering, University Parthenope, Naples, Italy
15 shared publications - Giuseppe Guarino
Department of Electrical Engineering and Information Technology, University Federico II, Naples, Italy
9 shared publications - Gemine Vivone
Institute of Methodologies for Environmental Analysis, National Research Council, CNR-IMAA, Tito, Italy
6 shared publications - Giovanni Poggi
Department of Electrical Engineering and Information Technology, University Federico II, Naples, Italy
6 shared publications - Dino Ienco
INRAE, UMR TETIS, University of Montpellier, Montpellier, France
2 shared publications - Raffaele Gaetano
CIRAD, Centre de Coopération Internationale en Recherche Agronomique pour le Développement, UMR TETIS, Montpellier, Paris, France
2 shared publications
Latest Publications
- Open Access
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 15854-15871
A Novel Deep Learning Framework Combining Optical-SAR Data to Generate Gap-Free 20 m LAI Time Series for Crop Monitoring
The Leaf Area Index (LAI) is a critical indicator of crop growth, making high-resolution, gap-free LAI data of paramount importance for crop monitoring. However, the availability of optical data is significantly reduced due to cloud contamination. Conversely, SAR data can be acquired under all weather conditions. Additionally, both Sentinel-2 optical data and Sentinel-1 SAR data provide high spatial resolution and high temporal frequency. Therefore, with the aim of generating gap-free LAI time series at both high spatial and temporal resolutions, we propose a novel deep learning framework that combines Sentinel-2 optical and Sentinel-1 SAR data. This framework employs a Convolutional Neural Network (CNN) backbone to incorporate spatial context into LAI estimation. It implements a year-to-acquisition-date training strategy to enhance estimation accuracy. First, the model is pre-trained on data collected over a one-year time series. Then, it is progressively fine-tuned for each specific acquisition date on which LAI estimation is required. These components are integrated into an operational workflow for automated generation of gap-free LAI time series data. We validate the proposed framework through a comprehensive evaluation across diverse climate types, crop percentages, and gap length scenarios. Results demonstrate the superior performance of the proposed framework ($\bm R^{2}$: 0.86, RMSE: 0.26, $\bm |Bias|$: 0.07, SSIM: 0.85) compared to seven standard optical-based gap-filling approaches and two deep learning–based methods that integrate optical and SAR data, with gap length being the primary factor affecting performance. Furthermore, the framework shows robust LAI reconstruction under undetected residual clouds, ensuring accurate temporal and spatial information over croplands.
- 31 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 15854-15871
Open AccessA Novel Deep Learning Framework Combining Optical-SAR Data to Generate Gap-Free 20 m LAI Time Series for Crop Monitoring
The Leaf Area Index (LAI) is a critical indicator of crop growth, making high-resolution, gap-free LAI data of paramount importance for crop monitoring. However, the availability of optical data is significantly reduced due to cloud contamination. Conversely, SAR data can be acquired under all weather conditions. Additionally, both Sentinel-2 optical data and Sentinel-1 SAR data provide high spatial resolution and high temporal frequency. Therefore, with the aim of generating gap-free LAI time series at both high spatial and temporal resolutions, we propose a novel deep learning framework that combines Sentinel-2 optical and Sentinel-1 SAR data. This framework employs a Convolutional Neural Network (CNN) backbone to incorporate spatial context into LAI estimation. It implements a year-to-acquisition-date training strategy to enhance estimation accuracy. First, the model is pre-trained on data collected over a one-year time series. Then, it is progressively fine-tuned for each specific acquisition date on which LAI estimation is required. These components are integrated into an operational workflow for automated generation of gap-free LAI time series data. We validate the proposed framework through a comprehensive evaluation across diverse climate types, crop percentages, and gap length scenarios. Results demonstrate the superior performance of the proposed framework ($\bm R^{2}$: 0.86, RMSE: 0.26, $\bm |Bias|$: 0.07, SSIM: 0.85) compared to seven standard optical-based gap-filling approaches and two deep learning–based methods that integrate optical and SAR data, with gap length being the primary factor affecting performance. Furthermore, the framework shows robust LAI reconstruction under undetected residual clouds, ensuring accurate temporal and spatial information over croplands.
