This dissertation introduces a novel automated book reader as an assistive technology tool for persons with blindness. The literature shows extensive work in the area of optical character recognition, but the current methodologies available for the automated reading of books or bound volumes remain inadequate and are severely constrained during document scanning or image acquisition processes. The goal of the book reader design is to automate and simplify the task of reading a book while providing a user-friendly environment with a realistic but affordable system design. This design responds to the main concerns of (a) providing a method of image acquisition that maintains the integrity of the source (b) overcoming optical character recognition errors created by inherent imaging issues such as curvature effects and barrel distortion, and (c) determining a suitable method for accurate recognition of characters that yields an interface with the ability to read from any open book with a high reading accuracy nearing 98%. This research endeavor focuses in its initial aim on the development of an assistive technology tool to help persons with blindness in the reading of books and other bound volumes. But its secondary and broader aim is to also find in this design the perfect platform for the digitization process of bound documentation in line with the mission of the Open Content Alliance (OCA), a nonprofit Alliance at making reading materials available in digital form. The theoretical perspective of this research relates to the mathematical developments that are made in order to resolve both the inherent distortions due to the properties of the camera lens and the anticipated distortions of the changing page curvature as one leafs through the book. This is evidenced by the significant increase of the recognition rate of characters and a high accuracy read-out through text to speech processing. This reasonably priced interface with its high performance results and its compatibility to any computer or laptop through universal serial bus connectors extends greatly the prospects for universal accessibility to documentation.

Peatlands are an important store of soil carbon, play a vital role in global carbon cycling, and can also act as sinks of atmospherically deposited heavy metals. Large areas of the UK’s blanket peat are significantly degraded and actively eroding, which negatively impacts carbon and pollutant storage. The restoration of eroding UK peatlands is a major conservation concern, and over the last decade measures have been taken to control erosion and restore large areas of degraded peat. In severely eroded peatlands, topography is highly variable, and an appreciation of geomorphological form and process is key in understanding the controls on peatland function, and in mitigating the negative impacts of peatland erosion. The blanket peats of the Peak District, Southern Pennines, UK, embody many problems and pressures faced by peatlands globally, and are amongst the most heavily eroded and contaminated in the world. The near-surface layer of the peat is contaminated by high concentrations of anthropogenically derived, atmospherically deposited heavy metals, which are released into the fluvial system as a consequence of widespread erosion. Whilst not desirable, this legacy of lead pollution and its release, offer a unique opportunity to trace peatland sediment movements and thus investigate the controls on sediment and contaminant mobility. A suite of established field, analytical and modelling techniques have been modified and adapted for use in peatland environments: (i) by incorporating a simple correction for moisture content, field portable XRF has been shown to be an accurate, cost-effective, and rapid tool for assessing in situ lead concentrations in wet organic sediments; (ii) a lightweight time integrated mass flux sampler has been developed for deployment at multiple remote peatland field sites, and has been used to explore spatial and temporal suspended sediment dynamics; and (iii) sediment source fingerprinting and numerical mixing models, traditionally used to determine sources of fine sediment in minerogenic systems, have been used to investigate suspended sediment composition in contaminated organic rich catchments. These modified methods have been successfully employed in combination to address issues of sediment and contaminant release. Several mechanisms and controls have been shown to be important influences on sediment dynamics and Pb release across a range of spatial and temporal scales: (i) the presence of vegetation is key in stabilising the peat’s surface and trapping mobilised sediment; (ii) sediment preparation influences the timing of POC and Pb release; (iii) antecedent water tables may control the timing and the nature of sediment entering the fluvial system during storm events; and (iv) the degree of degradation influences both Pb storage and release. At the landscape scale, peatland restoration significantly mitigates sediment production in eroding peatlands and substantially reduces carbon and pollutant export. At the catchment scale, sediment preparation and hydrological connectivity are important controls on the magnitude and timing of sediment and lead fluxes from eroding peatland catchments. At the plot scale, complex small scale spatial patterns of contaminant storage in eroding headwater catchments can be explained by interactions between topographic setting and vegetation cover, and the mobilisation of sediment by wind and water. This deeper understanding of the multi-scalar dynamics of sediment movements in eroding peatlands is important in the context of: (i) the release and reworking of legacy contamination in organic rich systems; (ii) the response of blanket peats to climate change; (iii) informing future restoration strategies that aim to manage peatland sediment and contaminant fluxes.

