Refine
Year of publication
- 2021 (40) (remove)
Document Type
- Doctoral Thesis (40) (remove)
Language
- English (40) (remove)
Keywords
- Umweltpsychologie (2)
- optimal control (2)
- Acceleration Structures (1)
- Amphibia (1)
- Aphid predator (1)
- Arzneimittel (1)
- Basic psychological needs (1)
- Bees (1)
- Bestäuber (1)
- Bestäubung (1)
Institute
- Fachbereich 7 (6)
- Institut für Integrierte Naturwissenschaften, Abt. Chemie (5)
- Institut für Computervisualistik (4)
- Institut für Integrierte Naturwissenschaften, Abt. Biologie (4)
- Institut für Wirtschafts- und Verwaltungsinformatik (3)
- Institute for Web Science and Technologies (3)
- Mathematisches Institut (3)
- Fachbereich 6 (2)
- Fachbereich 8 (2)
- Institut für Informatik (2)
Augmented reality (AR) applications typically extend the user's view of the real world with virtual objects.
In recent years, AR has gained increasing popularity and attention, which has led to improvements in the required technologies. AR has become available to almost everyone.
Researchers have made great progress towards the goal of believable AR, in which the real and virtual worlds are combined seamlessly.
They mainly focus on issues like tracking, display technologies and user interaction, and give little attention to visual and physical coherence when real and virtual objects are combined. For example, virtual objects should not only respond to the user's input; they should also interact with real objects. Generally, AR becomes more believable and realistic if virtual objects appear fixed or anchored in the real scene, appear indistinguishable from the real scene, and response to any changes within it.
This thesis examines on three challenges in the field of computer vision to meet the goal of a believable combined world in which virtual objects appear and behave like real objects.
Firstly, the thesis concentrates on the well-known tracking and registration problem. The tracking and registration challenge is discussed and an approach is presented to estimate the position and viewpoint of the user so that virtual objects appear fixed in the real world. Appearance-based line models, which keep only relevant edges for tracking purposes, enable absolute registration in the real world and provide robust tracking. On the one hand, there is no need to spend much time creating suitable models manually. On the other hand, the tracking can deal with changes within the object or the scene to be tracked. Experiments have shown that the use of appearance-based line models improves the robustness, accuracy and re-initialization speed of the tracking process.
Secondly, the thesis deals with the subject of reconstructing the surface of a real environment and presents an algorithm to optimize an ongoing surface reconstruction. A complete 3D surface reconstruction of the target scene
offers new possibilities for creating more realistic AR applications. Several interactions between real and virtual objects, such as collision and occlusions, can be handled with physical correctness. Whereas previous methods focused on improving surface reconstructions offline after a capturing step, the presented method de-noises, extends and fills holes during the capturing process. Thus, users can explore an unknown environment without any preparation tasks such as moving around and scanning the scene, and without having to deal with the underlying technology in advance. In experiments, the approach provided realistic results where known surfaces were extended and filled in plausibly for different surface types.
Finally, the thesis focuses on handling occlusions between the real and virtual worlds more realistically, by re-interpreting the occlusion challenge as an alpha matting problem. The presented method overcomes limitations in state-of-the-art methods by estimating a blending coefficient per pixel of the rendered virtual scene, instead of calculating only their visibility. In several experiments and comparisons with other methods, occlusion handling through alpha matting worked robustly and overcame limitations of low-cost sensor data; it also outperformed previous work in terms of quality, realism and practical applicability.
The method can deal with noisy depth data and yields realistic results in regions where foreground and background are not strictly separable (e.g. caused by fuzzy objects or motion blur).
Despite the significant presence of neuroactive substances in the environment, bioassays that allow to detect diverse groups of neuroactive mechanisms of action are not well developed and not properly integrated into environmental monitoring and chemical regulation. Therefore, there is a need to develop testing methods which are amenable for fast and high-throughput neurotoxicity testing. The overall goal of this thesis work is to develop a test method for the toxicological characterization and screening of neuroactive substances and their mixtures which could be used for prospective and diagnostic hazard assessment.
In this thesis, the behavior of zebrafish embryos was explored as a promising tool to distinguish between different neuroactive mechanisms of action. Recently, new behavioral tests have been developed including photomotor response (PMR), locomotor response (LMR) and spontaneous tail coiling (STC) tests. However, the experimental parameters of these tests lack consistency in protocols such as exposure time, imaging time, age of exposure, endpoint parameter etc. To understand how experimental parameters may influence the toxicological interpretation of behavior tests, a systematic review of existing behavioral assays was conducted in Chapter 2. Results show that exposure concentration and exposure duration highly influenced the comparability between different test methods and the spontaneous tail coiling (STC) test was selected for further testing based on its relative higher sensitivity and capacity to detect neuroactive substances (Chapter 2).
STC is the first observable motor activity generated by the developing neural network of the embryo which is assumed to occur as a result of the innervation of the muscle by the primary motor neurons. Therefore, STC could be a useful endpoint to detect effect on the muscle innervation and also the on the whole nervous system. Consequently, important parameters of the STC test were optimized and an automated workflow to evaluate the STC with the open access software KNIME® was developed (Chapter 3).
To appropriately interpret the observed effect of a single chemical and especially mixture effects, requires the understanding of toxicokinetics and biotransformation. Most importantly, the biotransformation capacity of zebrafish embryos might be limited and this could be a challenge for assessment of chemicals such as organophosphates which require a bioactivation step to effectively inhibit the acetylcholinesterase (AChE) enzyme. Therefore, the influence of the potential limited biotransformation on the toxicity pathway of a typical organophosphate, chlorpyrifos, was investigated in Chapter 5. Chlorpyrifos could not inhibit AChE and this was attributed to possible lack of biotransformation in 24 hpf embryos (Chapter 5).
Since neuroactive substances occur in the environment as mixtures, it is therefore more realistic to assess their combined effect rather than individually. Therefore, mixture toxicity was predicted using the concentration addition and independent action models. Result shows that mixtures of neuroactive substances with different mechanisms of action but similar effects can be predicted with concentration addition and independent action (Chapter 4). Apart
from being able to predict the combined effect of neuroactive substances for prospective risk assessment, it is also important to assess in retrospect the combined neurotoxic effect of environmental samples since neuroactive substances are the largest group of chemicals occurring in the environment. In Chapter 6, the STC test was found to be capable of detecting neurotoxic effects of a wastewater effluent sample. Hence, the STC test is proposed as an effect based tool for monitoring environmental acute and neurotoxic effects.
Overall, this thesis shows the utility and versatility of zebrafish embryo behavior testing for screening neuroactive substances and this allows to propose its use for prospective and diagnostic hazard assessment. This will enhance the move away from expensive and demanding animal testing. The information contained in this thesis is of great potential to provide precautionary solutions, not only for the exposure of humans to neuroactive chemicals but for the environment at large.
