ALBERT

Description: The use of hyperspectral data to determine the abundance of constituents in a certain portion of the Earth's surface relies on the capability of imaging spectrometers to provide a large amount of information at each pixel of a certain scene. Today, hyperspectral imaging sensors are capable of generating unprecedented volumes of radiometric data. The Airborne Visible/Infrared Imaging Spectrometer (AVIRIS), for example, routinely produces image cubes with 224 spectral bands. This undoubtedly opens a wide range of new possibilities, but the analysis of such a massive amount of information is not an easy task. In fact, most of the existing algorithms devoted to analyzing multispectral images are not applicable in the hyperspectral domain, because of the size and high dimensionality of the images. The application of neural networks to perform unsupervised classification of hyperspectral data has been tested by several authors and also by us in some previous work. We have also focused on analyzing the intrinsic capability of neural networks to parallelize the whole hyperspectral unmixing process. The results shown in this work indicate that neural network models are able to find clusters of closely related hyperspectral signatures, and thus can be used as a powerful tool to achieve the desired classification. The present work discusses the possibility of using a Self Organizing neural network to perform unsupervised classification of hyperspectral images. In sections 3 and 4, the topology of the proposed neural network and the training algorithm are respectively described. Section 5 provides the results we have obtained after applying the proposed methodology to real hyperspectral data, described in section 2. Different parameters in the learning stage have been modified in order to obtain a detailed description of their influence on the final results. Finally, in section 6 we provide the conclusions at which we have arrived.

Description: During the last several years, a number of airborne and satellite hyperspectral sensors have been developed or improved for remote sensing applications. Imaging spectrometry allows the detection of materials, objects and regions in a particular scene with a high degree of accuracy. Hyperspectral data typically consist of hundreds of thousands of spectra, so the analysis of this information is a key issue. Mathematical morphology theory is a widely used nonlinear technique for image analysis and pattern recognition. Although it is especially well suited to segment binary or grayscale images with irregular and complex shapes, its application in the classification/segmentation of multispectral or hyperspectral images has been quite rare. In this paper, we discuss a new completely automated methodology to find endmembers in the hyperspectral data cube using mathematical morphology. The extension of classic morphology to the hyperspectral domain allows us to integrate spectral and spatial information in the analysis process. In Section 3, some basic concepts about mathematical morphology and the technical details of our algorithm are provided. In Section 4, the accuracy of the proposed method is tested by its application to real hyperspectral data obtained from the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) imaging spectrometer. Some details about these data and reference results, obtained by well-known endmember extraction techniques, are provided in Section 2. Finally, in Section 5 we expose the main conclusions at which we have arrived.

Description: This report describes the transitional activities of the JPL Analysis Center.

Description: The 3.0-micrometers water of hydration absorption feature observed in the IR photometry of many low-albedo and some medium-albedo asteroids strongly correlates with the 0.7-micrometers Fe(+2) to Fe(+3) oxidized iron absorption feature observed in narrowband spectrophotometry of these asteroids. Using this relationship, an empirical algorithm for predicting the presence of water of hydration in the surface material of a Solar System body using photometry obtained through the Eight-Color Asteroid Survey nu (0.550 micrometers), w (0.701 micrometers), and x (0.853 micrometers) filters was developed and applied to the ECAS photometry of asteroids and outer planet satellites. The percentage of objects in low-albedo, outer main-belt asteroid classes that test positively for water of hydration increases from P to B to C to G class and correlates linearly with the increasing mean albedos of those objects testing positively. The medium-albedo M-class asteroids do not test positively in large number using this algorithm. Aqueously altered asteroids dominate the Solar System population between heliocentric distances of 2.6 to 3.5 AU, bracketing the Solar System region where the aqueous alteration mechanism operated most strongly. One jovian satellite, J VI Himalia, and one saturnian satellite. Phoebe, tested positively for water of hydration, supporting the hypothesis that these may be captured C-class asteroids from a postaccretional dispersion. The proposed testing technique could be applied to an Earth-based survey of asteroids or a space-probe study of an asteroid's surface characteristic in order to identify a potential water source.