- 20 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 23 (1545598X) , 1-5
SAHARA: Heterogeneous Semi-Supervised Transfer Learning With Adversarial Adaptation and Dynamic Pseudo-Labeling
Semi-supervised domain adaptation (SSDA) aims to transfer knowledge from a labeled source domain to a scarcely labeled target domain, despite distribution shifts. The challenge becomes greater when source and target data differ in acquisition modality, as in remote sensing where variations in sensor type (e.g., optical versus radar), spectral properties [e.g., RGB versus multispectral (MS)], or spatial resolution are common. This challenging scenario, known as semi-supervised heterogeneous domain adaptation (SSHDA), requires learning across modalities with limited target labels. In this work, we propose semi-supervised adaptation in heterogeneous domains via conditional adversarial representation disentanglement and adaptive pseudo-labeling (SAHARA), a new method for SSHDA that combines conditional adversarial feature adaptation with dynamic pseudo-labeling to learn domain-invariant features and handle extremely scarce target annotations. Experiments on two heterogeneous remote sensing benchmarks for scene classification, conducted with both convolutional and transformer-based backbones, demonstrate that SAHARA consistently outperforms existing SSHDA and semi-supervised methods. The code is available at https://github.com/giu-guarino/SAHARA
- 20 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 23 (1545598X) , 1-5
SAHARA: Heterogeneous Semi-Supervised Transfer Learning With Adversarial Adaptation and Dynamic Pseudo-Labeling
Semi-supervised domain adaptation (SSDA) aims to transfer knowledge from a labeled source domain to a scarcely labeled target domain, despite distribution shifts. The challenge becomes greater when source and target data differ in acquisition modality, as in remote sensing where variations in sensor type (e.g., optical versus radar), spectral properties [e.g., RGB versus multispectral (MS)], or spatial resolution are common. This challenging scenario, known as semi-supervised heterogeneous domain adaptation (SSHDA), requires learning across modalities with limited target labels. In this work, we propose semi-supervised adaptation in heterogeneous domains via conditional adversarial representation disentanglement and adaptive pseudo-labeling (SAHARA), a new method for SSHDA that combines conditional adversarial feature adaptation with dynamic pseudo-labeling to learn domain-invariant features and handle extremely scarce target annotations. Experiments on two heterogeneous remote sensing benchmarks for scene classification, conducted with both convolutional and transformer-based backbones, demonstrate that SAHARA consistently outperforms existing SSHDA and semi-supervised methods. The code is available at https://github.com/giu-guarino/SAHARA
- Open Access
- 27 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-19
Zero-Shot Hyperspectral Pansharpening Using Hysteresis-Based Tuning for Spectral Quality Control
Hyperspectral (HS) pansharpening has received much attention in recent years due to technological and methodological advances that open the door to new application scenarios. However, research on this topic is only now gaining momentum. The most popular methods are still borrowed from the more mature field of multispectral pansharpening and often overlook the unique challenges posed by HS data fusion, such as: 1) the very large number of bands; 2) the overwhelming noise in selected spectral ranges; 3) the significant spectral mismatch between panchromatic (PAN) and HS components; and 4) a typically high resolution ratio. Imprecise data modeling especially affects spectral fidelity. Even state-of-the-art (SotA) methods perform well in certain spectral ranges and much worse in others, failing to ensure consistent quality across all bands, with the risk of generating unreliable results. Here, we propose an HS pansharpening method that explicitly addresses this problem and ensures uniform spectral quality. To this end, a single lightweight neural network is used, with weights that adapt on the fly to each band. During fine-tuning, the spatial loss is turned on and off to ensure a fast convergence of the spectral loss to the desired level, according to a hysteresis-like dynamic. Furthermore, the spatial loss itself is appropriately redefined to account for nonlinear dependencies between PAN and spectral bands. Overall, the proposed method is fully unsupervised, with no prior training on external data, flexible, and low-complexity. Experiments on a recently published benchmarking toolbox show that it ensures excellent sharpening quality, competitive with the SotA, consistently across all bands. The software code and the full set of results are shared online on https://github.com/giu-guarino/rho-PNN
