The recent advancements in hyperspectral satellite missions have significantly expanded Earth Observation capabilities. Availability of high-resolution hyperspectral data and products thereof, such as those delivered by the PRISMA mission of the Italian Space Agency (ASI), provides unprecedented opportunities for monitoring Earth processes in a wide range of applications. Enhancement of hyperspectral image products plays a crucial role in further increasing data exploitation possibilities. In this work, we propose a unified, physics-based APproAch for joint detection of Clouds, water, and sHadows in hypErspectral images (APACHE), focusing on PRISMA L1 product enhancement. Unlike existing approaches that typically address detection of one individual surface type at a time, our methodology jointly extracts the three surfaces of interest in an automated, stepwise, and unsupervised manner. More importantly, there is no need for parameter tuning, labeled data, or multi-source images. APACHE only leverages spectral properties of clouds, water, and shadows combined with a non-parametric clustering technique. Experimental results on a variety of PRISMA images from diverse datasets demonstrate the effectiveness of the proposed methodology in accurately detecting the surfaces of interest, both with respect to state-of-the-art methods and enhancing PRISMA products. On average, APACHE achieves accuracies of 96.42% for cloud detection, 99.23% for water detection, and 97.55% for shadow detection, with improvements ranging from 2% to 30% over the best performing comparison method. These results highlight APACHE potential to improve hyperspectral satellite data exploitation across a wide range of remote sensing applications.

This paper provides an overview of the main activities and results of HYPERHEALTH project (funded by the Italian Space Agency). Specific focus of this paper is hyperspectral PRISMA data exploitation, mostly as regards PRISMA-based atmospheric constituent estimation and allergenic vegetation monitoring.

The recent development of new hyperspectral satellite missions provides newer and wider Earth Observation opportunities. Exploitation and integration of high-resolution hyper- spectral data, such as those provided by PRISMA sensor of the Italian Space Agency (ASI), have been changing the way we sense and monitor the environment. The great availability of hyperspectral data and products thereof allows for a variety of applications.Enhancement of hyperspectral image products plays a crucial role in increasing data exploitation possibilities. In this work, we describe methodologies to refine PRISMA data products and present some results on PRISMA Level 1 and Level 2 images.

HYPERHEALTH project is co-funded by Italian Space Agency (ASI) in the framework of the "PRISMA Scienza" program. The program supports R&D projects proposed by experts in hyperspectral remote sensing sector from national public research institutions to industries, also in the framework of international partnerships. The aim is designing, developing and testing innovative methods, techniques and algorithms for exploitation of hyperspectral data, with reliable perspectives as to engineering and pre-operational development, thus contributing to the improvement of socio-economic benefits of the end-user community. This paper outlines HYPERHEALTH main goals and activities.

A closed-form non-parametric detector for rare targets in hyperspectral images is developed, extending the so-called Veritas approach, which until now had been explored only for parametric background models. By treating the typically unknown target fill-factor as known and equal to the smallest value that can be detected at a given level of detectability, an admissible NP detector is obtained. Experimental results on a hyperspectral target detection scenario reveals the potential of the proposed approach, and sets the path for developing more complex but still mathematically and computationally tractable non-parametric admissible detectors based on multiple discrete target fill-factors.

The unique spectral content provided by PRISMA's hyperspectral sensor gives the possibility to study the Earth's surface and environment from space in unprecedented detail. In this respect, our work presents the results of an Italian Space Agency-funded project aiming to develop eight prototypes for providing Value Added products based on such data. Prototypes focus on vegetation, urban areas, water quality, material detection, and natural hazards, combining multiple state-of-the-art techniques based on Machine Learning, physical models, and index-based algorithms. This is particularly relevant given the increasing demand for accurate information to address sustainable policies and support decision-making processes. Through a series of case studies, we highlight the versatility and utility of PRISMA's hyperspectral data for various scientific and operational applications.