Wild bees are essential for the pollination of wild and cultivated plants. However, within the
last decades, the increasing intensification of modern agriculture has led to both a reduction and fragmentation as well as a degradation of the habitats wild bees need. The resulting loss of pollinators and their pollination poses an immense challenge to global food production. To support wild bees, the availability of flowering resources is essential. However, the flowering period of each resource is temporally limited and has different effects on pollinators and their pollination, depending on the time of their flowering.
Therefore, to efficiently promote and manage wild bee pollinators in agricultural landscapes, we identified species-specific key floral resources of three selected wild bee species and their spatial and temporal availability (CHAPTERS 2, 3 & 4). We examined, which habitat types predominantly provide these resources (CHAPTERS 3 & 4). We also investigated whether floral resource maps based on the use of these key resources and their spatial and temporal availability explain the abundance and development of the selected wild bees (CHAPTERS 3 & 4) and pollination (CHAPTER 5) better than habitat maps, that only indirectly account for the availability of floral resources.
For each of the species studied, we were able to identify different key pollen sources, predominantly woody plants in the early season (April/May) and increasingly herbaceous plants in the later season (June/July; CHAPTERS 2, 3 & 4). The open woody semi-natural habitats of our agricultural landscapes provided about 75% of the floral resources for the buff-tailed bumblebees, 60% for the red mason bees, and 55% for the horned mason bees studied, although they accounted for only 3% of the area (CHAPTERS 3 & 4). In addition, fruit orchards provided about 35% of the floral resources for the horned mason bees on 4% of the landscape area (CHAPTER 3). We showed that both mason bee species benefited from the resource availability in the surrounding landscapes (CHAPTER 3). Yet this was not the case for the bumblebees (CHAPTER 4). Instead, the weight gain of their colonies, the number of developed queen cells and their colony survival were higher with increasing proximity to forests. The proximity to forests also had a positive effect on the mason bees studied (CHAPTER 3). In addition, the red mason bees benefited from herbaceous semi-natural habitats. The proportion of built-up areas had a negative effect on the horned mason bees, and the proportion of arable land on the red mason bees. The habitat maps explained horned mason bee abundances equally well as the floral resource maps, but red mason bee abundances were distinctly better explained by key floral resources. The pollination of field bean increased with higher proportions of early floral resources, whereas synchronous floral resources showed no measurable reduction in their pollination (CHAPTER 5). Habitat maps also explained field bean pollination better than floral resource maps. Here, pollination increased with increasing proportions of built-up areas in the landscapes and decreased with increasing proportions of arable land.
Our results highlight the importance of the spatio-temporal availability of certain key species as resource plants of wild bees in agricultural landscapes. They show that habitat maps are ahead of, or at least equal to, spatio-temporally resolved floral resource maps in predicting wild bee development and pollination. Nevertheless, floral resource maps allow us to draw more accurate conclusions between key floral resources and the organisms studied. The proximity to forest edges had a positive effect on each of the three wild bee species studied. However, besides pure food availability, other factors seem to co-determine the occurrence of wild bees in agricultural landscapes.
Efficient Cochlear Implant (CI) surgery requires prior knowledge of the cochlea’s size and its characteristics. This information helps to select suitable implants for different patients. Registered and fused images helps doctors by providing more informative image that takes advantages of different modalities. The cochlea’s small size and complex structure, in addition to the different resolutions and head positions during imaging, reveals a big challenge for the automated registration of the different image modalities. To obtain an automatic measurement of the cochlea length and the volume size, a segmentation method of cochlea medical images is needed. The goal of this dissertation is to introduce new practical and automatic algorithms for the human cochlea multi-modal 3D image registration, fusion, segmentation and analysis. Two novel methods for automatic cochlea image registration (ACIR) and automatic cochlea analysis (ACA) are introduced. The proposed methods crop the input images to the cochlea part and then align the cropped images to obtain the optimal transformation. After that, this transformation is used to align the original images. ACIR and ACA use Mattes mutual information as similarity metric, the adaptive stochastic gradient descent (ASGD) or the stochastic limited memory Broyden–Fletcher–Goldfarb–Shanno (s-LBFGS) optimizer to estimate the parameters of 3D rigid transform. The second stage of nonrigid registration estimates B-spline coefficients that are used in an atlas-model-based segmentation to extract cochlea scalae and the relative measurements of the input image. The image which has segmentation is aligned to the input image to obtain the non-rigid transformation. After that the segmentation of the first image, in addition to point-models are transformed to the input image. The detailed transformed segmentation provides the scala volume size. Using the transformed point-models, the A-value, the central scala lengths, the lateral and the organ of corti scala tympani lengths are computed. The methods have been tested using clinical 3D images of total 67 patients: from Germany (41 patients) and Egypt (26 patients). The atients are of different ages and gender. The number of images used in the experiments is 217, which are multi-modal 3D clinical images from CT, CBCT, and MRI scanners. The proposed methods are compared to the state of the arts ptimizers related medical image registration methods e.g. fast adaptive stochastic gradient descent (FASGD) and efficient preconditioned tochastic gradient descent (EPSGD). The comparison used the root mean squared distance (RMSE) between the ground truth landmarks and the resulted landmarks. The landmarks are located manually by two experts to represent the round window and the top of the cochlea. After obtaining the transformation using ACIR, the landmarks of the moving image are transformed using the resulted transformation and RMSE of the transformed landmarks, and at the same time the fixed image landmarks are computed. I also used the active length of the cochlea implant electrodes to compute the error aroused by the image artifact, and I found out an error ranged from 0.5 mm to 1.12 mm. ACIR method’s RMSE average was 0.36 mm with a standard deviation (SD) of 0.17 mm. The total time average required for registration of an image pair using ACIR was 4.62 seconds with SD of 1.19 seconds. All experiments are repeated 3 times for justifications. Comparing the RMSE of ACIR2017 and ACIR2020 using paired T-test shows no significant difference (p-value = 0.17). The total RMSE average of ACA method was 0.61 mm with a SD of 0.22 mm. The total time average required for analysing an image was 5.21 seconds with SD of 0.93 seconds. The statistical tests show that there is no difference between the results from automatic A-value method and the manual A-value method (p-value = 0.42). There is no difference also between length’s measurements of the left and the right ear sides (p-value > 0.16). Comparing the results from German and Egypt dataset shows there is no difference when using manual or automatic A-value methods (p-value > 0.20). However, there is a significant difference when using ACA2000 method between the German and the Egyptian results (p-value < 0.001). The average time to obtain the segmentation and all measurements was 5.21 second per image. The cochlea scala tympani volume size ranged from 38.98 mm3 to 57.67 mm3 . The combined scala media and scala vestibuli volume size ranged from 34.98 mm 3 to 49.3 mm 3 . The overall volume size of the cochlea should range from 73.96 mm 3 to 106.97 mm 3 . The lateral wall length of scala tympani ranged from 42.93 mm to 47.19 mm. The organ-of-Corti length of scala tympani ranged from 31.11 mm to 34.08 mm. Using the A-value method, the lateral length of scala tympani ranged from 36.69 mm to 45.91 mm. The organ-of-Corti length of scala tympani ranged from 29.12 mm to 39.05 mm. The length from ACA2020 method can be visualised and has a well-defined endpoints. The ACA2020 method works on different modalities and different images despite the noise level or the resolution. In the other hand, the A-value method works neither on MRI nor noisy images. Hence, ACA2020 method may provide more reliable and accurate measurement than the A-value method. The source-code and the datasets are made publicly available to help reproduction and validation of my result.
