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: Our ecological footprint analyses of coral reef fish fisheries and, in particular, the live reef fish food trade (FT), indicate many countries' current consumption exceeds estimated sustainable per capita global, regional and local coral reef production levels. Hong Kong appropriates 25% of SE Asia's annual reef fish production of 135 260-286 560 tonnes (t) through its FT demand, exceeding regional biocapacity by 8.3 times; reef fish fisheries demand out-paces sustainable production in the Indo-Pacific and SE Asia by 2.5 and 6 times. In contrast, most Pacific islands live within their own reef fisheries means with local demand at 〈 20% of total capacity in Oceania. The FT annually requisitions up to 40% of SE Asia's estimated reef fish and virtually all of its estimated grouper yields. Our results underscore the unsustainable nature of the FT and the urgent need for regional management and conservation of coral reef fisheries in the Indo-Pacific.

Description: Since the launch of Landsat-1 28 years ago, remotely sensed data have been used to map features on the earth's surface. An increasing number of health studies have used remotely sensed data for monitoring, surveillance, or risk mapping, particularly of vector-borne diseases. Nearly all studies used data from Landsat, the French Systeme Pour l'Observation de la Terre, and the National Oceanic and Atmospheric Administration's Advanced Very High Resolution Radiometer. New sensor systems are in orbit, or soon to be launched, whose data may prove useful for characterizing and monitoring the spatial and temporal patterns of infectious diseases. Increased computing power and spatial modeling capabilities of geographic information systems could extend the use of remote sensing beyond the research community into operational disease surveillance and control. This article illustrates how remotely sensed data have been used in health applications and assesses earth-observing satellites that could detect and map environmental variables related to the distribution of vector-borne and other diseases.

Description: In August and September of 1995 the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) was deployed to Brazil as part of the NASA Smoke Cloud Aerosol and Radiation experiment in Brazil (SCAR-B). AVIRIS measures spectra from 400 to 2500 nm at 10-nm intervals. These spectra are acquired as images with dimensions of 11 by up to 800 km with 20-m spatial resolution. Spectral images measured by AVIRIS are spectrally, radiometrically, and spatially calibrated. During the SCAR-B deployment, AVIRIS measured more than 300 million spectra of regions of Brazil. A portion of these spectra were acquired over areas of actively burning fires. Actively burning fires emit radiance in the AVIRIS spectral range as a function of temperature. This emitted radiance is expressed from the 2500-nm end of the AVIRIS spectrum to shorter wavelengths as a function of intensity and modeled by the Planck function.. The objective of this research and analysis was to use spectroscopic methods to determine the minimum high temperature of the most intense fires measured in the SCAR-B AVIRIS data set. Spectra measured by AVIRIS with hot sources have been previously examined for volcanic lava and fires in Brazil.

Description: Various locations in the southwestern U.S. are used to calibrate remote sensing instruments. This study shows how some of these targets compare in terms of albedo and homogeneity, and records the variation of these factors for a single location (Ivanpah Playa) over a period of one year. Results indicate that there is a great deal of variation among these targets in albedo, spectral flatness, and surface uniformity, and that these factors can change throughout the year.

Description: Early observations with ERS-1 SAR image data revealed a large ice stream in North East Greenland (Fahnestock 1993). The ice stream has a number of the characteristics of the more closely studied ice streams in Antarctica, including its large size and gross geometry. The onset of rapid flow close to the ice divide and the evolution of its flow pattern, however, make this ice stream unique. These features can be seen in the balance velocities for the ice stream (Joughin 1997) and its outlets. The ice stream is identifiable for more than 700 km, making it much longer than any other flow feature in Greenland. Our research goals are to gain a greater understanding of the ice flow in the northeast Greenland ice stream and its outlet glaciers in order to assess their impact on the past, present, and future mass balance of the ice sheet. We will accomplish these goals using a combination of remotely sensed data and ice sheet models. We are using satellite radar interferometry data to produce a complete maps of velocity and topography over the entire ice stream. We are in the process of developing methods to use these data in conjunction with existing ice sheet models similar to those that have been used to improve understanding of the mechanics of flow in Antarctic ice streams.

Description: This research is focusing on two related areas that are fundamental to the NASA PARCA (Program for Arctic Regional Climate Assessment) program. The primary research area is the determination of the amount, rate, and timing of accumulation at distributed sites in the dry snow zone of Greenland and evaluation of these results in light of accumulation modeling results. The secondary research area is the calibration of the isotope "thermometer" at these ice sheet sites as well as the determination of long-term temperature trends in Greenland.

Description: The Long Valley Caldera located in the eastern Sierra Nevada (California) shows new signs of volcanic activity. This renewed activity is expressed by gas emissions, hydrothermal activity and frequent earthquakes. Analysis of the gas composition regarding the percentage biogenic carbon and the He-3/He-4 ratio revealed that the gas source is the magma body approximately 7 km beneath the Long Valley Caldera. The gas from the magma body surfaces not only via the fumaroles but also emerges along geological faults. Some of the spots where gas surfaces are marked by dead or stressed trees. Other spots may not yet be identified. It is only recently known, from research at 'Vulcano Island' in southern Italy, that volcanoes release abundant carbon dioxide from their flanks as diffuse soil emanations. Mammoth Mountain seems to behave in a similar manner. The research described in this paper is designed to determine whether AVIRIS (Airborne Visible/Infrared Imaging Spectrometer) can be used to identify areas of volcanic gas emissions.