The downstream space sector in Italy is confronted with distinctive challenges, marked by a fragmented landscape, inadequate stakeholder interactions, underdeveloped business planning, and a critical need for improved industrial and financial acumen. In response to these challenges, the Italian Space Agency has taken a central role in addressing these issues through the establishment of the Downstream and Integrated Application Unit. The primary objective of this unit is to elevate the competitiveness of national Small and Medium Enterprises (SMEs), industries, and academia operating in the space sector. Within this strategic framework, the PRISMA SCIENZA program emerges as a pivotal initiative, aiming to harness the scientific data generated by the PRISMA mission for the improvement of the Italian Community. Specifically, the program is designed to facilitate the complete exploitation of mission data while concurrently fostering the advancement of Italian expertise in the hyperspectral remote sensing sector.By operating within this comprehensive framework, the PRISMA SCIENZA program contributes to the overarching goal of addressing challenges within the Italian downstream space sector. It not only serves as a mechanism for maximizing the utilization of mission data but also plays a vital role in promoting the growth and proficiency of the Italian space industry, aligning with broader strategic objectives set forth by the Italian Space Agency.Aim of this paper is to analyze the effects the PRISMA SCIENZA program brought to the Italian industrial and scientific communities.

The scarcity of a sufficiently large and representative hyperspectral image dataset is a substantial obstacle to the effective development of algorithms for remote sensing applications. Hyperspectral images can provide rich spectral information for various tasks, such as land cover classification, vegetation monitoring, and environmental assessment. However, the limited availability of diverse and well-annotated hyperspectral datasets hinders the development and optimization of these models in this domain. For this purpose, the generation of synthetic hyperspectral images has emerged as a pivotal area of research.This paper aims to introduce a preliminary analysis of various AI-based methodologies specifically crafted to generate synthetic PRISMA hyperspectral images derived from Sentinel-2 data. By exploring innovative approaches, this study aims to develop novel techniques for creating synthetic datasets, providing valuable insights into the potential of synthetic hyperspectral imagery for algorithm training and evaluation in the absence of extensive real-world hyperspectral datasets.

The activities undertaken within the EcoNet project aim at the design and development of an integrated system for the monitoring of changes in surface waters natural status based on different sensoristic techniques. The proposed integration approach combines ground measurements and hyperspectral satellite images. The promising dialogue that occurs between these two multi-sensoristic technologies requires the implementation of appropriate tools for data handling and analysis which in this work are represented by Artificial Intelligence (AI), particularly suitable to retrieve very subtle relationships among the data. This integration can open enormous potential for overcoming the limits of traditional environmental monitoring and diagnostic techniques.

The PRIMARY (PRIsma for Monitoring AiR quality) project objective is to address air quality monitoring, especially in urban areas, exploiting the PRISMA (PRecursore IperSpettrale della Missione Applicativa) mission. Utilizing PRISMA's hyperspectral data, the project aims to gain insights into atmospheric aerosol content and composition, crucial for understanding environmental and health impacts, especially in urban areas. Overcoming spatial resolution limitations and the inverse problem's complexity in satellite-based characterization, PRISMA's decametric spatial resolution and artificial intelligence play crucial roles. A synthetic PRISMA-like dataset, relying on data provided by the Copernicus Atmosphere Monitoring service (CAMS), was generated for training neural networks for estimating aerosol characteristic exploiting PRISMA data. Preliminary results are encouraging. Properly field campaigns were performed in Rome (autumn 2022) and Milan (winter to summer 2023) to support the validation of the PRIMARY project's outcomes. In addition, drone-based campaigns are currently ongoing.

This paper introduces a hybrid approach to dimension reduction of PRISMA hyperspectral data, employing both linear and non-linear techniques: Principal Component Analysis (PCA) and autoencoders. The study aims to validate the efficacy of autoencoders by comparing results with the well-established PCA method. Our primary objective is to harness the complementary strengths of both methods in a hybrid framework, wherein certain bands may exhibit superior performance with autoencoders, while others fare better with PCA in dimension reduction. This strategic amalgamation not only accelerates data transfers and lowers computational costs for real-time applications but also leverages the specific advantages offered by each technique. The paper underscores the potential of this hybrid approach for optimizing hyper-spectral data for enhanced feature extraction in various neural network applications.