Over the past few decades, Single-Particle Analysis (SPA), in combination with cryo-transmission electron microscopy, has evolved into one of the leading technologies for structural analysis of biological macromolecules. It allows the investigation of biological structures in a close to native state at the molecular level. Within the last five years the achievable resolution of SPA surpassed 2°A and is now approaching atomic resolution, which so far has only been possible with Xray crystallography in a far from native environment. One remaining problem of Cryo-Electron Microscopy (cryo-EM) is the weak image contrast. Since the introduction of cryo-EM in the 1980s phase plates have been investigated as a potential tool to overcome these contrast limitations. Until now, technical problems and instrumental deficiencies have made the use of phase plates difficult; an automated workflow, crucial for the acquisition of 1000s of micrographs needed for SPA, was not possible. In this thesis, a new Zernike-type Phase Plate (PP) was developed and investigated. Freestanding metal films were used as a PP material to overcome the ageing and contamination problems of standard carbon-based PPs. Several experiments, evaluating and testing various metals, ended with iridium as the best-suited material. A thorough investigation of the properties of iridium PP followed in the second part of this thesis. One key outcome is a new operation mode, the rocking PP. By using this rocking-mode, fringing artifacts, another obstacle of Zernike PPs, could be solved. In the last part of this work, acquisition and reconstruction of SPA data of apoferritin was performed using the iridium PP in rocking-mode. A special semi-automated workflow for the acquisition of PP data was developed and tested. The recorded PP data was compared to an additional reference dataset without a PP, acquired following a conventional workflow.
The stands surveyed are among the last closed canopy forests in Rwanda. Their exploration began in the early twentieth century and is still ongoing. Previous studies were mainly concerned with plant sociological issues and presented references to environmental factors in anecdotal form, at best using indirect ordination methods. The present study undertakes a classification of the vegetation with numerical methods and establishes quantitative relationships of the species’ distributional structure to environmental parameters using spatially explicit procedures. For this purpose, 94 samples were taken in 100 m² hexagonal plots. Of these, 70 samples are from Nyungwe, 14 are from Gishwati, and 10 are from Cyamudongo. Given the homogeneity of the terrain and vegetation, all vegetation types encountered, all types of stands, and all vegetation strata were included. The beta diversity is expressed by an average Bray-Curtis dissimilarity of 0.92, and in JOST’S (2007) numbers equivalents, 37.90 equally likely samples would be needed to represent the diversity encountered. Within the survey, 1198 species in 127 families were collected. Among the specimens are 6 local endemics and 40 Albertine Rift endemics. Resulting from UPGMA and FCM-NC, 20 to 40 plant communities were established depending on the level of resolution. It can be inferred by means of a Mantel correlogram that the mean zone of influence of a single vegetation stand, as sampled by a 100 m² plot in Nyungwe Forest, ranges between 0.016 and 3.42 km. Of the communities compiled using FCM-NC and UPGMA, 50% consist of individual samples. Beyond undersampling, natural small-scale discontinuities are reflected by this result. Partial db-RDA resulted in an explained variation of 9.60% and 14.41% for environmental and soil factors, respectively. Utilising variation partitioning analyses based on CCA and tb-RDA, between 21.70% and 37.80% of the variation in vegetation data could be explained. The spatially structured fraction of these parameters accounts for between 30.50% and 49.80% of the explained variation (100%). The purely environmental parameters account for a share of 10.30% to 16.30%, whereby the lower limit originates from the unimodal approach and has lost its statistical significance. The soil variables, also after partial analysis, account for a share of 19.00% to 35.70%. While the residual impact of the climatic parameters is hardly significant, the effect of the soil properties is prevalent. In general, the spatially structured fraction of the parameters is predominant here. While on the broad-scale climatic factors, the altitude a.s.l. and the geology are determining factors, some soil parameters and matrix components also show their impacts here. In the mid-range of the scale, it is the forest matrix, the soil types, and the geology that determine species distribution. While in the fine range of the scale, some unrecorded parameters seem to have an effect, there are also neutral processes that determine species composition.
The protected areas of Rwanda are facing various challenges resulting from the anthropogenic activities of the surrounding communities especially in the adjacent area to Cyamudongo isolated rain forest, which results in climate change, soil degradation, and loss of biodiversity. Therefore, this study aims to broaden current knowledge on the impact of sustainable Agroforestry (AF) on the Carbon (C) stock and Biodiversity conservation on the surroundings of Cyamudongo isolated rain forest and Ruhande Arboretum.
To understand this, the permanent sample plots (PSPs) were established mainly in the designed four transects of four km long originating on the boundary of the Cyamudongo isolated rain forest following the slope gradient ranging from 1286 to 2015 m asl. A total number of 73 PSPs were established in the Cyamudongo study area while 3 PSPs were established in the Ruhande AF plot. The Arc Map GIS 10.4 was used to design and map the sampling areas while GPS was used for localization of collected items. Statistical significance was analyzed through the R-software especially for wood and soil variables while for biodiversity indicator species, MVSP Software 3.0 was used to determine the Shannon Diversity indices and similarities among species.
In this study, I have obtained comprehensive results demonstrating that in all study areas, the various AF tree species contribute differently to C stock and C sequestration and the amount of C stored and removed from the atmosphere depends on different factors such as tree species, plantation density, growth stage, or the age of establishment, applied management practices, wood specific density (WSD), wood C concentration, and climatic conditions. The estimated quantity of sequestrated C for 2 years and 34 years AF species were 13.11 t C ha -1 yr-1 (equivalent to 48 t CO2 ha -1 yr-1) and 6.85 t ha-1 yr-1 (equivalent to 25.1 t CO2 ha -1 yr-1) in Cyamudongo and Ruhande respectively. The estimated quantity of C stored by the Ruhande AF plot is 232.94 t ha-1. In Cyamudongo, the overall C stored by the AF systems was 823 t ha-1 by both young tree species established by the Cyamudongo Project (35.84 t ha-1) and C stored by existed AF species before the existence of the Project (787.12 t ha-1). In all study areas, the Grevillea robusta was found to contribute more to overall stored C compared to other species under this study.
The tests revealed differences in terms of nutrient contents (C, N, C: N ratio, K, Na, Ca, and Mg) for various AF tree species of Cyamudongo and Ruhande study areas. The differences in terms of correlation for various variables of AF tree species in different study areas varied with tree species, age, stage of growth, and tree shape. By comparing the correlation coefficients for various tree variables for young and mature AF tree species, the results showed a high correlation variability for young species than mature or old species recorded in different environmental conditions of Cyamudongo and Ruhande study areas.