- 27 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-19
Open AccessZero-Shot Hyperspectral Pansharpening Using Hysteresis-Based Tuning for Spectral Quality Control
Hyperspectral (HS) pansharpening has received much attention in recent years due to technological and methodological advances that open the door to new application scenarios. However, research on this topic is only now gaining momentum. The most popular methods are still borrowed from the more mature field of multispectral pansharpening and often overlook the unique challenges posed by HS data fusion, such as: 1) the very large number of bands; 2) the overwhelming noise in selected spectral ranges; 3) the significant spectral mismatch between panchromatic (PAN) and HS components; and 4) a typically high resolution ratio. Imprecise data modeling especially affects spectral fidelity. Even state-of-the-art (SotA) methods perform well in certain spectral ranges and much worse in others, failing to ensure consistent quality across all bands, with the risk of generating unreliable results. Here, we propose an HS pansharpening method that explicitly addresses this problem and ensures uniform spectral quality. To this end, a single lightweight neural network is used, with weights that adapt on the fly to each band. During fine-tuning, the spatial loss is turned on and off to ensure a fast convergence of the spectral loss to the desired level, according to a hysteresis-like dynamic. Furthermore, the spatial loss itself is appropriately redefined to account for nonlinear dependencies between PAN and spectral bands. Overall, the proposed method is fully unsupervised, with no prior training on external data, flexible, and low-complexity. Experiments on a recently published benchmarking toolbox show that it ensures excellent sharpening quality, competitive with the SotA, consistently across all bands. The software code and the full set of results are shared online on https://github.com/giu-guarino/rho-PNN
- 16 June 2025
- Published by Springer Nature
- p. 77-91
A Spectral-Preserving Zero-Shot Technique for Hyperspectral Pansharpening
- 16 June 2025
- Published by Springer Nature
- p. 77-91
A Spectral-Preserving Zero-Shot Technique for Hyperspectral Pansharpening
- Open Access
- 7 June 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (12) , 1983
A Comprehensive Benchmarking Framework for Sentinel-2 Sharpening: Methods, Dataset, and Evaluation Metrics
The advancement of super-resolution and sharpening algorithms for satellite images has significantly expanded the potential applications of remote sensing data. In the case of Sentinel-2, despite significant progress, the lack of standardized datasets and evaluation protocols has made it difficult to fairly compare existing methods and advance the state of the art. This work introduces a comprehensive benchmarking framework for Sentinel-2 sharpening, designed to address these challenges and foster future research. It analyzes several state-of-the-art sharpening algorithms, selecting representative methods ranging from traditional pansharpening to ad hoc model-based optimization and deep learning approaches. All selected methods have been re-implemented within a consistent Python-based (Version 3.10) framework and evaluated on a suitably designed, large-scale Sentinel-2 dataset. This dataset features diverse geographical regions, land cover types, and acquisition conditions, ensuring robust training and testing scenarios. The performance of the sharpening methods is assessed using both reference-based and no-reference quality indexes, highlighting strengths, limitations, and open challenges of current state-of-the-art algorithms. The proposed framework, dataset, and evaluation protocols are openly shared with the research community to promote collaboration and reproducibility.
