E. V. Dmitriev

Parameterization of volume scattering function of coastal waters based on the statistical approach.

A parameterization of the volume scattering function (VSF) specific to coastal waters is proposed. We have found that the standard VSF parameterizations proposed by Fournier-Forand and Petzold do not fit our measurements obtained with a high angular resolution VSF-meter for water samples taken in the Black Sea coastal zone. We propose modeling VSF as a linear function of scattering, backscattering and particulate absorption. The statistical techniques employed allow us to retrieve the variability of VSF and to demonstrate the significance of the estimates obtained. The results of independent validation and the comparison with other commonly used parameterizations are provided.

A system for processing hyperspectral imagery: application to detecting forest species

Although there are many approaches to hyperspectral data processing, they are typically based on an intuitive search of the appropriate spectral channels to solve the pattern recognition problem. To account for the accuracy of the computational procedures used, optimization techniques are needed to select the most useful spectral channels and to find contextual links for neighbouring pixels within a particular class of observed objects. We describe a system that merges both these types of mathematical formalism using a step-up method to extract the optimal channels from their entire set and to explain the contextual constraints on the images of high spatial resolution. The method is applied to forests of different species and age, which include areas illuminated by the Sun and shaded areas; these are the main classes recognized in this study. The proposed improvements in finding the specific information layers serve to enhance the computational efficiency of the techniques applied. These layers are formed by the sunlit forest canopy, sunlit background, and shaded background for a particular solar zenith angle during an aerial survey. The original system is created based on the relevant construction of the classifier employed, bearing in mind the signal to noise ratio of the hyperspectral device, its calibration, and the elaborated procedures of imagery processing. Results are shown of the related applications using the proposed system, which reveal the higher diversity in mapping forest classes due to the separation of the pixels in accordance with the indicated information layers. The accuracy of the pattern recognition for the processed scenes is shown to increase as the listed procedures are realized.

Improved technique for retrieval of forest parameters from hyperspectral remote sensing data.

This paper describes an approach of machine-learning pattern recognition procedures for the land surface objects using their spectral and textural features on remotely sensed hyperspectral images together with the biological parameters retrieval for the recognized classes of forests. Modified Bayesian classifier is used to improve the related procedures in spatial and spectral domains. Direct and inverse problems of atmospheric optics are solved based on modeling results of the projective cover and density of the forest canopy for the selected classes of forests of different species and ages. Applying the proposed techniques to process images of high spectral and spatial resolution, we have detected object classes including forests within their contours on a particular image and can retrieve the phytomass amount of leaves/needles as well as the relevant total biomass amount for the forest canopy.

Testing different classification methods in airborne hyperspectral imagery processing

To enhance the efficiency of machine-learning algorithms of optical remote sensing imagery processing, optimization techniques are evolved of the land surface objects pattern recognition. Different methods of supervised classification are considered for these purposes, including the metrical classifier operating with Euclidean distance between any points of the multi-dimensional feature space given by registered spectra, the K-nearest neighbors classifier based on a majority vote for neighboring pixels of the recognized objects, the Bayesian classifier of statistical decision making, the Support Vector Machine classifier dealing with stable solutions of the mini-max optimization problem and their different modifications. We describe the related techniques applied for selected test regions to compare the listed classifiers.

Cognitive technologies for processing optical images of high spatial and spectral resolution

The main stages of the development of technologies for natural and anthropogenic object recognition (cognitive technologies for optical image processing) using remote sensing data are considered together with computational procedures for atmospheric correction of multispectral and hyperspectral air-space images. The main focus is on recognizing forest ecosystems of various species and age, based on inflight testing of domestic hyperspectral equipment for a selected test area, where ground-based forest inventory and other observations were carried out. High accuracies of the recognition of separate gradations of ages for the selected pure birch and pine stands are revealed using elaborated software for airborne hyperspectral image processing.

System for processing of airborne images of forest ecosystems using high spectral and spatial resolution data

The developed hardware and software system for the recognition of natural and man-made objects based on the airborne hyperspectral sensing implements flight tasks on selected survey routes and computational procedures for solving applied problems that occur in data processing. The basics of object recognition based on obtained images of high spectral and spatial resolution in mathematical terms of sets of sites and labels and the basics of interrelations between separate resolution elements (pixels) for selected object classes are presented. Features of energy minimization of the processed scene are depicted as a target function of the optimization of computation and regularization of the solution of the considered problems as a theoretical basis for distinguishing between classes of objects in the presence of boundaries between them. Examples of the formation of information layers of recorded spectra for selected “pure species” of pine and birch forests are cited, with the separation of illuminated and shaded pixels, which increases the accuracy of object recognition in the processing of the images.

Recognition of natural and man-made objects in airborne hyperspectral images

New approaches to the processing of airborne hyperspectral images are implemented in order to develop emerging applications based on high-performance computing resources. The focus is on solving the problem of recognizing forest vegetation of different species composition and age based on high spectral and spatial resolution airborne sensing data. Examples of the formation of information layers of recorded spectra for “pure species” of pine and birch forests are given with the selection of illuminated and shaded pixels, which increases the accuracy of recognition of objects in the processing of these images.

Simulation of Local Atmospheric Dynamics in the Coastal Region of Dunkerque

The structure of the lower troposphere has been studied during the sea-breeze and post sea-breeze events in an industrialized coastal area of the North Sea. Atmospheric dynamics and dispersion of pollutants in the lower troposphere have been analyzed by the experimental results of the 3D nonhydrostatic Meso-NH model in Dunkerque area (51°N, 2.20°E), in the north of France. The simulations were verified and extended by data of the measurement campaign. Ground-based remote sensing systems (lidar and sodar), surface meteorology and air quality network stations data have been employed. We illustrate the different pollution scenarios and breeze structure by the analysis of Lagrangian tracers and back trajectories.

Comparison Analysis of Recognition Algorithms of Forest-Cover Objects on Hyperspectral Air-Borne and Space-Borne Images

The basic model for the recognition of natural and anthropogenic objects using their spectral and textural features is described in the problem of hyperspectral air-borne and space-borne imagery processing. The model is based on improvements of the Bayesian classifier that is a computational procedure of statistical decision making in machine-learning methods of pattern recognition. The principal component method is implemented to decompose the hyperspectral measurements on the basis of empirical orthogonal functions. Application examples are shown of various modifications of the Bayesian classifier and Support Vector Machine method. Examples are provided of comparing these classifiers and a metrical classifier that operates on finding the minimal Euclidean distance between different points and sets in the multidimensional feature space. A comparison is also carried out with the “K-weighted neighbors” method that is close to the nonparametric Bayesian classifier.

Statistical model for downscaling extreme surface temperatures at the weather station network in the Moscow region

Investigation of predictability of extreme meteorological values is an urgent problem of the present time. The purpose of this work is to demonstrate possibilities of reconstructing daily maximum and minimum air temperatures on a city scale using short-range weather forecasts. A statistical model is suggested, with which more than 85% of the natural variability of the extreme temperature at the Moscow weather stations can be reconstructed. A possibility to predict the maximum outliers in the solutions is demonstrated. The necessity to use the procedures of filling up the available gaps in observational data is emphasized. A classification of extreme situations in the atmosphere is suggested, which will help to increase the accuracy of the solution.