The main objective of the project SCIA (Sviluppo di algoritmi per lo studio della Criosfera mediante Immagini PrismA) is the development and optimization of methods for generating products related to the cryosphere. The project foresees the development of a robust processing chain of PRISMA hyperspectral data for the estimation of snow and glacier parameters in Alpine areas, through a combined use of satellite images, field data and radiative transfer models (RTMs). The image spectroscopy measurements provided by PRISMA will make possible to investigate radiometrically complex surfaces and obtain geophysical parameters currently only achievable through airborne hyperspectral sensors.

Space technologies and applications have the potential to significantly benefit European society and economy, addressing major societal challenges while bolstering synergies between public and private sectors. These technologies can stimulate economic growth and spawn new markets by fostering the development of innovative products and services tailored to meet user needs. However, the downstream space market poses unique challenges due to its disjointed nature, characterized by insufficient interactions between stakeholders, underdeveloped business planning, and the need for a stronger industrial and financial acumen. The Italian space industry, primarily comprised of SMEs (80%), encapsulates these issues, necessitating a robust ecosystem to streamline the space venture value chain.Italy has a strategic advantage, being one of the few nations with a comprehensive space industry production chain, spanning both upstream and downstream sectors. The Italian Space Agency is central to this development, having established the Downstream and Integrated Application Unit to enhance the competitiveness of national SMEs, industries, and academia. This initiative supports the development of space-based products and services, not just through financial aid, but by serving as a liaison between the space industry and traditional markets, fostering mutually beneficial interactions.

EcoNet is a joint project between the Italian Space Agency (ASI) and the Institute of Nanostructured Materials of the National Research Council of Italy (CNR-ISMN) with the participation of University of Tor Vergata (UTOV) that aims to develop an integrated sensor-driven system managed by artificial intelligence (AI) for monitoring surface waters near human settlements. The present paper outlines the foundational concepts of the project, the methodology and the ongoing activities, with an outlook towards the innovation for downstream applications in the field of aquatic ecosystem monitoring and management.

Air pollution and its effects on human health pose a significant challenge in modern society. The PRIMARY (PRIsma for Monitoring AiR quality) research project aims to utilize the capabilities of the Italian Space Agency's (ASI) PRISMA (PRecursor HyperSpectral Application Mission) to enhance air quality monitoring, particularly in urban areas. In particular, the project focuses on the exploitation of the hyperspectral PRISMA data to obtain detailed qualitative and quantitative data on atmospheric aerosol load and composition in urban environments. Current satellite-based characterization of particulate matter is limited due to spatial resolution constraints and to the complexities of the underlying inverse problem involving multiple variables. The PRIMARY project addresses the first issue through the decametric spatial resolution of PRISMA images, while the second issue is tackled by leveraging artificial intelligence approaches.

The Italian Space Agency (ASI) and the Indian Space Research Organisation (ISRO) established the joint Earth Observation Working Group (EOWG) that is currently focusing on hyperspectral (HYP) activity. Eleven projects are investigating the use of ASI’s PRISMA data for agriculture, land cover classification, mineral and soil mapping, Martian analogues, urban lakes, biodiversity and calibration. The paper provides an overview and first results after one year and half since the EOWG HYP was launched.

The intrinsic dimensionality (ID) of multivariate data is a very important concept in spectral unmixing of hyperspectral images. A good estimation of the ID is crucial for a correct retrieval of the number of endmembers (the spectral signatures of macroscopic materials) in the image, for dimensionality reduction or for subspace learning, among others. Recently, some approaches to perform spectral unmixing and superresolution locally have been proposed, which require a local estimation of the number of endmembers to use. However, the role of ID in local regions of hyperspectral images has not been properly addressed. Some important issues when dealing with small regions of hyperspectral data can seriously affect the performance of conventional hyperspectral ID estimators. We show that three factors mainly affect local ID estimation: the number of pixels in the local regions, which has to be high enough for the estimations to be relevant, the number of hyperspectral bands which complicates the estimations if the ambient space has a high dimensionality, and the noise, which can be misinterpreted as a signal when its power is important. Here, we review the hyperspectral ID estimators on the literature for local ID estimation, we show how they behave in a local setting on synthetic and real data sets, and we provide some guidelines to make proper use of these estimators in local approaches.