- 7 June 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (12) , 1983
Open AccessA Comprehensive Benchmarking Framework for Sentinel-2 Sharpening: Methods, Dataset, and Evaluation Metrics
The advancement of super-resolution and sharpening algorithms for satellite images has significantly expanded the potential applications of remote sensing data. In the case of Sentinel-2, despite significant progress, the lack of standardized datasets and evaluation protocols has made it difficult to fairly compare existing methods and advance the state of the art. This work introduces a comprehensive benchmarking framework for Sentinel-2 sharpening, designed to address these challenges and foster future research. It analyzes several state-of-the-art sharpening algorithms, selecting representative methods ranging from traditional pansharpening to ad hoc model-based optimization and deep learning approaches. All selected methods have been re-implemented within a consistent Python-based (Version 3.10) framework and evaluated on a suitably designed, large-scale Sentinel-2 dataset. This dataset features diverse geographical regions, land cover types, and acquisition conditions, ensuring robust training and testing scenarios. The performance of the sharpening methods is assessed using both reference-based and no-reference quality indexes, highlighting strengths, limitations, and open challenges of current state-of-the-art algorithms. The proposed framework, dataset, and evaluation protocols are openly shared with the research community to promote collaboration and reproducibility.
- 17 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Magazine
- Vol. 13 (1) , 311-338
Hyperspectral Pansharpening: Critical review, tools, and future perspectives
Hyperspectral (HS) pansharpening consists of fusing a high-resolution panchromatic (PAN) band and a low-resolution HS image to obtain a new image with high resolution in both the spatial and spectral domains. These remote sensing products are valuable for a wide range of applications, driving ever-growing research efforts. Nonetheless, results still do not meet application demands. In part, this comes from the technical complexity of the task: compared to multispectral (MS) pansharpening, many more bands are involved, in a spectral range only partially covered by the PAN component and with overwhelming noise. However, another major limiting factor is the absence of a comprehensive framework for the rapid development and accurate evaluation of new methods. This article attempts to address this issue. We started by designing a dataset large and diverse enough to allow reliable training (for data-driven methods) and testing of new methods. Then, we selected a set of state-of-the-art (SoTA) methods, following different approaches characterized by promising performance, and reimplemented them in a single PyTorch framework. Finally, we carried out a critical comparative analysis of all methods, using the most accredited quality indicators. The analysis highlights the main limitations of current solutions in terms of spectral/spatial quality and computational efficiency, and it suggests promising research directions. To ensure full reproducibility of the results and support future research, the framework (comprising methods and assessment codes and dataset setup procedures) is shared on https://github.com/matciotola/hyperspectral_pansharpening_toolbox , as a single Python-based reference benchmark toolbox.
- 17 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Magazine
- Vol. 13 (1) , 311-338
Hyperspectral Pansharpening: Critical review, tools, and future perspectives
Hyperspectral (HS) pansharpening consists of fusing a high-resolution panchromatic (PAN) band and a low-resolution HS image to obtain a new image with high resolution in both the spatial and spectral domains. These remote sensing products are valuable for a wide range of applications, driving ever-growing research efforts. Nonetheless, results still do not meet application demands. In part, this comes from the technical complexity of the task: compared to multispectral (MS) pansharpening, many more bands are involved, in a spectral range only partially covered by the PAN component and with overwhelming noise. However, another major limiting factor is the absence of a comprehensive framework for the rapid development and accurate evaluation of new methods. This article attempts to address this issue. We started by designing a dataset large and diverse enough to allow reliable training (for data-driven methods) and testing of new methods. Then, we selected a set of state-of-the-art (SoTA) methods, following different approaches characterized by promising performance, and reimplemented them in a single PyTorch framework. Finally, we carried out a critical comparative analysis of all methods, using the most accredited quality indicators. The analysis highlights the main limitations of current solutions in terms of spectral/spatial quality and computational efficiency, and it suggests promising research directions. To ensure full reproducibility of the results and support future research, the framework (comprising methods and assessment codes and dataset setup procedures) is shared on https://github.com/matciotola/hyperspectral_pansharpening_toolbox , as a single Python-based reference benchmark toolbox.