We investigate the use of Bayesian methods for hyperspectral subpixel target detection, where the uncertainty associated with the target fill factor is “probabilized” by a suitable prior. Specifically, we present a general framework for Bayesian target detection by employing different models for the background distribution, comparing different choices for the Bayesian prior, and investigating different numerical schemes for evaluating the Bayesian integral. The Bayesian methods are furthermore compared to their Generalized Likelihood Ratio Test (GLRT)-based counterparts. Experiments performed over real hyperspectral imagery, with both real and implanted subpixel targets, show that incorporating prior knowledge by means of non-uniform priors emphasizing smaller target fill factors outperforms usage of the “noninformative” uniform prior and enhances Bayes performance beyond the GLRT, a result observed for both parametric and non-parametric background models. We find that even “rough” priors can successfully leverage the context-based information by emphasizing target sizes that are of most interest. We further observe that the Gauss-Legendre numerical integration scheme provides efficient integral approximation while maintaining the desirable admissibility property of Bayesian methods.

A Bayesian Likelihood Ratio Test (LRT) detector is analytically derived here for the replacement target model and using the non-parametric variable-bandwidth kernel density estimator to model the hyperspectral background. The detector is compared to the recent Generalized LRT detector, based on the same non-parametric model for the background. Experimental results obtained on two hyperspectral sub-pixel target detection scenarios reveal the great potential of the proposed detector and set the basis for future investigations.

This work deals with submerged object recognition methodologies with fluorescence LIDAR that can be applied when no prior information about environmental conditions is available. Previous invariant methods rely upon conventional unconstrained and constrained subspace projection concepts. This paper investigates application of sparsity-based concepts within this framework. A method enforcing both L1and L2norm penalties is investigated. Both synthetic and real data are employed to evaluate the potential of the method. Experimental results reveal that sparsity-based methods can be useful in this context and deserve further investigation.

We implement and evaluate a Bayesian detector for opaque subpixel hyperspectral targets of unknown abundance. Using both simulated and real hyperspectral backgrounds, we compare this detector to the more conventional generalized likelihood ratio test (GLRT) approach, identifying theoretical differences and observing numerical similarities. Among the theoretical advantages provided by the Bayesian detector is admissibility, which means that no detector can be uniformly superior to it. Potential disadvantages include the need to choose a prior distribution, and the computation required to integrate that distribution. For solid subpixel targets, the uniform prior is a natural choice, and we find that adequately-accurate numerical integration can be achieved with only a few evaluations of the likelihood function. We show results for targets implanted in both simulated and real data.

Water quality assessment plays an important role in sustainable development of natural resources. Fluorescence LIDAR is among the preferred remote sensors used in water quality monitoring. This work presents an underwater fluorescence LIDAR simulator for water quality assessment developed to design, test, and validate methodologies for the retrieval of key biophysical parameters or to discriminate dissolved substances and pollutants. Experimental results illustrating the usefulness of the simulator are provided focusing, as a case-study, on a key water quality indicator playing a major role in the global carbon balance, namely the Chromophoric/colored Dissolved Organic Matter (CDOM).

This work examines classical, more recent, and new hyperspectral detection algorithms that stem from the common framework of the decision-theory based statistical likelihood ratio test (LRT). Within this context, the tradeoffs involve improving models of target spectral variability, accurately characterizing the background, and producing a detector with closed-form solution. There is no algorithm that has shown universally best performance, but each of the algorithms can be specifically suited to deal with a given target detection scenario. Experimental results featuring real hyperspectral data are shown to compare the detection performance of the examined algorithms on two case-study target detection scenarios.

This article presents an automated underwater material recognition methodology for fluorescence light detection and ranging (LIDAR) invariant to environmental conditions (ARTEMIdE). Contrary to other state-of-the-art methods for submerged object recognition, ARTEMIdE can be applied when no a priori knowledge about environmental conditions is available and without resorting to any additional data besides the received signal and the fluorescence spectral signatures of the materials of interest. Experimental results over synthetic and real data show that ARTEMIdE is effective at automatically recognizing various object materials submerged at different depths within the water column. The presented approach reveals to provide great potential for many marine and submarine applications.

This work presents a submerged object recognition method with fluorescence LIDAR that can be applied when no a priori information about environmental conditions is available. Whereas conventional methods require the availability of either LIDAR measurements of water samples or accurate knowledge about environmental conditions, the approach investigated here remove such assumptions. Experimental results on real data acquired in laboratory show the potential of the approach.

