Segmentation-based classification of hyperspectral imagery using projected and correlation clustering techniques

A partitional clustering-based segmentation is used to carry out supervised classification for hyperspectral images. The main contribution of this study lies in the use of projected and correlation partitional clustering techniques to perform image segmentation. These types of clustering techniques have the capability to concurrently perform clustering and feature/band reduction, and are also able to identify different sets of relevant features for different clusters. Using these clustering techniques segmentation map is obtained, which is combined with the pixel-level support vector machines (SVM) classification result, using majority voting. Experiments are conducted over two hyperspectral images. Combination of pixel-level classification result with the segmentation maps leads to significant improvement of accuracies in both the images. Additionally, it is also observed that, classified maps obtained using SVM combined with projected and correlation clustering techniques results in higher accuracies as compared to classified maps obtained from SVM combined with other partitional clustering techniques.     

Coastal upwelling off the southwest coast of India: observations and simulations

Spatial hydrographic data collected from August 2007 through May 2008 and the Princeton Ocean Model (POM) are utilized to explain and document the upwelling in the south eastern Arabian Sea (SEAS). The decrease in the magnitude of winds towards the coast favors local anti-cyclonic vorticity, resulting in the formation of cyclonic eddy and reversal of coastal currents. The Ekman transport due to alongshore winds, Ekman pumping due to wind stress curl, cyclonic eddy and southward West Indian Coastal Currents play different roles in the SEAS upwelling. In the offshore, wind stress curl leads to the formation of meso-scale eddies, resulting in Ekman pumping at the center and consequent upwelling. The rapid depth variation on the western side of Quilon Mount intensifies upwelling in the offshore. The upwelling Kelvin waves generated at the equator terminates in the Bay of Bengal and do not contribute to the SEAS upwelling. The possible role of local and remote winds, especially around Sri Lanka on the upwelling dynamics of SEAS is described utilizing POM. The Kelvin waves generated off the coast of Sri Lanka during the active phase of the summer monsoon also contribute to the upwelling dynamics off the west coast of India.     

Strongly perturbed quasi-periodic dynamical systems

It is almost impossible to construct a general theory of the motion of a strongly perturbed dynamical system using classical perturbation theory because this approach uses a reference orbit (e.g. a Keplerian ellipse) which is very different from the actual orbit.A general method, pioneered by Jefferys, is presented here. This method allows each quasi-periodic orbit (for instance a strongly perturbed two body problem: JVIII is the typical example) to specify the coordinates to be used. These coordinates are discovered by a truncated infinite series of coordinate transformations. The transformations are implemented using the idea that the nature of a dynamical system is embodied in the symplectic form. The method is illustracted by a simple example.With modern algebraic and series manipulation languages on present day computers all one needs to begin using this approach is a good numerical integration, the end product being a series for each coordinate. Further weak perturbations are easily incorporated into this semi-analytical solution by the usual methods.Proceedings of the Conference on Analytical Methods and Ephemerides: Theory and Observations of the Moon and Planets. Facultés universitaires Notre Dame de la Paix, Namur, Belgium, 28–31 July, 1980.     

Isolating type-specific phenologies through spectral unmixing of satellite time series

Vegetation phenology is commonly studied using time series of multi-spectral vegetation indices derived from satellite imagery. Differences in reflectance among land-cover and/or plant functional types are obscured by sub-pixel mixing, and so phenological analyses have typically sought to maximize the compositional purity of input satellite data by increasing spatial resolution. We present an alternative method to mitigate this ‘mixed-pixel problem’ and extract the phenological behavior of individual land-cover types inferentially, by inverting the linear mixture model traditionally used for sub-pixel land-cover mapping. Parameterized using genetic algorithms, the method takes advantage of the discriminating capacity of calibrated surface reflectance measurements in red, near infrared, and short-wave infrared wavelengths, as well as the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Water Index. In simulation, the unmixing procedure reproduced the reflectances and phenological signals of grass, crop, and deciduous forests with high fidelity (RMSE?相似文献   

Bringing communities at the Centre of Impact Assessment of Road Projects: integrating AHP with qualitative research

GeoJournal - This paper proposes a framework for measuring impacts associated with road projects on rural communities situated on either side of the road alignment. The extant approaches for...     

Estimation of wave directional spreading in shallow water

This paper describes wave directional spreading in shallow water. Waves were measured for a period of 2 months using the Datawell directional waverider buoy at 15 m water depth on the east coast of India in the Bay of Bengal. The study also showed that in shallow water wave directional spreading was narrowest at peak frequency and widened towards lower and higher frequencies. The wind direction was found to deviate from the wave direction during most of the time. The unidirectional spectrum was found to be satisfactorily represented by Scott spectra.     