- 15 October 2024
- Published by Springer Nature
- p. 233-278
Unsupervised Pansharpening Using ConvNets
- Open Access
- 16 August 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (16) , 3014
An Unsupervised CNN-Based Pansharpening Framework with Spectral-Spatial Fidelity Balance
In recent years, deep learning techniques for pansharpening multiresolution images have gained increasing interest. Due to the lack of ground truth data, most deep learning solutions rely on synthetic reduced-resolution data for supervised training. This approach has limitations due to the statistical mismatch between real full-resolution and synthetic reduced-resolution data, which affects the models’ generalization capacity. Consequently, there has been a shift towards unsupervised learning frameworks for pansharpening deep learning-based techniques. Unsupervised schemes require defining sophisticated loss functions with at least two components: one for spectral quality, ensuring consistency between the pansharpened image and the input multispectral component, and another for spatial quality, ensuring consistency between the output and the panchromatic input. Despite promising results, there has been limited investigation into the interaction and balance of these loss terms to ensure stability and accuracy. This work explores how unsupervised spatial and spectral consistency losses can be reliably combined preserving the outocome quality. By examining these interactions, we propose a general rule for balancing the two loss components to enhance the stability and performance of unsupervised pansharpening models. Experiments on three state-of-the-art algorithms using WorldView-3 images demonstrate that methods trained with the proposed framework achieve good performance in terms of visual quality and numerical indexes.
- 16 August 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (16) , 3014
Open AccessAn Unsupervised CNN-Based Pansharpening Framework with Spectral-Spatial Fidelity Balance
In recent years, deep learning techniques for pansharpening multiresolution images have gained increasing interest. Due to the lack of ground truth data, most deep learning solutions rely on synthetic reduced-resolution data for supervised training. This approach has limitations due to the statistical mismatch between real full-resolution and synthetic reduced-resolution data, which affects the models’ generalization capacity. Consequently, there has been a shift towards unsupervised learning frameworks for pansharpening deep learning-based techniques. Unsupervised schemes require defining sophisticated loss functions with at least two components: one for spectral quality, ensuring consistency between the pansharpened image and the input multispectral component, and another for spatial quality, ensuring consistency between the output and the panchromatic input. Despite promising results, there has been limited investigation into the interaction and balance of these loss terms to ensure stability and accuracy. This work explores how unsupervised spatial and spectral consistency losses can be reliably combined preserving the outocome quality. By examining these interactions, we propose a general rule for balancing the two loss components to enhance the stability and performance of unsupervised pansharpening models. Experiments on three state-of-the-art algorithms using WorldView-3 images demonstrate that methods trained with the proposed framework achieve good performance in terms of visual quality and numerical indexes.
- Open Access
- 4 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-18
Band-Wise Hyperspectral Image Pansharpening Using CNN Model Propagation
Hyperspectral (HS) pansharpening has received a growing interest in the last few years as testified by a large number of research papers and challenges. It consists in a pixel-level fusion between a lower resolution HS datacube and a higher resolution single-band image, the panchromatic (PAN) image, with the goal of providing an HS datacube at PAN resolution. Due to their powerful representational capabilities, deep learning models have succeeded to provide unprecedented results on many general-purpose image processing tasks. However, when moving to domain-specific problems, as in this case, the advantages with respect to traditional model-based approaches are much lesser clear-cut due to several contextual reasons. Scarcity of training data, lack of ground truth (GT), and data shape variability are some such factors that limit the generalization capacity of the state-of-the-art deep learning networks for HS pansharpening. To cope with these limitations, in this work, we propose a new deep learning method, which inherits a simple single-band unsupervised pansharpening model nested in a sequential band-wise adaptive scheme, where each band is pansharpened refining the model tuned on the preceding one. By doing so, a simple model is propagated along the wavelength dimension, adaptively and flexibly, with no need to have a fixed number of spectral bands, and, with no need to dispose of large, expensive, and labeled training datasets. The proposed method achieves very good results on our datasets, outperforming both traditional and deep learning reference methods. The implementation of the proposed method can be found at https://github.com/giu-guarino/R-PNN .
