Unsupervised clustering methods on remote sensing images have shown good results. However, this type of machine learning needs additional labelling to be an end-to-end classification in the same manner as traditional supervised classification. The automation of the labelling needs further exploration. We want to investigate the robustness of a supervised automatic labelling scheme by comparing a ...

Classic target decomposition methods use scattering space in their approaches. However, the goal for this project is to investigate whether a different approach to retrieve accurate and reliable estimates on the earth composition is possible when using the feature space with covariance matrix-based features. The approach consists of four steps. Generating multidimensional feature space data from ...

Sediment cores from Kløverbladvatna, a threshold lake in Wahlenbergfjorden, Nordaustlandet, Svalbard were used to reconstruct Holocene glacier fluctuations. Meltwater from Etonbreen spills over a threshold to the lake, only when the glacier is significantly larger than at present. Lithological logging, loss-on-ignition, ITRAX scanning and radiocarbon dating of the cores show that Kløverbladvatna ...

Remote sensing of planetary surfaces is an effective method for gaining knowledge of the processes that shape the planetary bodies in our solar system. This is useful for uncovering the environment of the primordial solar system and to study the current state of the upper crusts of the other planets in our neighborhood. A recent 6-meter wavelength polarimetric radar map of the Moon showed unexpectedly ...

Denne masteroppgaven tar for seg praktisk arbeid i fysikk i form av en åpen øvelse på universitetsnivå, hvordan øvelsen legger til rette for kreativitet og planlegging av egne metoder. Forskningen på utforskende arbeid, derunder åpne øvelser, er delt. Oppgaven tar hensyn til kritikken og presenterer områder hvor åpne øvelser kan bidra til utvikling av kreativitet og kompetanse i utforskende arbeid. I ...

This thesis aims to better understand the variability of glacial erosion along glacial troughs located in high latitude settings, through a case study of Kongsfjorden, Svalbard. The research question of this thesis is to identify the controlling factors on variable glacial modification of the bedrock dominated areas in Kongsfjorden. Fieldwork was conducted to analyze bedrock structures and ...

The Zermatt-Saas Fee zone represents an ophiolitic sequence within the Western Alps derived from the paleo ocean Liguro-Piemontaise. The melting of the Unterer Theodulgletcher glacier newly uncovered some Tc-Mgs and Atg-Dol veins within serpentinite host rock. Both the serpentinite and veins are fresh as they have been protected from weathering by the glacier, which makes this area ideal for ...

Smoke Particle Impact Detector (SPID) is a faraday cup impact probe designed and built by the University of Tromso (UIT). Its main purpose is to measure nanometer sized smoke particles (in-situ) in the atmosphere, and to do that it needs to be launched on a sounding rocket. Its design is an open faraday cup with grids to shield out ambient plasma and a larger slanted impact grid to measure the ...

This thesis presents the first sub-radar beam resolution determination of the angular width of wide altitude ion line enhancements (WAILEs), found to be 0.5 degrees around magnetic zenith, observed during a HF radio wave heating experiment at the EISCAT Tromsø Heating facility on 27 November 2014. The results of ray tracing simulations are detailed. The simulations are based on the hypothesis that ...

Deep image clustering is a rapidly growing branch of machine learning and computer vision, in which deep neural networks are trained to discover groups within a set of images, in an unsupervised manner. Deep neural networks have proven to be immensely successful in several machine learning tasks, but the majority of these advances have been in supervised settings. The process of labeling data for ...