This work presents a novel target detector that combines a nonparametric approach for conditional probability density function (pdf) estimation and an adaptive estimation of the target strength of the additive model it is based on. The variable bandwidth kernel density estimator is employed for pdf estimation within the Generalized Likelihood Ratio Test (GLRT) framework and a closed-form solution is found. Experimental results featuring hyperspectral data of a real subpixel target detection scenario reveal the potential of the proposed approach.

In this paper we propose novel semi- and non- parametric detectors to be used with the additive target signal model within the general detection framework of the likelihood ratio test. In the semi-parametric detector, the Gaussian mixture model is employed to estimate a lower dimensional approximation of the background probability density function (PDF), whereas a multivariate kernel density estimator is employed to estimate the PDF in the multidimensional space within the non-parametric approach. Target detection experiments are carried out using the hyperspectral airborne “Viareggio 2013 trial” data set. The detectors are shown to provide promising results for the detection of the targets of interest deployed in the scene and outperform the well-known Adaptive Matched Filter detector.

Discriminating different underwater objects based on their material is of great interest in many maritime applications, such as safe navigation and seafloor mapping. Fluorescence Light Detection And Ranging (LIDAR) systems allow different materials to be discriminated based on their spectral fluorescence properties. However, acquisition conditions play an important role in underwater material discriminability, and a scarce and/or inaccurate knowledge of such conditions may impair final performance. In this work, we investigate underwater material discriminability in unknown environmental conditions. Experimental results obtained with synthetic data reveal the potential of the explored approach.

In this work a novel target detector is proposed that is nonparametric in terms of conditional probability density function (pdf) estimation and parametric with respect to the target strength of the additive model it relies upon. The variable bandwidth kernel density estimator is employed to estimate the conditional pdfs, whereas the target strength is estimated via the Maximum Likelihood approach. Experimental results over real hyperspectral data show that the detector succeeds in detecting target objects embedded in a complex background and in providing reasonable estimates for the target strengths.

Target detection experiments with a novel non-parametric detector are carried out exploiting the availability of a new hyperspectral data set featuring a suburban scene with several different targets. Benefiting from its non-parametric nature and from its data adaptivity deriving from the variable-bandwidth approach, the detector is shown to provide promising results for the detection of the targets of interest both in global and local configurations.

Detecting and recognizing underwater objects is a topic of great interest in many maritime applications, such as harbor security, safe navigation of autonomous underwater vehicles, and safety of the littoral zone [1]. Fluorescence Light Detection And Ranging (LIDAR) systems play an important role in this context. In this work, a framework for predicting the performance of underwater object recognition by means of fluorescence LIDAR is proposed with the aim of assisting the user during operations such as mission planning and LIDAR system design. Experimental results obtained within a Monte Carlo simulation framework reveal the potential of the proposed framework.

An experimental study is presented with the goal of investigating application of the forward modeling (FM) approach to target detection to hyperspectral images acquired over an urban scenario, where the effects of variability and uncertainty of atmospheric and illumination conditions are enhanced. The study exploits the availability of a new hyperspectral data set acquired in May 2013 over the city of Viareggio, Italy. A physics-based model is assumed for radiation transfer in the atmosphere and variability around nominal acquisition conditions is accounted for by means of multiple executions of a radiative transfer code and a subspace-based target detection framework. Experimental results featuring several ground-truthed targets in an urban scenario confirm the effectiveness of the FM target detection approach.

This paper presents a multi-temporal approach to compensate the atmospheric effects in spaceborne hyperspectral images. It focuses on applications where a hyperspectral sensor periodically acquires images of the same scene and such images have to be classified. The method assumes that a reference reflectance image of the region of interest, equipped with an accurate classification map, is available. Such an image and the corresponding classification map represent the database that will be a reference for the classification of temporally subsequent images over the same scene. The multi-temporal approach allows the classification to be performed on at sensor radiance data and it does not require the knowledge of the atmospheric conditions at the acquisition time.

