Application of Wavelet Decomposition to Document Line Segmentation

ACM Computing Classification System (1998): I.7, I.7.5.

Stability in Terms of Two Measures for Differential Equations with “Maxima”

Атанаска Георгиева, Стела Глухчева, Снежана Христова - Изследвана е устойчивостта на нелинейни диференциални уравнения с “максимуми” по отношение на две мерки. Приложени са две различни мерки за началните условия и за решението. Използван е методът на Разумихин, а също така и методът на сравнението на обикновени скаларни диференциални уравнения. Приложението на получените резултати и достатъчни условия за устойчивост е илюстрирано с пример.

Functional Transfer Theorems for Maxima of Stationary Processes

2000 Mathematics Subject Classification: 60G70, 60F12, 60G10.

Limit Theorems for Maxima of Heavy-Tailed Terms with Random Dependent Weights

2000 Mathematics Subject Classification: Primary 60F17, 60G52, 60G70 secondary 60E07, 62E20.

Revisiting Offspring Maxima in Branching Processes

2000 Mathematics Subject Classification: 60J80; 60G70.

Lexicographic allocations and extreme core payoffs: the case of assignment games

We consider various lexicographic allocation procedures for coalitional games with transferable utility where the payoffs are computed in an externally given order of the players. The common feature of the methods is that if the allocation is in the core, it is an extreme point of the core. We first investigate the general relationship between these allocations and obtain two hierarchies on the class of balanced games. Secondly, we focus on assignment games and sharpen some of these general relationship. Our main result is the coincidence of the sets of lemarals (vectors of lexicographic maxima over the set of dual coalitionally rational payoff vectors), lemacols (vectors of lexicographic maxima over the core) and extreme core points. As byproducts, we show that, similarly to the core and the coalitionally rational payoff set, also the dual coalitionally rational payoff set of an assignment game is determined by the individual and mixed-pair coalitions, and present an efficient and elementary way to compute these basic dual coalitional values. This provides a way to compute the Alexia value (the average of all lemacols) with no need to obtain the whole coalitional function of the dual assignment game.

Rigid and non-rigid point-based medical image registration

The primary goal of this dissertation is to develop point-based rigid and non-rigid image registration methods that have better accuracy than existing methods. We first present point-based PoIRe, which provides the framework for point-based global rigid registrations. It allows a choice of different search strategies including (a) branch-and-bound, (b) probabilistic hill-climbing, and (c) a novel hybrid method that takes advantage of the best characteristics of the other two methods. We use a robust similarity measure that is insensitive to noise, which is often introduced during feature extraction. We show the robustness of PoIRe using it to register images obtained with an electronic portal imaging device (EPID), which have large amounts of scatter and low contrast. To evaluate PoIRe we used (a) simulated images and (b) images with fiducial markers; PoIRe was extensively tested with 2D EPID images and images generated by 3D Computer Tomography (CT) and Magnetic Resonance (MR) images. PoIRe was also evaluated using benchmark data sets from the blind retrospective evaluation project (RIRE). We show that PoIRe is better than existing methods such as Iterative Closest Point (ICP) and methods based on mutual information. We also present a novel point-based local non-rigid shape registration algorithm. We extend the robust similarity measure used in PoIRe to non-rigid registrations adapting it to a free form deformation (FFD) model and making it robust to local minima, which is a drawback common to existing non-rigid point-based methods. For non-rigid registrations we show that it performs better than existing methods and that is less sensitive to starting conditions. We test our non-rigid registration method using available benchmark data sets for shape registration. Finally, we also explore the extraction of features invariant to changes in perspective and illumination, and explore how they can help improve the accuracy of multi-modal registration. For multimodal registration of EPID-DRR images we present a method based on a local descriptor defined by a vector of complex responses to a circular Gabor filter.

Controls on Ecosystem Carbon Dioxide Exchange in Short- and Long-Hydroperiod Florida Everglades Freshwater Marshes

Although freshwater wetlands are among the most productive ecosystems on Earth, little is known of carbon dioxide (CO2) exchange in low latitude wetlands. The Everglades is an extensive, oligotrophic wetland in south Florida characterized by short- and long-hydroperiod marshes. Chamber-based CO2 exchange measurements were made to compare the marshes and examine the roles of primary producers, seasonality, and environmental drivers in determining exchange rates. Low rates of CO2 exchange were observed in both marshes with net ecosystem production reaching maxima of 3.77 and 4.28 μmol CO2 m−2 s−1 in short- and long-hydroperiod marshes, respectively. Fluxes of CO2 were affected by seasonality only in the short-hydroperiod marsh, where flux rates were significantly lower in the wet season than in the dry season. Emergent macrophytes dominated fluxes at both sites, though this was not the case for the short-hydroperiod marsh in the wet season. Water depth, a factor partly under human control, significantly affected gross ecosystem production at the short-hydroperiod marsh. As Everglades ecosystem restoration proceeds, leading to deeper water and longer hydroperiods, productivity in short-hydroperiod marshes will likely be more negatively affected than in long-hydroperiod marshes. The Everglades stand in contrast to many freshwater wetlands because of ecosystem-wide low productivity rates.