- 4 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-18
Open AccessBand-Wise Hyperspectral Image Pansharpening Using CNN Model Propagation
Hyperspectral (HS) pansharpening has received a growing interest in the last few years as testified by a large number of research papers and challenges. It consists in a pixel-level fusion between a lower resolution HS datacube and a higher resolution single-band image, the panchromatic (PAN) image, with the goal of providing an HS datacube at PAN resolution. Due to their powerful representational capabilities, deep learning models have succeeded to provide unprecedented results on many general-purpose image processing tasks. However, when moving to domain-specific problems, as in this case, the advantages with respect to traditional model-based approaches are much lesser clear-cut due to several contextual reasons. Scarcity of training data, lack of ground truth (GT), and data shape variability are some such factors that limit the generalization capacity of the state-of-the-art deep learning networks for HS pansharpening. To cope with these limitations, in this work, we propose a new deep learning method, which inherits a simple single-band unsupervised pansharpening model nested in a sequential band-wise adaptive scheme, where each band is pansharpened refining the model tuned on the preceding one. By doing so, a simple model is propagated along the wavelength dimension, adaptively and flexibly, with no need to have a fixed number of spectral bands, and, with no need to dispose of large, expensive, and labeled training datasets. The proposed method achieves very good results on our datasets, outperforming both traditional and deep learning reference methods. The implementation of the proposed method can be found at https://github.com/giu-guarino/R-PNN .
- Open Access
Band-wise Hyperspectral Image Pansharpening using CNN Model Propagation
Hyperspectral pansharpening is receiving a growing interest since the last few years as testified by a large number of research papers and challenges. It consists in a pixel-level fusion between a lower-resolution hyperspectral datacube and a higher-resolution single-band image, the panchromatic image, with the goal of providing a hyperspectral datacube at panchromatic resolution. Thanks to their powerful representational capabilities, deep learning models have succeeded to provide unprecedented results on many general purpose image processing tasks. However, when moving to domain specific problems, as in this case, the advantages with respect to traditional model-based approaches are much lesser clear-cut due to several contextual reasons. Scarcity of training data, lack of ground-truth, data shape variability, are some such factors that limit the generalization capacity of the state-of-the-art deep learning networks for hyperspectral pansharpening. To cope with these limitations, in this work we propose a new deep learning method which inherits a simple single-band unsupervised pansharpening model nested in a sequential band-wise adaptive scheme, where each band is pansharpened refining the model tuned on the preceding one. By doing so, a simple model is propagated along the wavelength dimension, adaptively and flexibly, with no need to have a fixed number of spectral bands, and, with no need to dispose of large, expensive and labeled training datasets. The proposed method achieves very good results on our datasets, outperforming both traditional and deep learning reference methods. The implementation of the proposed method can be found on https://github.com/giu-guarino/R-PNN
Open AccessBand-wise Hyperspectral Image Pansharpening using CNN Model Propagation
Hyperspectral pansharpening is receiving a growing interest since the last few years as testified by a large number of research papers and challenges. It consists in a pixel-level fusion between a lower-resolution hyperspectral datacube and a higher-resolution single-band image, the panchromatic image, with the goal of providing a hyperspectral datacube at panchromatic resolution. Thanks to their powerful representational capabilities, deep learning models have succeeded to provide unprecedented results on many general purpose image processing tasks. However, when moving to domain specific problems, as in this case, the advantages with respect to traditional model-based approaches are much lesser clear-cut due to several contextual reasons. Scarcity of training data, lack of ground-truth, data shape variability, are some such factors that limit the generalization capacity of the state-of-the-art deep learning networks for hyperspectral pansharpening. To cope with these limitations, in this work we propose a new deep learning method which inherits a simple single-band unsupervised pansharpening model nested in a sequential band-wise adaptive scheme, where each band is pansharpened refining the model tuned on the preceding one. By doing so, a simple model is propagated along the wavelength dimension, adaptively and flexibly, with no need to have a fixed number of spectral bands, and, with no need to dispose of large, expensive and labeled training datasets. The proposed method achieves very good results on our datasets, outperforming both traditional and deep learning reference methods. The implementation of the proposed method can be found on https://github.com/giu-guarino/R-PNN