This paper presents an end-to-end fluorescence LIDAR system modeling approach for predicting the performance of underwater object recognition. The overall LIDAR system chain is modeled, starting from LIDAR pulse generation up to collection and processing of the inelastic backscattering signal returned by the underwater object. A comprehensive simulator to reproduce in-water LIDAR propagation phenomena is included in the model. The proposed approach is capable of predicting underwater object recognition performance under various different fluorescence LIDAR system characteristics as well as viewing geometries and environmental conditions. Experimental results featuring immersion of an object in Case 1 and Case 2 waters show the potential of the proposed modeling approach.

The SAP4PRISMA project research activities aimed at supporting the Italian hyperspectral PRISMA mission by developing preliminary processing chains suitable for PRISMA to obtain high level hyperspectral data products for agriculture, land degradation, natural and human hazards.

This paper focuses on automated recognition of underwater objects by means of light detection and ranging (LIDAR) systems. Differently from most works involved in underwater object recognition with LIDAR, where objects are recognized by their shape, here the interest is distinguishing objects on the basis of physical/chemical properties of object materials. To this aim, laser-induced fluorescence (LIF) spectroscopy is exploited, and an ad hoc signal processing chain is presented to effectively analyze the LIF spectra extracted at the detected object-range. Specifically, the goal is that of automatically recognizing the detected object with respect to a database (DB) of objects of interest, which have been previously spectrally characterized by means of laboratory fluorescence measurements. To this aim, suitable physics-based methodologies are proposed to compensate the signal for water-column effects. A decision-theory-based framework is developed to approach spectral recognition of the detected object with respect to the object DB. Experimental results from a laboratory test-bed show that the proposed processing chain is effective at automatically recognizing objects submerged in an artificial water column at different depths, based on a diverse DB of sample materials. The presented approach is shown to provide great potential for automated object recognition in marine and other water environments.

A FluoRescence lidar simulator for underwater Object DeTEction applications (AFRODiTE) is proposed to generate inelastic backscattering signals returned from a water column both in the presence and in the absence of an underwater object. The simulator models the interaction of the transmitted laser pulse with the water medium, an underwater object, and the bottom. Specifically, AFRODiTE enables simulation of fluorescence backscattering signals for a variety of light detection and ranging (lidar) system characteristics, acquisition geometries, and water environmental conditions. With respect to models for the elastic backscattering lidar signals developed in the literature, AFRODiTE may be used to test and improve not only underwater object detection methodologies based on time-resolved lidar waveform analysis but also object recognition methodologies based on spectral analysis of the object fluorescence spectral signature. Experimental comparison with real signals measured by an advanced prototypal fluorescence lidar in a laboratory artificial water column shows that AFRODiTE is effective at reproducing the inelastic backscattering signals received by a lidar system from an underwater object. Furthermore, simulations of the received signals obtained reproducing immersion of objects in the waters of the Gulf of Mexico and the North Atlantic Ocean highlight AFRODiTE potential and flexibility for generating fluorescence lidar signals acquired in different operational scenarios on the basis of various system parameters, acquisition geometries, and water environments.

This letter presents a scheme for detecting global anomalies, in which a likelihood ratio test based decision rule is applied in conjunction with an automated data-driven estimation of the background probability density function (PDF). The latter is reliably estimated with a nonparametric variable-band width kernel density estimator (VKDE), without making any distributional assumption. With respect to conventional fixed bandwidth KDE (FKDE), which lacks adaptivity due to the use of a bandwidth that is fixed across the entire feature space, VKDE lets the bandwidths adaptively vary pixel by pixel, tailoring the amount of smoothing to the local data density. Two multispectral images are employed to explore the potential of VKDE background PDF estimation for detecting anomalies in a scene with respect to conventional nonadaptive FKDE.

We propose a local anomaly detection strategy for multi-hyperspectral images in which the background probability density function is estimated with a kernel density estimator and locally adaptive information extracted from the image is injected into the bandwidth selection process. Results for multispectral images of different scenarios show the benefits of the proposed strategy regarding its effectiveness both at detecting anomalies and at avoiding the crucial issue of properly selecting the kernel-width parameter.