Cette thèse porte sur la restauration de champs d'ondes sismiques perturbés par trois sources de dégradation. Ces sources sont dues à des trajets de propagation complexes, au dispositif d'acquisition, à des sources liées ou non à l'acquisition, et potentiellement présentes simultanément : des réflexions multiples (ou échos), une dégradation de la réponse impulsionnelle attendue (ou flou) et des perturbations plus aléatoires (ou bruits).Nous avons considéré dans un premier temps le problème des réflexions multiples, réflexions qui se sont réfléchies plusieurs fois sur au moins une interface. Nous nous intéressons ici au filtrage adaptatif de ces réflexions sismiques multiples à partir de modèles approximatifs issus de modélisation sismique. Ce filtrage est réalisé dans un domaine de trames d'ondelettes discrètes, mono- et bidimensionnelles, sous contraintes de parcimonie et de variation lente des filtres adaptatifs. Ceci est intéressant en réflexion sismique, car les méthodes standard peuvent produire des filtres très mal conditionnés, du fait notamment du caractère passe-bande des données sismiques. Dans ce travail, une formulation variationnelle des problèmes de réflexions multiples est proposée. Nous utilisons des algorithmes proximaux, dont la convergence est garantie lorsqu'il s'agit d'optimiser dans un cadre convexe. L'avantage de ces approches est l'utilisation d'une régularisation sophistiquée, permettant de considérer la parcimonie à la fois a) dans le domaine d'ondelettes, b) via des a priori sur les filtres pour lesquels nous avons utilisé différentes fonctions de régularisation (norme l1, l2, mixte l1-2 et nucléaire). Notre méthode vise à étendre et améliorer certains aspects de la méthode proposée par S. Ventosa en collaboration avec CGG en 2012, et testée avec succès sur plusieurs campagnes sismiques. Les résultats que nous avons obtenus démontrent la performance de notre méthode non seulement sur des données synthétiques bruitées mais également sur des données réelles. Nous nous intéressons ensuite au problème de déconvolution aveugle. En géophysique, un modèle simplifié de la Terre souvent utilisé fait l'hypothèse d'un nombre de couches localement parallèles, chacune avec des propriétés constantes. Mais la vitesse, la densité ou les deux peuvent varier d'une couche à l'autre. L'impédance acoustique est calculée pour chaque couche ; puis les coefficients de réflexion pour une incidence normale sont calculés aux endroits où il y a des changements d'impédance acoustique. Chaque changement d'impédance acoustique opère une modification d'amplitude et de polarisation liée aux coefficients de réflexion. Ainsi, la séquence de réflectivité (réponse impulsionnelle) est convoluée avec la forme d'onde descendante pour donner une trace sismique. Ce problème constitue un contexte de déconvolution aveugle où l'on recherche un signal inconnu, convolué avec une forme d'onde elle aussi inconnue en présence de bruit additif. La déconvolution requiert souvent d'introduire des hypothèses complémentaires sous forme de pénalisation, notamment non convexe. L'ambiguïté d'échelle en déconvolution aveugle suggère l'usage de fonctions de contraste invariantes en échelle. Dans cette thèse, nous proposons un algorithme de minimisation alternée, de type explicite-implicite à métrique variable. Il traite une approximation lisse du rapport l1/l2 (SOOT pour "Smoothed One-Over-Two penalty") pour des données réelles signées. Nous étudions les propriétés de convergence de la méthode proposée, basées sur l'inégalité de Kurdyka-Lojasiewicz. Les performances de cette nouvelle approche sont illustrées à travers un exemple en déconvolution aveugle de données sismiques, mais aussi sur des images / This thesis deals with two different problems within the framework of convex and non convex optimization. The first one is an application to multiple removal in seismic data with adaptive filters and the second one is an application to blind deconvolution problem that produces characteristics closest to the Earth layers. More precisely : unveiling meaningful geophysical information from seismic data requires to deal with both random and structured “noises”. As their amplitude may be greater than signals of interest (primaries), additional prior information is especially important in performing efficient signal separation. We address here the problem of multiple reflections, caused by wave-field bouncing between layers. Since only approximate models of these phenomena are available, we propose a flexible framework for time-varying adaptive filtering of seismic signals, using sparse representations, based on inaccurate templates. We recast the joint estimation of adaptive filters and primaries in a new convex variational formulation. This approach allows us to incorporate plausible knowledge about noise statistics, datas parsity and slow filter variation in parsimony-promoting wavelet transforms. The designed primal-dual algorithm solves a constrained minimization problem that alleviates standard regularization issues in finding hyper parameters. The approach demonstrates significantly good performance in low signal-to-noise ratio conditions, both for simulatedand real field seismic data. In seismic exploration, a seismic signal (e.g. primary signal) is often represented as the results of a convolution between the “seismic wavelet” and the reflectivity series. The second goal of this thesis is to deconvolve them from the seismic signal which is presented in Chapter 6. The main idea of this work is to use an additional premise that the reflections occur as sparsely restricted, for which a study on the “sparsity measure”is considered. Some well known methods that fall in this category are proposed such as[Sacchi et al., 1994; Sacchi, 1997]. We propose a new penalty based on a smooth approximation of the l1/l2 function that makes a difficult non convex minimization problem. We develop a proximal-based algorithm to solve variational problems involving this function and we derive theoretical convergence results. We demonstrate the effectiveness of our method through a comparison with a recent alternating optimization strategy dealing with the exact l1/l2 term