The recorded soil pH mean value across in Cyamudongo study area is 4.2, which is very strongly acidic. The tests revealed that the soil pH, C, C: N ratio, OM, NH4+, NO3-+NO2-, PO43-, and CEC were significantly (P < 0.05) different in various soil depths whereas the N was not statistically significant. The pH, N, C: N ratio, CEC, NH4+, PO43-, and Al3+ showed a significant difference across land uses whereas the C and NO3-+NO2- did not show any statistical difference. All tested chemical elements showed a statistical difference as far as altitude ranges are concerned. The only NH4+, PO43-, and CEC showed significant differences with time whereas all other remaining chemical elements did not show any statistical significance. The bulk density of soil was statistically different across land uses and altitude ranges. The soil pH was very strongly correlated with CEC, Mg, and Ca in cropland (CL) whereas it was strongly correlated in both AF and natural forest (NF) except for Mg, which was moderately correlated in AF. Furthermore, its correlation with K was strong in CL, moderate in AF while it was weak in NF. Finally, the pH correlation with Na was weak in both AF and CL whereas it was negligible in NF. The overall estimated soil C stock of the study area was 16848 t ha -1.
The sustainable AF practices changed significantly the frequency of reptiles, amphibians, and flowering plants while there was no statistical change observed on ferns with time. In terms of species richness, 16 flowering plants, 14 ferns, 5 amphibians, and 3 reptiles were recorded and monitored. These findings add to a growing body of literature on the impact of AF on the C stock, soil improvement, and Biodiversity. It is recommended that further researches should be undertaken for the contribution of other AF tree species to the C stock found in the agricultural landscape around all protected areas of Rwanda and the impact on them on the soil and biodiversity.
Human population pressure increased with the population growth around the NNP and Cyamudongo with disturbance impacts on the forests isolating populations into fragments and today, Cyamudongo natural forest is located a way at a distance of at least 8.5 km horizontal distance to Nyungwe main block with a surface area estimated at 300 ha. Under Cyamudongo project implementation, there was a need to understand how the flora diversity responded to human imposed challenges and to forest restoration initiatives. Three physiognomic landscapes forest were identified and considered for three phases of vegetation survey in Cyamudongo and related to the closest area of Nyungwe main block. In this study, 15 transects were laid in each physiognomic forest landscape and 10 and 5 plots were set respectively in Cyamudongo and Gasumo study area. In total, 315 phytosociological relevés were performed and the Braun-Blanquet methods used for three times vegetation surveys. Species life-forms and chorophyte were evaluated and tree species dbh and height have been measured. Data were subjected to different statistical analyses using different softwares such as PAST, R 3.5.2, and SPSS. The mapping was done using Arc GIS and the Multi-Spectral Remote Sensing used to find NDVI for the vegetation classification.
NDVI trends showed that there has been fluctuations in vegetation classifications of the studied area. In this study, 494 vascular plant species from 106 families were harbored in the study area and distributed differently among forest landscapes and study phases. Although, 43.54% were common to Cyamudongo and Gasumo landscapes while 48.54 % of species diversity were hold only by Cyamudongo and 7.92% confined to Gasumo and 12 in total were found new records for Rwanda while several others suspected require detailed research for identification showing how the flora diversity of Cyamudongo is of special interest and extremely important for discoveries.
The finding of the study on diversity indices, the PCA, CA and the Cluster analysis, all statistical analyses (MANOVA, ANOVA) and life form spectra unanimously showed that the anthropogenic disturbance shaped the vegetation cover, the floristic composition, the species diversity, the forest landscapes community structure, the life form spectrum and the phytoclimate of Cyamudongo and Gasumo forest landscapes. Although, the vegetation analysis couldn’t clearly identify communities and sub-communities at the initial and final vegetation surveys and cluster groups were heterogeneous as well as overlapping and species associations not clearly defined due to the high level of similarities in species composition among forest landscapes and vegetation surveys. The species diversity was found high in secondary forest and Gasumo landscape forest and low in the primary forest and the buffer zone of Cyamudongo and the disturbance with gaps openings was found to be associated to the species diversity with a seasonal variation. The patterns of dbh for the buffer zone and of the size classes of all landscapes with an inverted ‘J’ indicated a healthy regeneration in the forest landscapes and tree species explained a good regeneration and recruitment capacity. Different shapes in the pattern of dbh with respectively an inverted ‘J’, ‘J ‘and ‘U’ for the buffer zone, primary and together the secondary and Gasumo forest landscapes indicated differences in the landscapes health and degree of regeneration and recruitment capacity.
Findings from differents measuements showed at which extend human activities have shaped the flora diversity and structure of forest landcapes studied. For instance, disturbances due human activities were daily oberved and trees were logged by neighboring communities such as Batwa populations at Cyamudongo and local populations at Gasumo. Some species were evenly observed targeted for their barks such as Ocotea usambarensis, Parinari excelsa for medicines and many others for their wood quality, fire wood collection and for agricultural purposes.
In the period of Cyamudongo project implementation, important achievements included the increase of forest biomass and therefore the photosynthetic capacity and the evapotranspiration potential that influence the rainfall regime; the regulation of weather conditions and then species diversity; supporting local communities and limiting human activities; raising awareness on conservation and protection of biodiversity and improving of living conditions of neighboring populations by providing paid employment and so to restore to the Cyamudongo forest ecosystem functions. Moreover, Cyamudongo forest remains vulnerable as surrounded by local communities with a high population pressure relying on forest resources for its survival. Cyamudongo harbors a high level of endemism and is a small hotspot for biodiversity conservation. It is therefore recommended to strengthen conservation and protection measures and continue the support of local communities.
Climate change is an existential threat to human survival, the social organization of society, and the stability of ecosystems. It is thereby profoundly frightening. In the face of threat, people often want to protect themselves instead of engaging in mitigating behaviors. When psychological resources are insufficient to cope, people often respond with different forms of denial. In this dissertation, I contribute original knowledge to the understanding of the multifaceted phenomenon of climate denial from a psychological perspective.
There are four major gaps in the literature on climate denial: First, the spectrum of climate denial as a self-protective response to the climate crisis has not received attention within psychology. Second, basic psychological need satisfaction, a fundamental indicator of human functioning and the ability to cope with threat, has not been investigated as a predictor of climate denial. Third, relations of the spectrum of climate denial to climate-relevant emotions, specifically climate anxiety, have not been examined empirically. Forth, it has not been investigated how the spectrum of climate denial relates to established predictors of climate denial, namely right-wing ideological convictions and male gender. To address those gaps, I investigate what the spectrum of climate denial looks like in the German context and how it relates to basic psychological need satisfaction and frustration, pro-environmental behavior, climate anxiety, ideological conviction, and gender.