Leaf Functional Traits and Forest Structure of Tropical Dry Forest Species Along a Rainfall Gradient in Florida and Puerto Rico

This study examined how different rainfall regimes affect a set of leaf functional traits related to plant stress and forest structure in tropical dry forest (TDF) species on limestone substrate. One hundred fifty eight individuals of four tree species were sampled in six ecological sites in south Florida and Puerto Rico, ranging in mean annual rainfall from 858 to 1933 mm yr-1. Leaf nitrogen content, specific leaf area (SLA), and N:P ratio of evergreen species, but not deciduous species, responded positively to increasing rainfall. Phosphorus content was unaffected in both groups. Canopy height and basal area reached maxima of 10.3 m and 31.4 m2 ha-1, respectively, at 1168 mm annual rainfall. Leaf traits reflected soil properties only to a small extent. This led us to the conclusion that water is a major limiting factor in TDF and some species that comprise TDF ecosystems are limited by nitrogen in limestone sites with less than ~1012 mm rainfall, but organismal, biological and/or abiotic forces other than rainfall control forest structure in moister sites.

Rigid and Non-rigid Point-based Medical Image Registration

The primary goal of this dissertation is to develop point-based rigid and non-rigid image registration methods that have better accuracy than existing methods. We first present point-based PoIRe, which provides the framework for point-based global rigid registrations. It allows a choice of different search strategies including (a) branch-and-bound, (b) probabilistic hill-climbing, and (c) a novel hybrid method that takes advantage of the best characteristics of the other two methods. We use a robust similarity measure that is insensitive to noise, which is often introduced during feature extraction. We show the robustness of PoIRe using it to register images obtained with an electronic portal imaging device (EPID), which have large amounts of scatter and low contrast. To evaluate PoIRe we used (a) simulated images and (b) images with fiducial markers; PoIRe was extensively tested with 2D EPID images and images generated by 3D Computer Tomography (CT) and Magnetic Resonance (MR) images. PoIRe was also evaluated using benchmark data sets from the blind retrospective evaluation project (RIRE). We show that PoIRe is better than existing methods such as Iterative Closest Point (ICP) and methods based on mutual information. We also present a novel point-based local non-rigid shape registration algorithm. We extend the robust similarity measure used in PoIRe to non-rigid registrations adapting it to a free form deformation (FFD) model and making it robust to local minima, which is a drawback common to existing non-rigid point-based methods. For non-rigid registrations we show that it performs better than existing methods and that is less sensitive to starting conditions. We test our non-rigid registration method using available benchmark data sets for shape registration. Finally, we also explore the extraction of features invariant to changes in perspective and illumination, and explore how they can help improve the accuracy of multi-modal registration. For multimodal registration of EPID-DRR images we present a method based on a local descriptor defined by a vector of complex responses to a circular Gabor filter.

Water temperature reconstructions based on paired Uk'37 and Tex86 records, and paired surface and subsurface Mg/Ca records

This dataset contains the collection of available published paired Uk'37 and Tex86 records spanning multi-millennial to multi-million year time scales, as well as a collection of Mg/Ca-derived temperatures measured in parallel on surface and subsurface dwelling foraminifera, both used in the analyses of Ho and Laepple, Nature Geoscience 2016. As the signal-to-noise ratios of proxy-derived Holocene temperatures are relatively low, we selected records that contain at least the last deglaciation (oldest sample >18kyr BP).

Distribution of Upper Cretaceous deep water agglutinated foraminifera in sediments of the North Atlantic and its marginal seas (Tab. 1)

The stratigraphic and biogeographic distribution of more than 170 species of deep-water agglutinated benthic foraminifers (DWAF) from the North Atlantic and adjacent marginal seas has been compared with paleoenvironmental data (e.g. paleobathymetry, oxygenation of the bottom waters, amount of terrigenous input and substrate disturbance). Six general types of assemblages, in which deep water agglutinated taxa occur, are defined from the Turonian to Maastrichtian times: 1. High latitude slope assemblages 2. Low to mid latitude slope assemblages 3. Flysch-type assemblages 4. Deep water limestone assemblages (,,Scaglia,,-type) 5. Abyssal mixed calcareous-agglutinated assemblages 6. Abyssal purely agglutinated assemblages Latitudinal differences in faunal composition are observed, the most important of which is the lack or extreme paucity of calcareous forms in high latitude assemblages. East-to-west differences appear to be of comparatively minor importance. Most DWAF species occur in all studied regions and are thus considered as cosmopolitan. Biostratigraphic turnovers in the taxonomic content of assemblages are observed in the lowermost Turonian, mid-Campanian and in the upper Maastrichtian to lowermost Paleocene. These datum levels correspond to inter-regional and time-constant paleooceanographic events, which probably also affected the deep-water benthic biota. This allows us to use deep-water agglutinated foraminifers for biostratigraphy in the North Atlantic sequences deposited below CCD and to geographically extend the currently used zonal schemes which have been established in the Carpathian and Alpine areas.

