The capability to generate realistic hyperspectral imagery plays a prominent role in applications to sensor and mission development as well as in the training of machine learning models. Yet, it is a challenging task due to the high dimensionality and complex spectral–spatial structure of the data. This paper proposes a novel unsupervised deep-learning framework for generating realistic hyperspectral imagery based on blind hyperspectral unmixing and denoising diffusion probabilistic models. First, the approach extracts both endmembers and abundance maps from hyperspectral data through a dictionary of hyperspectral unmixing algorithms. The extracted abundances are then used as inputs for a guided diffusion model, which serves as the generative framework with the goal of producing realistic synthetic abundance maps. Finally, the generation of synthetic hyperspectral images is accomplished by integrating the generated abundance maps with the extracted endmember set and by suitably conditioning the probabilistic formulation of the guided diffusion model as a function of the unmixing algorithms in the aforementioned dictionary. By combining a collection of blind linear unmixing techniques with the generative capabilities of diffusion models, the proposed methodology aims to address key challenges in simulating hyperspectral sensor outputs. The methodology was validated experimentally using real satellite hyperspectral imagery from the PRISMA mission of the Italian Space Agency. The results confirm the effectiveness of the approach in generating realistic synthetic hyperspectral images associated with various land-covers. The code is available at: https://github.com/martinapastorino/HSI_DDPM.

This paper tackles the semantic segmentation of zones affected by forest fires by the introduction of methods fusing multimodal imagery collected from unmanned aerial vehicles (UAVs) and satellite platforms. The multiresolution fusion task is especially challenging in this case because the difference between the involved spatial resolutions is very large – a situation that is normally not addressed by traditional multiresolution schemes. Two novel multiresolution fusion approaches, based on Bayesian and probabilistic graphical fusion models and integrated with a deep fully convolutional network and with the expectation-maximization algorithm, are proposed. The application is to a real case of fire zone mapping and management in the area of Marseille, France.

The availability of multimodal remotely sensed images calls for the development of methods capable to jointly exploit the information deriving from images acquired at different spatial resolutions, frequencies, and bands, taking advantage from their possible complementary features. This letter proposes to address this task in the case of multimission synthetic aperture radar (SAR) images, through a combination of fully convolutional networks (FCNs), hierarchical probabilistic graphical models (PGMs), and decision tree ensembles. The objective is to model the multimodal information collected at multiple spatial resolutions by distinct space missions with SAR payloads through the nonparametric formulation of FCNs and decision trees, and the spatial and multiresolution modeling capabilities of FCNs and hierarchical PGMs. The experimental validation is conducted with multimission SAR imagery acquired at X-, L-, and C-band, respectively, by COSMO-SkyMed, SAOCOM, and Sentinel-1 over Northern Italy. The results suggest the advantages of incorporating multifrequency radar acquisitions to reach accurate classification maps and the multimodal fusion capabilities of the proposed methodology.

This article presents a method for the automatic learning of the potentials of a stochastic model, in particular a conditional random field (CRF), in a non-parametric fashion. The proposed model is based on a neural architecture, in order to leverage the modeling capabilities of deep learning (DL) approaches to directly learn semantic and spatial information from the input data. Specifically, the methodology is based on fully convolutional networks and fully connected neural networks. The idea is to access the multiscale information intrinsically extracted in the intermediate layers of a fully convolutional network through the integration of fully connected neural networks at different scales, while favoring the interpretability of the hidden layers as posterior probabilities. The potentials of the CRF are learned through an additional convolutional layer, whose kernel models the local spatial information considered. The loss function is computed as a linear combination of cross-entropy losses, accounting for the multiscale and the spatial information. To evaluate the capabilities of the proposed approach for the semantic segmentation of remote sensing images, the experimental validation was conducted with the ISPRS 2-D semantic labeling challenge Vaihingen and Potsdam datasets and with the IEEE GRSS data fusion contest Zeebruges dataset. As the ground truths of these benchmark datasets are spatially exhaustive, they have been modified to approximate the spatially sparse ground truths common in real remote sensing applications. The results are significant, as the proposed approach obtains higher average classification accuracies than recent state-of-the-art techniques considered in this article. The code is available athttps://github.com/Ayana-Inria/CRFNet-RS.