Remote sensing hyperspectral images (HSIs) are quite often low rank, in the sense that the data belong to a low dimensional subspace/manifold. This has been recently exploited for the fusion of low spatial resolution HSI with high spatial resolution multispectral images in order to obtain super-resolution HSI. Most approaches adopt an unmixing or a matrix factorization perspective. The derived methods have led to state-of-the-art results when the spectral information lies in a low-dimensional subspace/manifold. However, if the subspace/manifold dimensionality spanned by the complete data set is large, i.e., larger than the number of multispectral bands, the performance of these methods mainly decreases because the underlying sparse regression problem is severely ill-posed. In this paper, we propose a local approach to cope with this difficulty. Fundamentally, we exploit the fact that real world HSIs are locally low rank, that is, pixels acquired from a given spatial neighborhood span a very low-dimensional subspace/manifold, i.e., lower or equal than the number of multispectral bands. Thus, we propose to partition the image into patches and solve the data fusion problem independently for each patch. This way, in each patch the subspace/manifold dimensionality is low enough, such that the problem is not ill-posed anymore. We propose two alternative approaches to define the hyperspectral super-resolution through local dictionary learning using endmember induction algorithms. We also explore two alternatives to define the local regions, using sliding windows and binary partition trees. The effectiveness of the proposed approaches is illustrated with synthetic and semi real data.

In this letter, we propose a method aiming at reducing the noise in hyperspectral images based on the nonlinear generalization of principal component analysis (NLPCA). NLPCA is performed by an autoassociative neural network (AANN) that has the hyperspectral image as input and is trained to reconstruct the same image at the output. Due to its topology, characterized by a bottleneck layer, the nonlinear AANN forces the hyperspectral image to be projected in a lower dimensionality feature space by removing noise and both linear and nonlinear correlations between spectral bands. This process permits to obtain enhancements in terms of the quality of the reconstructed hyperspectral image. The results conducted on different hyperspectral images are qualitatively and quantitatively discussed and demonstrate the potentialities of the proposed method, as compared with similar approaches such as PCA and kernel PCA.

This letter presents a novel technique for the forecast of the ground horizontal irradiance (GHI) from satellite-based images. To enhance the forecast accuracy, spatial information in addition to temporal information has been considered. This produced an increase in the computational load of the forecast process. Dimensionality reduction techniques based on nonlinear principal component analysis (PCA) are used to project the original data set into low-dimension feature space. A multilayer feedforward neural network classifier is used to model the signal through a training operation involving past history of the considered spatiotemporal signal. Experiments have been carried out on two different data sets. Comparisons with classical forecasting techniques demonstrate that the introduction of the spatial information permits to obtain better short-term forecast measurements for all types of sky conditions. Moreover, further analysis demonstrates that, compared with linear PCA, the nonlinear PCA is more appropriate for dimensionality reduction of spatiotemporal GHI data set.

Pansharpening aims at fusing a multispectral and a panchromatic image, featuring the result of the processing with the spectral resolution of the former and the spatial resolution of the latter. In the last decades, many algorithms addressing this task have been presented in the literature. However, the lack of universally recognized evaluation criteria, available image data sets for benchmarking, and standardized implementations of the algorithms makes a thorough evaluation and comparison of the different pansharpening techniques difficult to achieve. In this paper, the authors attempt to fill this gap by providing a critical description and extensive comparisons of some of the main state-of-the-art pansharpening methods. In greater details, several pansharpening algorithms belonging to the component substitution or multiresolution analysis families are considered. Such techniques are evaluated through the two main protocols for the assessment of pansharpening results, i.e., based on the full- and reduced-resolution validations. Five data sets acquired by different satellites allow for a detailed comparison of the algorithms, characterization of their performances with respect to the different instruments, and consistency of the two validation procedures. In addition, the implementation of all the pansharpening techniques considered in this paper and the framework used for running the simulations, comprising the two validation procedures and the main assessment indexes, are collected in a MATLAB toolbox that is made available to the community.