Description: The infrared transmission spectra and photochemical behavior of various organic compounds isolated in solid N2 ices, appropriate for applications to Triton ad Pluto, are presented. It is shown that excess absorption in the surface spectra of Triton and Pluto, i.e., absorption not explained by present models incorporating molecules already identified on these bodies (N2, CH4, CO, and CO2), that starts near 4450/cm (2.25 microns) and extends to lower frequencies, may be due to alkanes (C(n)H(2n+2)) and related molecules frozen in the nitrogen. Branched and linear alkanes may be responsible. Experiments in which the photochemstry of N2: CH4 and N2: CH4: CO ices was explored demonsrtrate that the surface ices of Triton and Pluto may contain a wide variety of additional species containing H, C, O, and N. Of these, the reactive molecule diazomethane, CH2N2, is particularly important since it may be largely responsible for the synthesis of larger alkanes from CH4 and other small alkanes. Diazomethane would also be expected to drive chemical reactions involving organics in the surface ices of Triton and Pluto toward saturation, i.e., to reduce multiple CC bonds. The positions and intrinsic strengths (A values) of many of the infrared absorption bands of N2 matrix-isolated molecules of relevance to Triton and Pluto have also been determined. These can be used to aid in their search and to place constraints on their abundances.

Description: Estimates of the mass of dust suspended in the Martian atmosphere are derived from global and regional 9-micrometer opacity maps produced from Viking Infrared Thermal Mapper data. During the peak of the 1977b storm, a total dust mass of approximately 4.3 x 10(exp 14) g was suspended, equivalent to 4.3 x 10(exp -4) g/sq cm, or a layer 1.4 micrometers thick. During a local dust storm near Solis Planum at L(sub s) 227 deg, approximately 1.3 x 10(exp 13) g of dust were lofted, equal to about a 6-micrometer layer in that vicinity.

Description: We model an infrared outburst on Io as being due to a large, erupting lava flow which increased its area at a rate of 1.5 x 10(exp 5)/sq m and cooled from 1225 to 555 K over the 2.583-hr period of observation. The inferred effusion rate of 3 x 10(exp 5) cu m/sec for this eruption is very high, but is not unprece- dented on the Earth and is similar to the high eruption rates suggested for early lunar volcanism. Eruptions occur approxi- mately 6% of the time on Io. These eruptions provide ample resurfacing to explain Io's lack of impact craters. We suggest that the large total radiometric heat flow, 10(exp 14) W, and the size and temperature distribution of the thermal anomalies (McEwen et al. 1992; Veeder et al. 1994) can be accounted for by a series of silicate lava flows in various stages of cooling. We propose that the whole suite of Io's currently observed thermal anomalies was produced by multiple, high-eruptive-rate silicate flows within the past century.

Description: This report presents the most recent spherical harmonic topography model of Venus developed at Jet Propulsion Laboratory. It was produced by a spherical harmonic analysis of the most complete set of Magellan altimetry data, augmented by Pioneer Venus and Venera data. The harmonic coefficients of the topography were computed to degree and order 360. Compared to previous topography models, this one has the highest correlation with the gravity field of Venus.

Description: The 500-Myr average crater retention age for Venus has raised questions about the present-day level of tectonic activity. In this study we examine the relationship between the gravity and topography of four large volcanic swells, Beta, Atla, Bell, and Western Eistla Regiones, for clues about their stage evolution. The Magellan line-of-sight gravity data are inverted using a point mass model of the anomalous mass to solve for the local vertical gravity field. Spectral admittance calculated from both the local gravity inversions and a spherical harmonic model is compared to three models of compensation: local compensation, a 'flexural' model with local and regional compensation of surface and subsurface loads, and a 'hotspot' model of compensation that includes top loading by volcanoes and subsurface loading due to a deep, low density mass anomaly. The coherence is also calculated in each region, but yields an elastic thickness estimate only at Bell Regio. In all models, the long wavelengths are compensated locally. Our results may indicate a relatively old, possibly inactive plume.

Description: We investigate the orbital dynamics of small dust particles generated via the continuous micrometeoroid bombardment of the Martian moons. In addition to Mar's oblateness, we also consider the radiation pressure perturbation that is complicated by the planet's eccentric orbit and tilted rotational axis. Considering the production rates and the lifetimes of dust grains, we show that particles from Deimos with radii of about 15 micrometers are expected to dominate the population of a permanently present and tilted dust torus. This torus has an estimated peak number density of approximately equals 5 x 10(exp -12)/cu cm and an optical depth of approximately equals 4 x 10(exp -8).