Quantitative evolved gas analysis: Winchcombe in comparison with other CM2 meteorites

Two bulk Winchcombe along with six other CM2 meteorite samples were subjected to quantitative evolved gas analysis. The observed release patterns for almost all volatile species demonstrate close similarity for all the samples and especially between those for Winchcombe. This can be considered as a fingerprint for this petrological type of meteorites. We identified several gases including H₂, H₂O, O₂, CO, CO₂, and SO₂ released in different temperature ranges. The sources and mechanisms of their release were also established. Some of the gases, H₂, CO, and CO₂, are released as a result of oxidation of macromolecular organic material from oxygen derived from oxygen-bearing minerals (a part of CO₂ is also released as a result of decomposition of carbonates). The others, O₂ and H₂O, are associated with the phase transformation/decomposition of phyllosilicates and (oxy)hydrates, while a high-temperature release of SO₂ is associated mostly with the decomposition of sulfides and in few cases also with sulfates. A low-temperature release of SO₂ is due to evaporation and oxidation of elemental sulfur from the meteoritic matrix and organic material. The total concentrations of H (mostly represented by H₂O), C, and S, calculated according to calibration of the quadrupole mass spectrometer with reference gases and decomposition of solid samples (CaSO₄·2H₂O and NaHCO₃) are in reasonable agreement with those determined by independent methods. Variations in the ratio of the carbon amounts released as CO₂ and CO ( $$/C_CO) between the samples could be an indicator of their terrestrial weathering.     

Wavelet frame-based feature extraction technique for improving classification accuracy

Classification of textures in remotely-sensed data has received considerable attention during the past decades. One difficulty of texture analysis in the past was lack of adequate tools to characterize different scales of textures effectively. Recent space-frequency analytical tools like the wavelet transform can effectively characterize the coupling of different scales in texture and helps to overcome the difficulty. This paper presents a wavelet-based texture classification technique that was applied to a Multi-Spectral Scanner (MSS) image of Madurai City, Tamil Nadu, India The feature extraction stage of the technique uses Lemarie-Battle orthogonal wavelets to derive a texture feature vector and this vector is input to a fuzzy-c means classifier, an unsupervised classification procedure. Four indices (user’s accuracy, producer’s accuracy, overall accuracy and Kappa co-efficient) are used to assess the accuracy of the classified data. The experiment results show that the performance of the presented technique is superior to the classical techniques.     

Shock‐induced microtextures in lunar apatite and merrillite

Apatite and merrillite are the most common phosphate minerals in a wide range of planetary materials and are key accessory phases for in situ age dating, as well as for determination of the volatile abundances and their isotopic composition. Although most lunar and meteoritic samples show at least some evidence of impact metamorphism, relatively little is known about how these two phosphates respond to shock‐loading. In this work, we analyzed a set of well‐studied lunar highlands samples (Apollo 17 Mg‐suite rocks 76535, 76335, 72255, 78235, and 78236), in order of displaying increasing shock deformation stages from S1 to S6. We determined the stage of shock deformation of the rock based on existing plagioclase shock‐pressure barometry using optical microscopy, Raman spectroscopy, and SEM‐based panchromatic cathodoluminescence (CL) imaging of plagioclase. We then inspected the microtexture of apatite and merrillite through an integrated study of Raman spectroscopy, SEM‐CL imaging, and electron backscatter diffraction (EBSD). EBSD analyses revealed that microtextures in apatite and merrillite become progressively more complex and deformed with increasing levels of shock‐loading. An early shock‐stage fragmentation at S1 and S2 is followed by subgrain formation from S2 onward, showing consistent decrease in subgrain size with increasing level of deformation (up to S5) and finally granularization of grains caused by recrystallization (S6). Starting with 2°–3° of intragrain crystal‐plastic deformation in both phosphates at the lowest shock stage, apatite undergoes up to 25° and merrillite up to 30° of crystal‐plastic deformation at the highest stage of shock deformation (S5). Merrillite displays lower shock impedance than apatite; hence, it is more deformed at the same level of shock‐loading. We suggest that the microtexture of apatite and merrillite visualized by EBSD can be used to evaluate stages of shock deformation and should be taken into account when interpreting in situ geochemically relevant analyses of the phosphates, e.g., age or volatile content, as it has been shown in other accessory minerals that differently shocked domains can yield significantly different ages.