Many powerful pansharpening approaches exploit the functional relation between the fusion of PANchromatic (PAN) and MultiSpectral (MS) images. To this purpose, the modulation transfer function of the MS sensor is typically used, being easily approximated as a Gaussian filter whose analytic expression is fully specified by the sensor gain at the Nyquist frequency. However, this characterization is often inadequate in practice. In this paper, we develop an algorithm for estimating the relation between PAN and MS images directly from the available data through an efficient optimization procedure. The effectiveness of the approach is validated both on a reduced scale data set generated by degrading images acquired by the IKONOS sensor and on full-scale data consisting of images collected by the QuickBird sensor. In the first case, the proposed method achieves performances very similar to that of the algorithm that relies upon the full knowledge of the degrading filter. In the second, it is shown to outperform several very credited state-of-the-art approaches for the extraction of the details used in the current literature.

Transform-based lossy compression has a huge potential for hyperspectral (HS) data reduction. The emerging JPEG2000 technology is based on the synergistic use of both spectral and spatial compression techniques. In this context the choice of the spectral decorrelation approach can have a strong impact on the quality of the compressed image. Since hyperspectral images are highly correlated within each spectral band and in particular across neighboring frequency bands, the choice of a spectral decorrelation method that allows to retain as much information content as possible is desirable. From this point of view, several methods based on PCA and Wavelet have been presented in the literature. In this paper, we propose the use of Nonlinear Principal Component Analysis (NLPCA) transform as a lossy spectral compression method applied to hyperspectral data. Being the NLPCA the nonlinear generalization of the standard principal component analysis (PCA), it permits to represent in a lower dimensional space the same information content with less features than the standard PCA.

In this paper, we propose a new method aiming at reducing the noise in hyperspectral images. It is based on the nonlinear generalization of Principal Component Analysis (NLPCA). The NLPCA is performed by an autoassociative neural network that have the hyperspectral image as input and is trained to reconstruct the same image at the output. Thanks to its bottleneck structure, the AANN forces the hyperspectral image to be projected in a lower dimensionality feature space where noise as well as both linear and nonlinear correlations between spectral bands are removed. This process permits to obtain enhancements in terms of hyperspectral image quality. Experiments are conducted on different real hyperspectral images, with different contexts and resolutions. The results are qualitatively and quantitatively discussed and demonstrate the interest of the proposed method as compared to traditional approaches.

Remote sensing hyperspectral images (HSI) are quite often locally low rank, in the sense that the spectral vectors acquired from a given spatial neighborhood belong to a low dimensional subspace/manifold. This has been recently exploited for the fusion of low spatial resolution HSI with high spatial resolution multispectral images (MSI) in order to obtain super-resolution HSI. Most approaches adopt an unmixing or a matrix factorization perspective. The derived methods have led to state-of-the-art results when the spectral information lies in a low dimensional subspace/manifold. However, if the subspace/manifold dimensionality spanned by the complete data set is large, the performance of these methods decrease mainly because the underlying sparse regression is severely ill-posed. In this paper, we propose a local approach to cope with this difficulty. Fundamentally, we exploit the fact that real world HSI are locally low rank, to partition the image into patches and solve the data fusion problem independently for each patch. This way, in each patch the subspace/manifold dimensionality is low enough to obtain useful super-resolution. We explore two alternatives to define the local regions, using sliding windows and binary partition trees. The effectiveness of the proposed approach is illustrated with synthetic and semi-real data.

The design of a pansharpening algorithm for enriching a MultiSpectral image with the spatial details of a Panchromatic image should preserve the characteristics of the original dataset. A widely employed quality check consists in verifying the consistency of the fused product, namely the similarity of the original image and a reduced resolution version of the sharpened product. We propose to improve this feature by applying an Iterative Back-Projection algorithm after the fusion procedure. The approach is validated on two datasets, acquired by the Ikonos and WorldView-2 sensors, showing remarkable improvements, especially in conjunction with Component Substitution pansharpening methods.

In this paper state-of-the-art and advanced methods for multispectral pansharpening are reviewed and evaluated on two very high resolution datasets acquired by IKONOS-2 (four bands) and WorldView-2 (eight bands). The experimental analysis allows us to highlight the performances of the two main pansharpening approaches (i.e. component substitution and multiresolution analysis).