Five manuscripts reveal that climate denial exists on a spectrum in the German context, ranging from the distortion of facts (interpretive climate denial, specifically denial of personal and global outcome severity) to the denial of the implications of climate change (implicatory climate denial, specifically avoidance, denial of guilt, and rationalization of one's own involvement). Across analyses, low basic psychological need satisfaction predicted the spectrum of climate denial, which was negatively related to pro-environmental behavior. Climate denial was generally negatively related to climate anxiety, except for a positive association of avoidance and climate anxiety. Right-wing ideological conviction was the strongest predictor of climate denial across the spectrum. However, low need satisfaction and male gender were additional weaker predictors of implicatory climate denial.
These findings suggest that the spectrum of climate denial serves many psychological functions. Climate denial is possibly both a self-protective strategy to downregulate emotions and to protect oneself from loss of privilege. In short, it represents a barrier to climate action that may only be resolved once people have sufficient psychological resources to face the threat of climate change and cope with their underlying self-protective, emotional responses.
Content and Language Integrated Learning (CLIL) has experienced growing importance in the last decades and an increasing number of schools have already implemented CLIL programmes or are planning to do so. Though the potentials of CLIL programmes are widely praised, first research results also raise doubts if CLIL students can live up to these high expectations. Both Fehling (2005) as well as Rumlich (2013; 2016), for example, found that CLIL programmes not inevitably show the expected results but that the CLIL students’ success might also be at least partially explained by other influences, such as the selection process of future CLIL students. Hence, CLIL students apparently fall short of the high expectations that are usually connected to the respective CLIL programmes and as this is mainly based on the unsatisfactory quality of these programmes, Rumlich concludes that “it is now high time to focus on the quality of CLIL provision” (Rumlich 2016: 452). He continues to explain that “the promises of CLIL do not materialise automatically owing to the fact that another language is used for learning in a non-language subject” (Rumlich 2016: 452). It must be assumed that the success of CLIL teaching also highly depends on the quality of the CLIL teachers.
In contrast to the continuously growing number of CLIL schools, however, the number of specifically trained CLIL teachers is comparably small. In Germany, CLIL teachers are not (yet) required to attend any special training in order to teach in a CLIL programme. Notwithstanding, is it sufficient for a CLIL teacher only to be trained in the content subject and the foreign language? Or does CLIL teaching require more than the sum of these two components? Do CLIL teachers need additional teaching competences to the ones of a content and a language teacher? In the light of the recent findings of CLIL programmes falling short of the high expectations, the answer to these questions must clearly be “Yes”. Hence, in order to appropriately train (future) CLIL teachers, special training programmes need to be developed which consider the teachers’ individual educational backgrounds, i.e. their qualifications as language and/or as content teachers and build up on these competences through adding the CLIL-specific teaching competences.
Therefore, this thesis aims at developing a German Framework for CLIL Teacher Education, which considers both the already published, theory-based standards of CLIL teacher education as well as the practical perspective of experienced CLIL teachers in Germany. This German Framework for CLIL Teacher Education classifies the different teaching competences, which are derived from integrating the theoretical and the practical perspective on CLIL teacher education, with regard to the three different competence areas, i.e. the general teaching competence, the language teaching competence and the subject teaching competence and is hence adaptable to different CLIL settings and educational backgrounds. In addition to developing this German Framework for CLIL Teacher Education, which provides the content of future CLIL teacher education programmes, this thesis discusses different forms of structurally implementing CLIL teacher education programmes in the existing structures of teacher education in Germany. This is achieved through analysing the current state of the art of CLIL teacher education at German universities and systematising the different forms of implementing these training programmes in the prevailing educational structures. Building on these first two steps, in the third and final step, this thesis develops a CLIL teacher education programme at a German university that is based on the results and elements of the German Framework for CLIL Teacher Education as well as the state of the art of CLIL teacher education in Germany. Thus, this thesis is allocated at the intersection between foreign language teaching as well as teacher education and is structured in eleven chapters.
SUMMARY
Buildings and infrastructures characterize the appearance of our cultural landscapes and provide essential services for the human society. However, they inevitably impact the natural environment e.g. by the structural change of habitats. Additionally, they potentially cause further negative environmental impacts due to the release of chemical substances from construction materials. Galvanic anodes and organic coatings regularly used for corrosion protection of steel structures are building materials of particular importance for the transport infrastructure. In direct contact with a water body or indirectly via the runoff after rainfall, numerous chemicals can be released into the environment and pose a risk to aquatic organisms. Up to now, there is no uniform investigation and evaluation approach for the assessment of the environmental compatibility of building products. Furthermore, galvanic anodes and organic coatings pose particular challenges for their ecotoxicological characterization due to their composition. Therefore, the objective of the presented thesis was the ecotoxicological assessment of emissions from galvanic anodes and protective coatings as well as the development of standardized assessment procedures for these materials.
The possible environmental hazard posed by the use of anodes on offshore installations was investigated on three trophic levels. To ensure a realistic and reliable evaluation, the experiments were carried out in natural seawater and under natural pH conditions. Moreover, the anode material and its main components zinc and aluminum were exposed while simulating a worst-case scenario. The anode material examined caused a weak inhibition of algae growth; no acute toxicity was observed on the luminescent bacteria and amphipods. However, an increase of aluminum and indium levels in the crustacean species was found. On the basis of these results, no direct threat has been identified for marine organisms from the use of galvanic aluminum anodes. However, an accumulation of metals in crustaceans and a resulting entry into the marine food web cannot be excluded.
The environmental compatibility of organic coating systems was exemplarily evaluated using a selection of relevant products based on epoxy resins (EP) and polyurethanes. For this purpose, coated test plates were dynamically leached over 64 days. The eluates obtained were systematically analyzed for their ecotoxicological effects (acute toxicity to algae and luminescent bacteria, mutagenic and estrogenic effects) and their chemical composition. In particular, the EP-based coatings caused significant bacterial toxicity and estrogen-like effects. The continuously released 4-tert-butylphenol was identified as a main contributor to these effects and was quantified in concentrations exceeding the predicted no effect concentration for freshwater in all samples. Interestingly, the overall toxicity was not governed by the content of 4-tert-butylphenol in the products but rather by the release mechanism of this compound from the investigated polymers. This finding indicates that an optimization of the composition can result in the reduction of emissions and thus of environmental impacts - possibly due to a better polymerization of the compounds.
Coatings for corrosion protection are exposed to rain, changes in temperature and sun light leading to a weathering of the polymer. To determine the influence of light-induced aging on the ecotoxicity of top coatings, the emissions and associated adverse effects of UV-irradiated and untreated EP-based products were compared. To that end, the investigation of static leachates was focused on estrogenicity and bacterial toxicity, which were detected in the classic microtiter plate format and in combination with thin-layer plates. Both materials examined showed a significant decrease of the ecotoxicological effects after irradiation with a simultaneous reduction of the 4-tert-butylphenol emission. However, bisphenol A and various structural analogues were detected as photolytic degradation products of the polymers, which also contributed to the observed effects. In this context, the identification of bioactive compounds was supported by the successful combination of in-vitro bioassays with chemical analysis by means of an effect-directed analysis. The presented findings provide important information to assess the general suitability of top coatings based on epoxy resins.