The problem of the semantic segmentation of multimodal images is characterized by the challenge of jointly exploiting information deriving from images possibly acquired at different spatial resolutions, frequencies, and bands. This paper proposes to address this task in the case of multimission synthetic aperture radar (SAR) images, through a combination of fully convolutional networks (FCNs), hierarchical probabilistic graphical models (PGMs), and decision tree ensembles. The objective is to model the spatial and multiresolution information contained in multimodal remote sensing images collected by distinct space missions with SAR payloads. The experimental validation is conducted with COSMO-SkyMed and SAOCOM images over Northern Italy. The results show that the proposed methodology is capable to reach accurate classification maps from input multimission SAR imagery.

This paper presents a novel semantic segmentation method of very high resolution remotely sensed images based on fully convolutional networks (FCNs) and feedforward neural networks (FFNNs). The proposed model aims to exploit the intrinsic multiscale information extracted at different convolutional blocks in an FCN by the integration of FFNNs, thus incorporating information at different scales. The purpose is to obtain accurate classification results with realistic data sets characterized by sparse ground truth (GT) data by taking benefit from multiscale and long-range spatial information. The final loss function is computed as a linear combination of the weighted cross-entropy losses of the FFNNs and of the FCN. The modeling of spatial-contextual information is further addressed by the introduction of an additional loss term which allows to integrate spatial information between neighboring pixels. The experimental validation is conducted with the ISPRS 2D Semantic Labeling Challenge data set over the city of Vaihingen, Germany. The results are promising, as the proposed approach obtains higher average classification results than the state-of-the-art techniques considered, especially in the case of scarce, suboptimal GTs.

This paper addresses the semantic segmentation of synthetic aperture radar (SAR) images through the combination of ful-ly convolutional networks (FCN s), hierarchical probabilistic graphical models (PGMs), and decision tree ensembles. The idea is to incorporate long-range spatial information together with the multiresolution information extracted by FCN s, through the multiresolution graph topology on which hierar-chical PGMs can be efficiently formulated. The objective is to obtain accurate classification results with small datasets and reduce problems of spatial inconsistency. The experimental validation is conducted with several COSMO-SkyMed satel-lite images over Northern Italy. The results are significant, as the proposed method obtains more accurate classification results than the standard FCN s considered.

Deep learning (DL) is currently the dominant approach to image classification and segmentation, but the performances of DL methods are remarkably influenced by the quantity and quality of the ground truth (GT) used for training. In this article, a DL method is presented to deal with the semantic segmentation of very-high-resolution (VHR) remote-sensing data in the case of scarce GT. The main idea is to combine a specific type of deep convolutional neural networks (CNNs), namely fully convolutional networks (FCNs), with probabilistic graphical models (PGMs). Our method takes advantage of the intrinsic multiscale behavior of FCNs to deal with multiscale data representations and to connect them to a hierarchical Markov model (e.g., making use of a quadtree). As a consequence, the spatial information present in the data is better exploited, allowing a reduced sensitivity to GT incompleteness to be obtained. The marginal posterior mode (MPM) criterion is used for inference in the proposed framework. To assess the capabilities of the proposed method, the experimental validation is conducted with the ISPRS 2D Semantic Labeling Challenge datasets on the cities of Vaihingen and Potsdam, with some modifications to simulate the spatially sparse GTs that are common in real remote-sensing applications. The results are quite significant, as the proposed approach exhibits a higher producer accuracy than the standard FCNs considered and especially mitigates the impact of scarce GTs on minority classes and small spatial details.

In this paper, a novel method to deal with the semantic segmentation of very high resolution remote sensing data is presented. Recent advances in deep learning (DL), especially convolutional neural networks (CNNs) and fully convolutional networks (FCNs), have shown outstanding performances in this task. However, the map accuracy depends on the quantity and quality of ground truth (GT) used to train them. At the same time, probabilistic graphical models (PGMs) have sparked even more interest in the past few years, because of the ever-growing need for structured predictions. The novel method proposed in this paper combines DL and PGMs to perform remote sensing image classification. FCNs can be exploited to deal with multiscale data through the integration with a hierarchical Markov model. The marginal posterior mode (MPM) criterion for inference is used in the proposed framework. Experimental validation is conducted on the ISPRS 2D Semantic Labeling Challenge Vaihingen dataset. The results are significant, as the proposed method has a higher recall than the standard FCNs considered and allows mitigating the impact of incomplete or suboptimal GT, especially with regard to the discrimination of minoritary classes.