Images with high spatial and spectral resolutions are desirable for remote sensing applications. Unfortunately, due to sensor physical constraints, this result cannot be obtained by a single sensor. To overcome these limitations, a great number of data fusion approaches have been developed in the last years. The fusion of panchromatic and multispectral images, also known as Pansharpening, is capturing a lot of attention in the literature. In this paper, we extend and analyze the use of some classical pansharpening techniques, belonging to the MultiResolution Analysis and Component Substitution families, for fusing hyperspectral data instead of multispectral ones. The experimental results, conducted on two real datasets acquired by the Hyperion/ALI and CHRIS-Proba/QuickBird sensors, point out the greater suitability of the algorithms into the MRA class thanks to a better spectral consistency of the final products, which is a desirable feature when the number of bands to fuse increases.

The pansharpening process has the purpose of building a high-resolution multispectral image by fusing low spatial resolution multispectral and high-resolution panchromatic observations. A very credited method to pursue this goal relies upon the injection of details extracted from the panchromatic image into an upsampled version of the low-resolution multispectral image. In this letter, we compare two different injection methodologies and motivate the superiority of contrast-based methods both by physical consideration and by numerical tests carried out on remotely sensed data acquired by IKONOS and Quickbird sensors.

The operational level reached by polarimetric data processing techniques has been demonstrated during the last decade. The next generation of spaceborne Synthetic Aperture Radar satellites will implement full- or dual- polarimetric capabilities. In few years a huge amount of data will have to be processed in a fast and reliable way, implementing polarimetric decompositions or accurate classifications. Two neural network approaches for fast and accurate processing of polarimetric data are presented. In the first approach a neural network based processing chain for fast model based polarimetric decomposition is developed, while in the second approach a Non-Linear Principal Component Analisys of polarimetric data has been performed using an Auto Associative Neural Network. The results show a considerable reduction of computational effort and a substantial data compression with a minimun loss of information.

Generally, for optical satellite sensors spatial and spectral resolutions are highly correlated factors. In fact, given the design constraints of these sensors, there is an inverse relation between their spatial and spectral resolution. Thus, the hyperspectral sensors have a high spectral resolution i.e. large number of bands covering the electromagnetic spectrum, but a lower spatial resolution. On the other hand, panchromatic (PAN) images have the highest spatial resolution but no spectral diversity. For better utilization and interpretation, hyperspectral images having both high spectral and spatial resolution are desired. This can be achieved by making use of a high spatial resolution PAN image in the context of pansharpening or image fusion. Several fusion approaches have been proposed in the literature. In this paper we propose the use of a hybrid algorithm combining substitution and injection methods. One of the main challenges in hyperspectral image fusion is the improvement of the spatial resolution, i.e. spatial details while preserving the original spectral information. This requires addition of pertinent spatial details to each band of the HS image. However, due to large number of bands the pansharpening of HS images is computationally expensive. Thus a dimensionality reduction preprocess, compressing the original number of measurements into a lower dimensional space, becomes mandatory. In this paper we propose the use of non-linear principal components instead of the original HS bands as input to a fusion process to enhance the spatial resolution of the HS image.

In the literature, for sake of simplicity it is usually assumed that the model ruling spectral mixture in a hyperspectral pixels is basically linear. However, in many real life cases the different materials are usually in intimate association, like sand grains, resulting in a nonlinear mixture. Unfortunately, modeling a nonlinear approach is not trivial, and a general procedure is still up to be found. Aim of this paper is to evaluate the potentialities of Nonlinear Principal Component Analysis (NLPCA) as an approach to perform a nonlinear unmixing for the unsupervised extraction and quantification of the end-members. From this point of view scope of this paper is to demonstrate that the NLPCs derived from the proposed process can be considered as end-members. To perform an accurate evaluation, the proposed algorithm has been tested on two different hyperspectral datasets and compared with other approaches found in the literature.