Within the scope of the present study, an investigation concept was developed and successfully applied to a selection of relevant construction materials. The adaptation of single standard methods allowed an individual evaluation of these products. At the same time, the suitability of the ecotoxicological methods used for the investigation of materials of unknown and complex composition was confirmed and the basis for a systematic assessment of the environmental compatibility of corrosion protection products was created. Against the background of the European Construction Products Regulation, the chosen approach can facilitate the selection of environmentally friendly products and contributes to the optimization of individual formulations by the simple comparison of different building materials e.g. within a product group.
Connected vehicles will have a tremendous impact on tomorrow’s mobility solutions. Such systems will heavily rely on information delivery in time to ensure the functional reliability, security and safety. However, the host-centric communication model of today’s networks questions efficient data dissemination in a scale, especially in networks characterized by a high degree of mobility. The Information-Centric Networking (ICN) paradigm has evolved as a promising candidate for the next generation of network architectures. Based on a loosely coupled communication model, the in-network processing and caching capabilities of ICNs are promising to solve the challenges set by connected vehicular systems. In such networks, a special class of caching strategies which take action by placing a consumer’s anticipated content actively at the right network nodes in time are promising to reduce the data delivery time. This thesis contributes to the research in active placement strategies in information-centric and computation-centric vehicle networks for providing dynamic access to content and computation results. By analyzing different vehicular applications and their requirements, novel caching strategies are developed in order to reduce the time of content retrieval. The caching strategies are compared and evaluated against the state-of-the-art in both extensive simulations as well as real world deployments. The results are showing performance improvements by increasing the content retrieval (availability of specific data increased up to 35% compared to state-of-the-art caching strategies), and reducing the delivery times (roughly double the number of data retrieval from neighboring nodes). However, storing content actively in connected vehicle networks raises questions regarding security and privacy. In the second part of the thesis, an access control framework for information-centric connected vehicles is presented. Finally, open security issues and research directions in executing computations at the edge of connected vehicle networks are presented.
Enterprise collaboration platforms are increasingly gaining importance in organisations. Integrating groupware functionality and enterprise social software (ESS), they have substantially been transforming everyday work in organisations. While traditional collaboration systems have been studied in Computer Supported Cooperative Work (CSCW) for many years, the large-scale, infrastructural and heterogeneous nature of enterprise collaboration platforms remains uncharted. Enterprise collaboration platforms are embedded into organisations’ digital workplace and come with a high degree of complexity, ambiguity, and generativity. When introduced, they are empty shells with no pre-determined purposes of use. They afford interpretive flexibility, and thus are shaping and being shaped by and in their social context. Outcomes and benefits emerge and evolve over time in an open-ended process and as the digital platform is designed through use. In order to make the most of the platform and associated continuous digital transformation, organisations have to develop the necessary competencies and capabilities.
Extant literature on enterprise collaboration platforms has proliferated and provide valuable insights on diverse topics, such as implementation strategies, adoption hurdles, or collaboration use cases, however, they tend to disregard their evolvability and related multiple time frames and settings. Thus, this research aims to identify, investigate, and theorise the ways that enterprise collaboration platforms are changing over time and space and the ways that organisations build digital transformation capabilities. To address this research aim two different case study types are conducted: i) in-depth longitudinal qualitative case study, where case narratives and visualisations capturing hard-to-summarise complexities in the enterprise collaboration platform evolution are developed and ii) multiple-case studies to capture, investigate, and compare cross-case elements that contribute to the shaping of enterprise collaboration platforms in different medium-sized and large organisations from a range of industries. Empirical data is captured and investigated through a multi-method research design (incl. focus groups, surveys, in-depth interviews, literature reviews, qualitative content analysis, descriptive statistics) with shifting units of analysis. The findings reveal unique change routes with unanticipated outcomes and transformations, context-specific change strategies to deal with multiple challenges (e.g. GDPR, works council, developments in the technological field, competing systems, integration of blue-collar workers), co-existing platform uses, and various interacting actors from the immediate setting and broader context. The interpretation draws on information infrastructure (II) as a theoretical lens and related sociotechnical concepts and perspectives (incl. inscriptions, social worlds, biography of artefacts). Iteratively, a conceptual model of the building of digital transformation capabilities is developed, integrating the insights gained from the study of enterprise collaboration platform change and developed monitoring change tools (e.g. MoBeC framework). It assists researchers and practitioners in understanding the building of digital transformation capabilities from a theoretical and practical viewpoint and organisations implement the depicted knowledge in their unique digital transformation processes.
Cultural eutrophication due to excessive inputs of nutrients seriously threatens aquatic ecosystems worldwide and is one of the major anthropogenic stressors on aquatic biota in European rivers. In streams and shallow rivers, its effects include excessive periphyton growth, which causes biological clogging and thereby oxygen depletion in the hyporheic zone. The result is a serious degradation of habitat quality for benthic invertebrates as well as for the eggs and larvae of gravel-spawning fish. Unlike in standing waters, efficient tools for controlling eutrophication in rivers are lacking. However, top-down control of the food-web by manipulating fish stocks, similar to the biomanipulation successfully applied in lakes, offers a promising approach to mitigating the effects of eutrophication in shallow rivers, especially those in which major reductions in nutrient inputs are not feasible. The overall aim of this thesis was to assess the potential for top-down control by two large cypriniform fish, the common nase (Chondrostoma nasus), the only obligate herbivorous fish species in European rivers, and the omnivorous European chub (Squalius cephalus), to mitigate the effects of eutrophication in medium-sized rivers. I therefore conducted field experiments on different spatial and temporal scales in the hyporhithral zone of a eutrophic gravel-bed river. Generally, the results of those experiments revealed the crucial role of fish-mediated top-down effects in river food webs. In a 4-year reach-scale experiment, the key contribution of my thesis, the enhancement of fish densities significantly increased both oxygen availability and water exchange in the upper layer of the hyporheic zone, even though the top-down effects of the fish on periphyton biomass were relatively small. These findings were supported by those of a 4-week mesocosm experiment, which also provided insights into the mechanisms underlying the mitigation of eutrophication effects by nase and chub. The top-down effects of both fish species reduced hyporheic oxygen depletion, suggesting a reduction of biological clogging. The positive effects of herbivorous nase on hyporheic oxygen availability could be attributed to benthic grazing, whereas the reduction of hyporheic oxygen depletion in the presence of omnivorous chub was best explained by the enhanced bioturbation induced by the fish’s benthic foraging. Overall, the results of my thesis demonstrate that biomanipulation achieved by enhancing herbivorous and omnivorous fish stocks can mitigate the effects of eutrophication in medium-sized European rivers. The results may be the first step towards the establishment of biomanipulation as a supportive management tool for eutrophication control in running waters and therefore as a strategy to preserve aquatic biodiversity.