The effects of signal contamination of secondary data are investigated in the framework of adaptive target detection in remotely sensed hyperspectral images. In contrast to previous studies on signal contamination, the focus of this paper is the detection of targets with unknown spectral signatures (i.e., anomalies) and adaptive detection methods based on a local estimation of the background covariance matrix. Contamination due to the target signal is expected to have a more severe impact when the number of secondary data is limited. An analytical model for signal contamination is developed that allows variability in the extent of contamination. Several parameters, such as the contamination fraction of secondary data and the contaminating signal energy, are introduced, and a contaminating signal-to-interference-plus-noise ratio is derived as an objective measure of contamination. The proposed model is employed to experimentally evaluate signal contamination effects and the impact of its variability on the performance of adaptive detection of local anomalies. The outcomes of the experimental study are substantiated by validation with real hyperspectral data. The results obtained highlight the relevance that the impact of signal contamination, assessed with respect to different system parameters, may have for practical applications. This paper represents a starting point for the development of detection performance forecasting models that consider signal contamination.

The main objectives of the PRISMA (Hyperspectral Precursor of the Application Mission) mission are: the implementation of an Earth Observation pre-operative payload, the in-orbit demonstration and qualification of an Italian state-of-the-art hyperspectral/panchromatic technology and the validation of end-to-end data processing system able to support the development of new applications based on high spectral resolution images. The aim of the paper is to provide an overview of the PRISMA mission by describing the current status of the program and giving a brief outline of the work done till now in the framework of the SAP4PRISMA project scientific studies in supporting the exploitation of the future PRISMA hyperspectral images for environmental applications.

Anomaly detection (AD) in remotely sensed hyperspectral images has been proven to be valuable in many applications. In this paper, we propose a scheme for detecting global anomalies in which a likelihood ratio test-based decision rule is applied in conjunction with automated data-driven estimation of the background probability density function (PDF). Specifically, the use of both semiparametric (finite mixtures) and nonparametric (Parzen windows) models is investigated for background PDF estimation. Although such approaches are well known in multivariate data analysis, they have been very seldom applied to estimate the hyperspectral image background PDF, mostly due to the difficulty of reliably learning the model parameters without operator intervention. In this paper, semi and nonparametric estimators have been successfully employed to estimate the image background PDF with the aim of detecting global anomalies in a scene benefiting from the application of ad hoc Bayesian learning strategies. Two real hyperspectral images have been used to experimentally evaluate the ability of the proposed AD scheme resulting from the application of different global background PDF models and learning methods.

In this paper we investigate the effects of the signal dependent noise on the CFAR property of the RX anomaly detector. The CFAR behaviour of the algorithm was proved under the assumption of spatial stationarity of the random noise affecting the hyperspectral image. In data collected by new generation sensors such an assumption is not valid because photon noise contribution, which depends on the spatially varying signal levels, is not negligible. In this paper, experiments on real data collected by a new hyperspectral camera are discussed in order to show that the signal dependent noise is one of the causes of the non-CFAR behaviour of the RX detector we have experienced in many practical situations.

This work investigates the effects of target signal contamination over local RX anomaly detection performance. An analytical model for signal contamination is introduced where several parameters, such as the contamination fraction and the contaminating signal strength, are employed to derive an objective measure of contamination. Results from Monte Carlo simulations show the relevance that signal contamination may have in practical applications.

The SAP4PRISMA is a four year research project which aims at developing algorithms and products for the future PRISMA mission. The project started on May 2010 and is now entering his full activities as the ”PRISMA like” data set has been defined and the test areas were selected. The paper describes the main project objectives and the activities realized in the first 9 months of the project.

Local background covariance matrix corruption due to outliers in the sample data may be one of the major causes that limit detection performance of those algorithms that detect local anomalies in hyperspectral images on the basis of the Mahalanobis distance. In this letter, an original detection scheme is presented that efficiently embeds covariance corruption mitigation. A kurtosis-based binary hypothesis test is first applied to each pixel to quickly determine the presence of outliers in the local neighborhood. Rejection of the null hypothesis triggers application of a robust-to-outlier covariance estimation technique. Results on real data exhibit good detection performance and robustness to outliers. Contrary to previous works, this is achieved without an unnecessary increase of the procedural complexity.