Aim of this paper is to show how fusing SAR images having different characteristics can improve the classification accuracy, in spite of the geometrical problems rising in the fusion operation. To this end, we fused multiple-frequency (Cand X-band), multiple-polarization (HH, HV, VH and VV) and multi-resolution images. The classification has been carried out by a neural network algorithm (NN), in which the backscattering coefficients at each polarization for each image have been fused to form the input to the classifier. The evaluation of the classification accuracy has been performed in terms of overall accuracy, per class accuracy and Kappa coefficient. The obtained results show not only an enhancement of the classification accuracies, but also that more land cover classes can be better identified with respect to a single acquisition.

In this paper we propose a new approach for the improvement of the spatial resolution of hyperspectral image classification maps combining both spectral unmixing and pansharpening approaches. The main idea is to use a spectral unmixing algorithm based on neural networks to retrieve the abundances of the endmembers present in the scene, and then use the spatial information retrieved from the pansharpened image to find the location of each endmember within the enhanced pixel according to the endmembers abundances. The proposed approach has been applied both to real and synthetic datasets.

The goal of this work is to investigate on the enhancement, in terms of accuracy and number of classes, of the vegetation mapping through the joint use of multi-sensors data. Several stacks of Spotlight COSMO-SkyMed, acquired both in HH and VV polarization, and Multispectral World-View2 images, taken in the same or different seasons, have been compounded and exploited to identify six types of natural surfaces by means of a Neural Network classifier. While the information content of the eight bands of the multispectral data may be sufficient to discriminate the classes of interest, the single polarization of each SAR image has to be integrated by extracting further features, such as textural parameters. The assessment of the provided vegetation maps has been carried out in terms of per class accuracy, overall accuracy and K coefficient. The achieved results demonstrate the improvement of the classifications obtained by fusing more information from multi-sensors acquisitions.

Morphological profiles (MPs) have been proposed in recent literature as aiding tools to achieve better results for classification of remotely sensed data. MPs are in general built using features containing most of the information content of the data, such as the components derived from principal component analysis (PCA). Recently, nonlinear PCA (NLPCA), performed by autoassociative neural network, has emerged as a good unsupervised technique to fit the information content of hyperspectral data into few components. The aim of this letter is to investigate the classification accuracies obtained using extended MPs built from the features of NPCA. A comparison of the two approaches has been validated on two different data sets having different spatial and spectral resolutions/coverages, over the same ground truth, and also using two different classification algorithms. The results show that NLPCA permits one to obtain better classification accuracies than using linear PCA.

This paper presents a novel method for the enhancement of spatial quality of Hyperspectral (HS) images while making use of a high resolution panchromatic (PAN) image. Due to the high number of bands the application of a pansharpening technique to HS images may result in an increase of the computational load and complexity. Thus a dimensionality reduction preprocess, compressing the original number of measurements into a lower dimensional space, becomes mandatory. To solve this problem we propose a pansharpening technique combining both dimensionality reduction and fusion, exploited by non-linear Principal Component Analysis (NLPCA) and Indusion respectively, to enhance the spatial resolution of a hyperspectral image.

Neural networks (NNs) are recognized as very effective techniques when facing complex retrieval tasks in remote sensing. In this paper, the potential of NNs has been applied in solving the unmixing problem in hyperspectral data. In its complete form, the processing scheme uses an NN architecture consisting of two stages: the first stage reduces the dimension of the input vector, while the second stage performs the mapping from the reduced input vector to the abundance percentages. The dimensionality reduction is performed by the so-called autoassociative NNs, which yield a nonlinear principal component analysis of the data. The evaluation of the whole performance is carried out for different sets of experimental data. The first one is provided by the Airborne Hyperspectral Scanner. The second set consists of images from the Compact High-Resolution Imaging Spectrometer on board the Project for On-Board Autonomy satellite, and it includes multiangle and multitemporal acquisitions. The third set is represented by Airborne Visible/InfraRed Imaging Spectrometer measurements. A quantitative performance analysis has been carried out in terms of effectiveness in the dimensionality reduction phase and in terms of the accuracy in the final estimation. The results obtained, when compared with those produced by appropriate benchmark techniques, show the advantages of this approach.

Feature extraction for the dimensionality reduction of hyperspectral data is performed by means of Auto-Associative Neural Networks. The algorithm performance is compared to the Principal Component Analysis and the Maximum Noise Fraction ones. Results of land cover pixel-based maps yielded by the reduced vector and a dedicated neural network classification algorithm are also reported.