The belief in a just world in face of injustice: victim, observer, and perpetrator perspectives
(2021)
Injustice happens every day either to us, to our neighbors, or people across the world. Yet, believing that the world is a fair place helps us to cope with this injustice and motivates us to behave fairly. Scholars have found that these functions that the belief in a just world (BJW) serves are crucial for maintaining mental health. However, the conditions under which BJW is functional and when people give up this belief are not well studied. The current dissertation aims to examine: when the BJW can be shattered, the role of the external world and other internal resources in face of injustice, and the role of BJW in predicting corrupt behavior. Three studies were conducted corresponding to each party of injustice: a victim, an observer, and a perpetrator.
Study 1 examined the effects of criminal victimization on BJW and buffering role of perceptions of justice in the criminal justice process. A cross-sectional study showed that victims of very severe crimes such as domestic violence and human trafficking had lower personal BJW than non-victims and victims of less severe crimes, and higher informational justice perceptions reduced the effect of victimization on the personal BJW. Study 2 aimed to test the changes in BJW after observing severe injustice. A longitudinal study showed that after observing school rampage attacks that happened at other schools, BJW of adolescent participants increased. Moreover, life satisfaction and perceived social support moderated the change of BJW. Study 3 examined relationships between BJW and corrupt behavior. A cross-sectional study showed that personal BJW can predict bribery behavior.
The findings of three studies provided evidence that BJW does not function in isolation. An external world and internal resources can reduce the threat of injustice on BJW. BJW plays an important role in predicting unfair behavior therefore authorities should aim to maintain the BJW of their citizens.
Vertebrate biodiversity is rapidly decreasing worldwide with amphibians being the most endangered vertebrate group. In the EU, 21 of 89 amphibian species are recognized as being endangered. The intensively used European agricultural landscape is one of the major causes for these declines. As agriculture represents an essential habitat for amphibians, exposure to pesticides can have adverse effects on amphibian populations. Currently, the European risk assessment of pesticides for vertebrates requires specific approaches for fish regarding aquatic vertebrate toxicity and birds as well as mammals for terrestrial vertebrate toxicity but does not address the unique characteristics of amphibians. Therefore, the overall goal of this thesis was to investigate the ecotoxicological effects of pesticides on Central European anuran amphibians. For this, effects on aquatic and terrestrial amphibian life stages as well as on reproduction were investigated. Then, in anticipation of a risk assessment of pesticides for amphibians, this thesis discussed potential regulatory risk assessment approaches.
For the investigated pesticides and amphibian species, it was observed that the acute aquatic toxicity of pesticides can be addressed using the existing aquatic risk assessment approach based on fish toxicity data. However, lethal as well as sublethal effects were observed in terrestrial juveniles after dermal exposure to environmentally realistic pesticide concentrations, which cannot be covered using an existing risk assessment approach. Therefore, pesticides should also be evaluated for potential terrestrial toxicity using risk assessment tools before approval. Additionally, effects of co-formulants and adjuvants of pesticides need specific consideration in a future risk assessment as they can increase toxicity of pesticides to aquatic and terrestrial amphibian stages. The chronic duration of combined aquatic and terrestrial exposure was shown to affect amphibian reproduction. Currently, such effects cannot be captured by the existing risk assessment as data involving field scenarios analysing effects of multiple pesticides on amphibian reproduction are too rare to allow comparison to data of other terrestrial vertebrates such as birds and mammals. In the light of these findings, future research should not only address acute and lethal effects, but also chronic and sublethal effects on a population level. As pesticide exposure can adversely affect amphibian populations, their application should be considered even more carefully to avoid further amphibian declines. Overall, this thesis emphasizes the urgent need for a protective pesticide risk assessment for amphibians to preserve and promote stable amphibian populations in agricultural landscapes.
The Stereotype Content Modell (SCM; Fiske et al., 2002) proposes two fundamental dimensions of social evaluation: Warmth, or the intentions of the target, and Competence, or the ability to enact these intentions. The practical applications of the SCM are very broad and have led to an assumption of universality of warmth and competence as fundamental dimensions of social evaluation.
This thesis has identified five mainly methodological shortcomings of the current SCM research and literature: (I) An insufficient initial scale development; (II) the usage of varying warmth and competence scales without sufficient scale property assessment in later research; (III) the dominant application of first-generation analytical approaches; (IV) the insufficient definition and empirical proof for the SCM’s assumption of universality; and (V) the limited application of the SCM for some social targets. These shortcomings were addressed in four article manuscripts strictly following open science recommendations.
Manuscript # 1 re-analysed published research using English SCM measures to investigate the measurement properties of the used warmth and competence scales. It reported the scales’ reliability, dimensionality and comparability across targets as well as the indicator-based parameter performance in a (multiple group) confirmatory factor analysis framework. The findings indicate that about two thirds of all re-analysed scales do not show the theoretically expected warmth and competence dimensionality. Moreover, only about eleven per cent allowed meaningful mean value comparisons between targets. Manuscript # 2 presents a replication of Manuscript # 1 in the national and language of German(y) generating virtually identical results as Manuscript # 1 did. Manuscript # 3 investigated the stereotype content of refugee subgroups in Germany. We showed that refugees was generally perceived unfavourably in terms of warmth and competence, but that the stereotype content varied based on the refugees’ geographic origin, religious affiliation, and flight motive. These results were generated using a reliability-corrected approach to compare mean values named alignment optimisation procedure. Manuscript # 4 developed and tested a high-performing SCM scale assessing occupational stereotypes a number of exploratory and confirmatory factor analyses.
The Web is an essential component of moving our society to the digital age. We use it for communication, shopping, and doing our work. Most user interaction in the Web happens with Web page interfaces. Thus, the usability and accessibility of Web page interfaces are relevant areas of research to make the Web more useful. Eye tracking is a tool that can be helpful in both areas, performing usability testing and improving accessibility. It can be used to understand users' attention on Web pages and to support usability experts in their decision-making process. Moreover, eye tracking can be used as an input method to control an interface. This is especially useful for people with motor impairment, who cannot use traditional input devices like mouse and keyboard. However, interfaces on Web pages become more and more complex due to dynamics, i.e., changing contents like animated menus and photo carousels. We need general approaches to comprehend dynamics on Web pages, allowing for efficient usability analysis and enjoyable interaction with eye tracking. In the first part of this thesis, we report our work on improving gaze-based analysis of dynamic Web pages. Eye tracking can be used to collect the gaze signals of users, who browse a Web site and its pages. The gaze signals show a usability expert what parts in the Web page interface have been read, glanced at, or skipped. The aggregation of gaze signals allows a usability expert insight into the users' attention on a high-level, before looking into individual behavior. For this, all gaze signals must be aligned to the interface as experienced by the users. However, the user experience is heavily influenced by changing contents, as these may cover a substantial portion of the screen. We delineate unique states in Web page interfaces including changing contents, such that gaze signals from multiple users can be aggregated correctly. In the second part of this thesis, we report our work on improving the gaze-based interaction with dynamic Web pages. Eye tracking can be used to retrieve gaze signals while a user operates a computer. The gaze signals may be interpreted as input controlling an interface. Nowadays, eye tracking as an input method is mostly used to emulate mouse and keyboard functionality, hindering an enjoyable user experience. There exist a few Web browser prototypes that directly interpret gaze signals for control, but they do not work on dynamic Web pages. We have developed a method to extract interaction elements like hyperlinks and text inputs efficiently on Web pages, including changing contents. We adapt the interaction with those elements for eye tracking as the input method, such that a user can conveniently browse the Web hands-free. Both parts of this thesis conclude with user-centered evaluations of our methods, assessing the improvements in the user experience for usability experts and people with motor impairment, respectively.