Over the past few years, hyperspectral data exploitation aimed at detecting spectral anomalies within remotely sensed images has been of growing interest in many applications. In this letter, we are interested in an anomaly detection (AD) scheme for hyperspectral images in which spectral anomalies are defined with respect to a statistical model of the background probability density function (PDF). Among the multitude of PDF estimators discussed in statistics literature, Parzen windowing (PW) has always attracted much attention. However, its ability to estimate the PDF of global background in order to detect anomalies in hyperspectral images has not been investigated yet. Here, we propose the use of PW to provide reliable background PDF estimation. As is widely recognized, PW performance is primarily affected by the choice of the bandwidth matrix, which controls the degree of smoothing for the resulting PDF approximation. In this letter, the bandwidth selection problem is approached, resorting to an unsupervised method based on a Bayesian approach. Once the background PDF is approximated through PW, it is employed to detect anomalous objects within the scene by using the likelihood ratio test.

This paper deals with subspace-based target detection in hyperspectral images. Specifically, it focuses on a general detection scheme where, first, background is suppressed through orthogonal-subspace projection and then target detection is accomplished. An adequate estimation of the background subspace is essential to a successful outcome. The background subspace has been typically estimated globally. However, global approaches may be ineffective for small-target-detection applications since they tend to overestimate the background interference affecting a given target. This may result in a low target residual energy after background suppression that is detrimental to detection performance. In this paper, we propose a novel and fully automatic algorithm for local background-subspace estimation (LBSE). Local background has typically a lower inherent complexity than that of global background. By estimating the background subspace over a local neighborhood of the test pixel, the resulting background-subspace dimension is expected to be low, thus resulting in a higher target residual energy after suppression which benefits the detection performance. Specifically, the proposed LBSE acts on a per-pixel basis, thus adaptively tailoring the estimated basis to the local complexity of background. Both simulated and real hyperspectral data are employed to investigate the detection-performance improvements offered by LBSE with respect to both global and local methodologies previously presented.

Accounting for radiative transfer within the atmosphere is usually necessary to accomplish target detection in airborne/satellite hyperspectral images. In this paper, two methods of accounting for the illumination and atmospheric effects-atmospheric compensation (AC) and forward modeling (FM)-are investigated in their application to target detection. Specifically, several crucial aspects are examined, such as the processing required, the computational complexity, and the flexibility accorded to an imperfect knowledge of acquisition conditions. Real ground-truthed hyperspectral data are employed in order to evaluate the operational applicability of such approaches in a target-detection scenario, as well as their impact on the processing-chain computational complexity. Results indicate that AC is recommended when accurate knowledge of the acquisition conditions is available, and the image has relatively uniform illumination and nonshadowed targets. Conversely, FM is preferred if scene conditions are not well known and when the targets may be subject to varying illumination conditions, including shadowing.

This paper proposes a fully unsupervised anomaly detection strategy in hyperspectral imagery based on mixture learning. Anomaly detection is conducted by adopting a Gaussian mixture model (GMM) to describe the statistics of the background in hyperspectral data. One of the key tasks in the application of mixture models is the specification in advance of the number of GMM components, the determination of which is essential and strongly affects detection performance. In this work, GMM parameters estimation was performed through a variation of the well-known expectation maximization (EM) algorithm that was developed within a Bayesian framework. Specifically, the adopted mixture learning technique incorporates a built-in mechanism for automatically assessing the number of components during the parameter estimation procedure. Then, generalized likelihood ratio test (GLRT) is considered for detecting anomalies. Real hyperspectral imagery acquired by an airborne sensor is used for experimental evaluation of the proposed anomaly detection strategy.

This paper deals with the problem of signal subspace estimation and dimensionality reduction (DR) in hyperspectral images. A new algorithm is presented which preserves both the abundant and the rare signal components and is therefore suitable for DR in target detection applications. Results obtained by applying the new procedure and a classical method based on the analysis of the second order statistics are presented and discussed with reference to real AVIRIS data.