Refine
Has Fulltext
- yes (4)
Year of publication
- 2021 (4) (remove)
Document Type
- Master's Thesis (4) (remove)
Language
- English (4) (remove)
Is part of the Bibliography
- yes (4) (remove)
Keywords
- Autismus (1)
- Brett Kavanaugh (1)
- Brock Turner (1)
- Buzzfeed victim impact statement (1)
- Chanel Miller (1)
- Core Field Modeling (1)
- Emotionserkennung (1)
- Evolution Strategies (1)
- Evolutionsstrategien (1)
- Geomagnetism (1)
DeepGeoMap
(2021)
In recent years, deep learning improved the way remote sensing data is processed. The classification of hyperspectral data is no exception. 2D or 3D convolutional neural networks have outperformed classical algorithms on hyperspectral image classification in many cases. However, geological hyperspectral image classification includes several challenges, often including spatially more complex objects than found in other disciplines of hyperspectral imaging that have more spatially similar objects (e.g., as in industrial applications, aerial urban- or farming land cover types). In geological hyperspectral image classification, classical algorithms that focus on the spectral domain still often show higher accuracy, more sensible results, or flexibility due to spatial information independence. In the framework of this thesis, inspired by classical machine learning algorithms that focus on the spectral domain like the binary feature fitting- (BFF) and the EnGeoMap algorithm, the author of this thesis proposes, develops, tests, and discusses a novel, spectrally focused, spatial information independent, deep multi-layer convolutional neural network, named 'DeepGeoMap’, for hyperspectral geological data classification. More specifically, the architecture of DeepGeoMap uses a sequential series of different 1D convolutional neural networks layers and fully connected dense layers and utilizes rectified linear unit and softmax activation, 1D max and 1D global average pooling layers, additional dropout to prevent overfitting, and a categorical cross-entropy loss function with Adam gradient descent optimization. DeepGeoMap was realized using Python 3.7 and the machine and deep learning interface TensorFlow with graphical processing unit (GPU) acceleration. This 1D spectrally focused architecture allows DeepGeoMap models to be trained with hyperspectral laboratory image data of geochemically validated samples (e.g., ground truth samples for aerial or mine face images) and then use this laboratory trained model to classify other or larger scenes, similar to classical algorithms that use a spectral library of validated samples for image classification. The classification capabilities of DeepGeoMap have been tested using two geological hyperspectral image data sets. Both are geochemically validated hyperspectral data sets one based on iron ore and the other based on copper ore samples. The copper ore laboratory data set was used to train a DeepGeoMap model for the classification and analysis of a larger mine face scene within the Republic of Cyprus, where the samples originated from. Additionally, a benchmark satellite-based dataset, the Indian Pines data set, was used for training and testing. The classification accuracy of DeepGeoMap was compared to classical algorithms and other convolutional neural networks. It was shown that DeepGeoMap could achieve higher accuracies and outperform these classical algorithms and other neural networks in the geological hyperspectral image classification test cases. The spectral focus of DeepGeoMap was found to be the most considerable advantage compared to spectral-spatial classifiers like 2D or 3D neural networks. This enables DeepGeoMap models to train data independently of different spatial entities, shapes, and/or resolutions.
From Brock to Brett
(2021)
This master's thesis in US American cultural studies posits that the phenomenon of rape culture represents a socio-cultural system of social power structures and cultural myths. Based on so-called rape myths, this system also constitutes an ideology. The thesis aims to demonstrate how these rape myths are instrumentalized in order to protect (primarily white, cis-male) perpetrators and instead assign responsibility to those affected by sexualized violence. In doing so, the thesis shows that young men like Brock Turner, who benefit from patriarchal power structures, grow up to become men like Brett Kavanaugh, who not only benefit from the fact that rape culture excuses their abusive behavior, but also from the fact that this enables them to reach positions of power through which they, as decision-makers, can in turn maintain the structures underlying rape culture.
The thesis focuses on the rape myths of so-called victim blaming and shaming as well as the victimization of perpetrators. These myths are examined by analyzing 19th-century newspaper articles and then traced into the 21st century. Based on Mary Douglas' theory on ideas of purity, the thesis shows the extent to which not only social categories, namely gender, race, socio-economic status, and age, but also the sexual purity or impurity of those affected have an impact on the societal response to rape cases.
Furthermore, the thesis demonstrates how female bodies function as an ideological battleground for political and social change in the US, and how perceived threats to the patriarchal status quo are framed in public discourse as moral dangers posed by female bodies. The paper argues that rape culture is driven by (white cis) male entitlement to female bodies but moreover to positions of power in the patriarchal system. The thesis shows how this system instrumentalizes rape culture to maintain its underlying structures that favor (cis) men and, in contrast, disadvantage (cis) women and other marginalized and non-heteronormative groups. This is illustrated by analyzing the 2016 Stanford rape case and the 2018 Kavanaugh hearing.
Emotions are a complex concept and they are present in our everyday life. Persons on the autism spectrum are said to have difficulties in social interactions, showing deficits in emotion recognition in comparison to neurotypically developed persons. But social-emotional skills are believed to be positively augmented by training. A new adaptive social cognition training tool “E.V.A.” is introduced which teaches emotion recognition from face, voice and body language. One cross-sectional and one longitudinal study with adult neurotypical and autistic participants were conducted. The aim of the cross-sectional study was to characterize the two groups and see if differences in their social-emotional skills exist. The longitudinal study, on the other hand, aimed for detecting possible training effects following training with the new training tool. In addition, in both studies usability assessments were conducted to investigate the perceived usability of the new tool for neurotypical as well as autistic participants. Differences were found between autistic and neurotypical participants in their social-emotional and emotion recognition abilities. Training effects for neurotypical participants in an emotion recognition task were found after two weeks of home training. Similar perceived usability was found for the neurotypical and autistic participants. The current findings suggest that persons with ASC do not have a general deficit in emotion recognition, but are in need for more time to correctly recognize emotions. In addition, findings suggest that training emotion recognition abilities is possible. Further studies are needed to verify if the training effects found for neurotypical participants also manifest in a larger ASC sample.
Geomagnetic field modeling using spherical harmonics requires the inversion for hundreds to thousands of parameters. This large-scale problem can always be formulated as an optimization problem, where a global minimum of a certain cost function has to be calculated. A variety of approaches is known in order to solve this inverse problem, e.g. derivative-based methods or least-squares methods and their variants. Each of these methods has its own advantages and disadvantages, which affect for example the applicability to non-differentiable functions or the runtime of the corresponding algorithm.
In this work, we pursue the goal to find an algorithm which is faster than the established methods and which is applicable to non-linear problems. Such non-linear problems occur for example when estimating Euler angles or when the more robust L_1 norm is applied. Therefore, we will investigate the usability of stochastic optimization methods from the CMAES family for modeling the geomagnetic field of Earth's core. On one hand, basics of core field modeling and their parameterization are discussed using some examples from the literature. On the other hand, the theoretical background of the stochastic methods are provided. A specific CMAES algorithm was successfully applied in order to invert data of the Swarm satellite mission and to derive the core field model EvoMag. The EvoMag model agrees well with established models and observatory data from Niemegk. Finally, we present some observed difficulties and discuss the results of our model.