We consider variational discretization of three different optimal control problems.
The first being a parabolic optimal control problem governed by space-time measure controls. This problem has a nice sparsity structure, which motivates our aim to achieve maximal sparsity on the discrete level. Due to the measures on the right hand side of the partial differential equation, we consider a very weak solution theory for the state equation and need an embedding into the continuous functions for the pairings to make sense. Furthermore, we employ Fenchel duality to formulate the predual problem and give results on solution theory of both the predual and the primal problem. Later on, the duality is also helpful for the derivation of algorithms, since the predual problem can be differentiated twice so that we can apply a semismooth Newton method. We then retrieve the optimal control by duality relations.
For the state discretization we use a Petrov-Galerkin method employing piecewise constant states and piecewise linear and continuous test functions in time. For the space discretization we choose piecewise linear and continuous functions. As a result the controls are composed of Dirac measures in space-time, centered at points on the discrete space-time grid. We prove that the optimal discrete states and controls converge strongly in L^q and weakly-* in Μ, respectively, to their smooth counterparts, where q ϵ (1,min{2,1+2/d}] is the spatial dimension. The variational discrete version of the state equation with the above choice of spaces yields a Crank-Nicolson time stepping scheme with half a Rannacher smoothing step.
Furthermore, we compare our approach to a full discretization of the corresponding control problem, precisely a discontinuous Galerkin method for the state discretization, where the discrete controls are piecewise constant in time and Dirac measures in space. Numerical experiments highlight the sparsity features of our discrete approach and verify the convergence results.
The second problem we analyze is a parabolic optimal control problem governed by bounded initial measure controls. Here, the cost functional consists of a tracking term corresponding to the observation of the state at final time. Instead of a regularization term for the control in the cost functional, we consider a bound on the measure norm of the initial control. As in the first problem we observe a sparsity structure, but here the control resides only in space at initial time, so we focus on the space discretization to achieve maximal sparsity of the control. Again, due to the initial measure in the partial differential equation, we rely on a very weak solution theory of the state equation.
We employ a dG(0) approximation of the state equation, i.e. we choose piecewise linear and continuous functions in space, which are piecewise constant in time for our ansatz and test space. Then, the variational discretization of the problem together with the optimality conditions induce maximal discrete sparsity of the initial control, i.e. Dirac measures in space. We present numerical experiments to illustrate our approach and investigate the sparsity structure
As third problem we choose an elliptic optimal control governed by functions of bounded variation (BV) in one space dimension. The cost functional consists of a tracking term for the state and a BV-seminorm in terms of the derivative of the control. We derive a sparsity structure for the derivative of the BV control. Additionally, we utilize the mixed formulation for the state equation.
A variational discretization approach with piecewise constant discretization of the state and piecewise linear and continuous discretization of the adjoint state yields that the derivative of the control is a sum of Dirac measures. Consequently the control is a piecewise constant function. Under a structural assumption we even get that the number of jumps of the control is finite. We prove error estimates for the variational discretization approach in combination with the mixed formulation of the state equation and confirm our findings in numerical experiments that display the convergence rate.
In summary we confirm the use of variational discretization for optimal control problems with measures that inherit a sparsity. We are able to preserve the sparsity on the discrete level without discretizing the control variable.
Human action recognition from a video has received growing attention in computer vision and has made significant progress in recent years. Action recognition is described as a requirement to decide which human actions appear in videos. The difficulties involved in distinguishing human actions are due to the high complexity of human behaviors as well as appearance variation, motion pattern variation, occlusions, etc. Many applications use human action recognition on captured video from cameras, resulting in video surveillance systems, health monitoring, human-computer interaction, and robotics. Action recognition based on RGB-D data has increasingly drawn more attention to it in recent years. RGB-D data contain color (Red, Green, and Blue (RGB)) and depth data that represent the distance from the sensor to every pixel in the object (object point). The main problem that this thesis deals with is how to automate the classification of specific human activities/actions through RGB-D data. The classification process of these activities utilizes a spatial and temporal structure of actions. Therefore, the goal of this work is to develop algorithms that can distinguish these activities by recognizing low-level and high-level activities of interest from one another. These algorithms are developed by introducing new features and methods using RGB-D data to enhance the detection and recognition of human activities. In this thesis, the most popular state-of-the-art techniques are reviewed, presented, and evaluated. From the literature review, these techniques are categorized into hand-crafted features and deep learning-based approaches. The proposed new action recognition framework is based on these two categories that are approved in this work by embedding novel methods for human action recognition. These methods are based on features extracted from RGB-D data that are
evaluated using machine learning techniques. The presented work of this thesis improves human action recognition in two distinct parts. The first part focuses on improving current successful hand-crafted approaches. It contributes into two significant areas of state-of-the-art: Execute the existing feature detectors, and classify the human action in the 3D spatio-temporal domains by testing a new combination of different feature representations. The contributions of this part are tested based on machine learning techniques that include unsupervised and supervised learning to evaluate this suitability for the task of human action recognition. A k-means clustering represents the unsupervised learning technique, while the supervised learning technique is represented by: Support Vector Machine, Random Forest, K-Nearest Neighbor, Naive Bayes, and Artificial Neural Networks classifiers. The second part focuses on studying the current deep-learning-based approach and how to use it with RGB-D data for the human action recognition task. As the first step of each contribution, an input video is analyzed as a sequence of frames. Then, pre-processing steps are applied to the video frames, like filtering and smoothing methods to remove the noisy data from each frame. Afterward, different motion detection and feature representation methods are used to extract features presented in each frame. The extracted features
are represented by local features, global features, and feature combination besides deep learning methods, e.g., Convolutional Neural Networks. The feature combination achieves an excellent accuracy performance that outperforms other methods on the same RGB-D datasets. All the results from the proposed methods in this thesis are evaluated based on publicly available datasets, which illustrate that using spatiotemporal features can improve the recognition accuracy. The competitive experimental results are achieved overall. In particular, the proposed methods can be better applied to the test set compared to the state-of-the-art methods using the RGB-D datasets.