Estimation of sea surface temperatures and salinity of ODP Site 177-1089 (Appendix A)

ODP Site 1089 is optimally located in order to monitor the occurrence of maxima in Agulhas heat and salt spillage from the Indian to the Atlantic Ocean. Radiolarian-based paleotemperature transfer functions allowed to reconstruct the climatic history for the last 450 kyr at this location. A warm sea surface temperature anomaly during Marine Isotope Stage (MIS) 10 was recognized and traced to other oceanic records along the surface branch of the global thermohaline (THC) circulation system, and is particularly marked at locations where a strong interaction between oceanic and atmospheric overturning cells and fronts occurs. This anomaly is absent in the Vostok ice core deuterium, and in oceanic records from the Antarctic Zone. However, it is present in the deuterium excess record from the Vostok ice core, interpreted as reflecting the temperature at the moisture source site for the snow precipitated at Vostok Station. As atmospheric models predict a subtropical Indian source for such moisture, this provides the necessary teleconnection between East Antarctica and ODP Site 1089, as the subtropical Indian is also the source area of the Agulhas Current, the main climate agent at our study location. The presence of the MIS 10 anomaly in the delta13C foraminiferal records from the same core supports its connection to oceanic mechanisms, linking stronger Agulhas spillover intensity to increased productivity in the study area. We suggest, in analogy to modern oceanographic observations, this to be a consequence of a shallow nutricline, induced by eddy mixing and baroclinic tide generation, which are in turn connected to the flow geometry, and intensity, of the Agulhas Current as it flows past the Agulhas Bank. We interpret the intensified inflow of Agulhas Current to the South Atlantic as responding to the switch between lower and higher amplitude in the insolation forcing in the Agulhas Current source area. This would result in higher SSTs in the Cape Basin during the glacial MIS 10, due to the release into the South Atlantic of the heat previously accumulating in the subtropical and equatorial Indian and Pacific Ocean. If our explanation for the MIS 10 anomaly in terms of an insolation variability switch is correct, we might expect that a future Agulhas SSST anomaly event will further delay the onset of next glacial age. In fact, the insolation forcing conditions for the Holocene (the current interglacial) are very similar to those present during MIS 11 (the interglacial preceding MIS 10), as both periods are characterized by a low insolation variability for the Agulhas Current source area. Natural climatic variability will force the Earth system in the same direction as the anthropogenic global warming trend, and will thus lead to even warmer than expected global temperatures in the near future.

Stable carbon and oxygen isotope ratios of planktonic foraminifera from the equatorial Atlantic

Carbon isotopic records of nutrient-depleted surface water place constraints on the past fertility of the oceans and on past atmospheric pCO2 levels. The best records of nutrient-depleted delta13C are obtained from planktonic foraminifera living in the thick mixed layers of the western equatorial and tropical Atlantic Ocean. We have produced a composite, stacked Globigerinoides sacculifer delta13C record from the equatorial Atlantic, which exhibits significant spectral power at the 100,000- and 41,000-year Milankovitch periods, but no power at the 23,000-year period. Similar to the record presented by Shackleton and Pisias [1985], surface-deep ocean Delta delta13C produced with the G. sacculifer record leads the delta18O ice volume record. However, the glacial-interglacial amplitudes of Delta delta13C differ between our record and Shackleton and Pisias [1985] record. Although large changes in Delta delta13C occur in the equatorial Atlantic during early stages of the last three glacial cycles, surface-deep Delta delta13C at glacial maxima (18O stage 2, late stage 6, and late stage 8) was only about 0.2? greater than during the subsequent interglacial. Our results imply that nutrient-driven pCO2 changes account for about one third of the pCO2 decrease observed in ice cores, and consequently, Delta delta13C should not be used as a proxy pCO2 index. Enough variance in the ice core pCO2 records remains to be explained that conclusions about pCO2 and ice volume phase relationships should also be reexamined. As much as 40 ppm pCO2 change still has not been accounted for by models of past physics and chemistry of the ocean.