Wetland loss and degradation have led to the development of restoration programs worldwide, many of which monitor wading bird populations as indicators of wetland quality. Therefore, efficient, standardized monitoring is integral to restoration progress. I tested the use of passive acoustic monitoring to estimate nest abundances and provisioning rates in wading bird colonies and examined regional nesting dynamics in the Florida Everglades, where a long monitoring record enables analysis of nesting patterns relative to hydrologic changes. I found that call rates can serve as indices of colony nest abundances and begging call rate and timing are indicative of provisioning events. Nesting dynamics suggested that resource availability is asynchronous between regions of the Everglades, but the degree of asynchrony varies with species. The conclusions of this study will facilitate the long-term monitoring of wading bird nesting trends, which are important measures of wetland restoration in Florida and worldwide. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2021. / FAU Electronic Theses and Dissertations Collection

Scotts Creek, located in northern Santa Cruz County, maintains the southernmost persistent population of Central California Coast (CCC) Coho Salmon (endangered) in addition to CCC steelhead (threatened). Fisheries biologists believe overwinter mortality due to lack of refuge habitat is the primary factor limiting salmonid production. Instream rearing habitat may also be limiting, especially during drought years. The legacy effects of historic land use practices, including dredging, wood removal, and the construction of levees, continued to limit refuge and rearing opportunities. A restoration project was implemented to improve refuge and rearing opportunities for salmonids along lower Scotts Creek by removing portions of the deteriorating levee, grading new connections with existing off-channel features, enhancing tributary confluences, constructing alcove habitat features at the margins of the stream channel, and constructing large wood complexes (LWCs) instream. Novel restoration techniques were employed on an experimental basis. Whole in-situ alder trees were pushed into the stream channel with their root systems left partially intact to establish living key pieces. Individual log, boulder, and rootwad LWC components were attached together with couplers that permitted some freedom of independent movement among the individual components. LWCs were braced against live, standing trees and stabilized with boulder ballasts placed on the streambed, which eliminated excavation of the streambed/banks and the need to dewater or divert the stream during construction. Project performance, changes to physical habitat characteristics, and changes to stream morphology associated with implementation were monitored using habitat assessment methods derived from the California Department of Fish and Wildlife’s (CDFW) salmonid habitat survey protocol (Flosi et al. 2010), and topographic survey techniques and data analysis adapted from Columbia Habitat Monitoring Protocol (Bouwes et al. 2011). Preliminary results indicated that LWCs remained stable and functional. In addition, implementation of the restoration project increased pool frequency, low-flow pool volume, instream cover, frequency of instream, alcove, and off-channel refuge habitat features, and frequency of points of connectivity with the floodplain. Long-term monitoring will be required to determine the survivorship, decay rates, and overall persistence of alder recruits.

The study deals with the restoration and the new use of the Classicist Chateau in Hrušovany nad Jevišovkou. The castle is now in the possession of the village, it is unused and degraded. The basic idea was to create a castle attractive not only for tourists, but also for the inhabitants of Hrušovany. Another idea was to combine several different functions that connect each other, but they can work independently. There are designed exhibition spaces, wine bar, multifunctional hall, seniors club or chateau-style accommodation. A new building, which serves as a guest house (guest house and restaurant), is proposed at the site of a no longer existing building.

The restoration of degraded tallgrass prairies can mitigate climate change due to the carbon accrued during the development of grasslands. The focal species, dominant grass Andropogon gerardii, can assist the recovery of grassland ecosystem functioning. Climate, local adaptation, and biodiversity have been found to impact the accrual of carbon in grasslands. This study examined the difference in ecosystem functioning between ecotypes along a dry to mesic precipitation scale. The study site for this project was at the Southern Illinois University Agriculture Research Center in Carbondale, Illinois. The field site was planted with seeds originating from dry to mesic ecotypes, and the resulting ecosystem functioning was analyzed. It was found that the Kansas non-local ecotypes had significantly higher biodiversity, while the local Illinois sites demonstrated local adaptation with A. gerardii. Aboveground plant biomass was higher in the local sites, but there was no difference in carbon accrual between any of the ecotypes. While ecotypic variation in a dominant species will usually differentially influence ecosystem functioning, in this case, high biodiversity and local adaptation result in similar carbon inputs in grassland soil. It is necessary to analyze the carbon content of the soil in the drier field sites in order to determine if major differences in rainfall leads to differences in carbon accrual. If the goal of restoring a tallgrass prairie in southern Illinois is to assist with climate change mitigation, then it does not make a significant difference if the dominant species is sourced locally or non-locally.