In this paper we test an unsupervised neural network approach for extracting features from very high resolution X-band SAR images. The purpose of this study is buildings recognition in images of low density urban areas, acquired by COSMO-Skymed and TerraSAR-X satellites, by means of Pulse Coupled Neural Network (PCNN), a relatively novel unsupervised algorithm based on models of the visual cortex of small mammals. The features retrieved from geo-referenced SAR images are compared against the ground truth provided by corresponding optical images. The accuracy yielded by PCNN is quantitatively evaluated and critically discussed, also in comparison with commonly used feature extraction techniques.

In this paper Autoassociative Neural Networks (AANN) are used to implement Nonlinear Principal Component Analysis (NLPCA) for dimension reduction of hyperspectral data. The nonlinear components are then considered as inputs for a Multi-Layer Perceptron (MLP) network to perform pixel-based classification. The methodology has been applied considering the test area of Tor Vergata — Frascati, Italy, and the hyperspectral data provided by the CHRIS-PROBA mission. Comparative analysis with a similar procedure considering a more standard dimensionality reduction technique such as Principal Component Analysis (PCA) has been carried out.

The increasing number of satellite missions providing more and more data for updating land cover and land use maps requires to upgrade the level of automatism for the processing of remotely sensed imagery. In this paper we try to pursue the ambitious goal of designing a completely automatic (no human interaction) supervised scheme for the classification, in terms of land cover, of a multi-spectral image. An expert system, using appropriate spectral and textural features, drives the selection of suitable training pixels in the image. These are used for the learning of a neural network algorithm that successively performs the pixel-based land cover classification of the whole image. The processing scheme has been tested on a set of Landsat images taken on different European urban areas.

The 2008 Data Fusion Contest organized by the IEEE Geoscience and Remote Sensing Data Fusion Technical Committee deals with the classification of high-resolution hyperspectral data from an urban area. Unlike in the previous issues of the contest, the goal was not only to identify the best algorithm but also to provide a collaborative effort: The decision fusion of the best individual algorithms was aiming at further improving the classification performances, and the best algorithms were ranked according to their relative contribution to the decision fusion. This paper presents the five awarded algorithms and the conclusions of the contest, stressing the importance of decision fusion, dimension reduction, and supervised classification methods, such as neural networks and support vector machines.

Since a few months TerraSAR-X has been acquiring X-band SAR images of the earth surface from space. This contribution reports on a study carried out to understand the main textural features of the X-band radar return from various kinds of surfaces and in particular to assess the potential of images acquired by X-band space borne radars in mapping fire scars and in classifying suburban/agricultural land cover. To this end, a novel unsupervised neural network algorithm, the Textural Self-Organizing Map (TexSOM), based on the textural features of the radar image, has been worked out and tested on areas in Greece and Italy.

This paper reports the study of supervised neural network algorithm for classification purposes. SPOT 5 and TerraSAR-X dataset are analyzed. Classification results are critically discussed and compared to ground truth map and unsupervised neural classification of the same area. The aim is to demonstrate the capability of neural networks in managing heterogeneous dataset and the accuracy improvement obtained by the use of the textural object based layers fused with the optical and radar data.

The paper aims at improving image automatic classification from remote sensing data using neural network algorithms (NN). NN have been found to have good generalization properties and their use is becoming increasingly prevalent in the field of remote sensing and in particular for image classification. However, the type of input to be considered for the algorithm in order to maximize the information available from the measurement is still an open issue. Using the mere spectral signature with no pre-processing is not an effective choice. Another point regards the use of textural features to improve the classification which involves taking decisions on how many and what specific features should be considered. More in general, minimizing the number of inputs of a neural network algorithm, avoiding significant loss of information, affects positively the NN mapping ability and computational efficiency. In this paper we propose a new methodology facing with the aforementioned problems and providing a solution to them.

Polarimetric SAR interferometric data can provide estimates of forest biomass density. There are different approaches to deal with the inversion problem, such as neural networks and the traditional optimal estimation approach. This paper presents a study to evaluate their performance by means of quantitative indexes addressing both the computation time and the retrieval accuracy. Better forest parameters estimates have been obtained when neural networks algorithms were used.