Exotic annual weeds have transformed western North America, particularly in sagebrush-steppe systems. Restoration of these invaded sites has been met with low levels of success. Pre-emergent herbicide provides a means to control annual weeds, but typically, this treatment does not allow for the concurrent seeding of desired species. Seeding within a deep, U-shaped furrow following herbicide application may be a method to reduce pre-emergent herbicide effects by transferring the herbicide away from the seed at the time of planting. We tested this potential planting technique by spraying plots with or without the pre-emergent herbicide imazapic, and planting bunchgrass seeds either with or without a deep furrow. Treatments (i.e. spraying and furrowing) were applied using mechanical equipment within a single pass, at six sites. In plots without imazapic, we found that deep furrows generally had higher seedling emergence, density of juvenile plants, and above-ground biomass when compared to no furrows. For plots with imazapic, deep furrows also generally improved measured plant metrics for the seeded species compared to plots without furrows. For example, the density of juvenile plants in deep furrows ranged, by study site, between 62% – 97% and 41% – 89% higher than the no furrow treatment, for plots with and without imazapic, respectively. Plots with imazapic and deep furrows was not always as effective as plots without imazapic and deep furrows. Deep furrows also reduced exotic annual weeds in the first year after planting, but weed reduction was generally more effective when this treatment was applied with imazapic. Overall, this research provides evidence that in most instances, the use of deep furrows alone is sufficient to improve seeding success. However, in areas with high weed cover, the application of herbicide followed by the creation of deep furrows in a one-pass system should be considered.

Abstract Nuclear Medicine (NM) images inherently suffer from large amounts of noise and blur. The purpose of this research is to reduce the noise and blur while maintaining image integrity for improved diagnosis. The proposal is to further improve image quality after the standard pre- and post-processing undertaken by a gamma camera system. Mean Field Annealing (MFA), the image processing technique used in this research is a well known image processing approach. The MFA algorithm uses two techniques to achieve image restoration. Gradient descent is used as the minimisation technique, while a deterministic approximation to Simulated Annealing (SA) is used for optimisation. The algorithm anisotropically diffuses an image, iteratively smoothing regions that are considered non-edges and still preserving edge integrity until a global minimum is obtained. A known advantage of MFA is that it is able to minimise to this global minimum, skipping over local minima while still providing comparable results to SA with significantly less computational effort. Image blur is measured using either a point or line source. Both allow for the derivation of a Point Spread Function (PSF) that is used to de-blur the image. The noise variance can be measured using a flood source. The noise is due to the random fluctuations in the environment as well as other contributors. Noisy blurred NM images can be difficult to diagnose particularly at regions with steep intensity gradients and for this reason MFA is considered suitable for image restoration. From the literature it is evident that MFA can be applied successfully to digital phantom images providing improved performance over Wiener filters. In this paper MFA is shown to yield image enhancement of planar NM images by implementing a sharpening filter as a post MFA processing technique.

Throughout midwestern North American ecosystems, semi-aquatic mammals including beaver (Castor canadensis), mink (Neovision vision), muskrat (Ondatra zibethicus), and river otter (Lontra canadensis) co-exist in wetlands. These species are ecologically important through their manipulation of habitats and interactions with other species present. The Lake County Forest Preserve District (Lake County, Illinois) is actively restoring forest preserves using several restoration practices and are interested in how these efforts may affect semi-aquatic mammal occupancy. We studied impacts of restoration practices and other environmental covariates on detection and occupancy of the 4 aforementioned focal species. Sign surveys were conducted during December-April in 2018-19 and 2019-20. Single species, multi-season models were run in RStudio using the package unmarked. Muskrat detection (p̂ = 0.60 ± 0.03) was positively affected by survey replicate, and negatively impacted by last precipitation amount and bank angle. Beaver detection (p̂ = 0.76 ± 0.03) was positively affected by survey replicate and previous 2-week precipitation amount, and negatively influenced by bank angle. Mink detection (p̂ = 0.26 ± 0.04) was positively impacted by survey replicate, and negatively affected by last precipitation amount and previous 2-week precipitation amount. River otter detection (p̂ = 0.10 ± 0.07) was not influenced by any covariates sampled. Muskrat had the highest occupancy (ψ ̂= 0.90 ± 0.05), followed by beaver (ψ ̂ = 0.57 ± 0.07), mink (ψ ̂ = 0.57 ± 0.10), and river otter (ψ ̂ = 0.28 ± 0.18). Both muskrat and river otter occupancy were not affected by any covariates sampled. Beaver occupancy was positively impacted by stream density and number of saplings. Mink occupancy was positively affected by stream density. No focal species’ occupancies were influenced by restoration practices. From these findings, I provide recommendations for future surveys and management for semi-aquatic furbearers in urban landscapes. Thank you to the Lake County Forest Preserve District for funding my research.