05 Fakultät Informatik, Elektrotechnik und Informationstechnik

Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/6

Browse

Subject: search.filters.subject.020

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    A method for describing, analyzing, and exploring visual art collections
    (2024) Pflüger, Hermann
    Visual art collections are more than just sets of images. Behind a collection or an exhibition are the curator’s concept and the specific characteristics of the collection. A controlled vocabulary that takes into account the concept and the specific characteristics of a collection requires individual terms/concepts, and specific relationships between the concepts of the vocabulary. Furthermore, the granularity of a controlled vocabulary should be adapted to a particular concept of the collection, i.e., for some areas, the number of terms/concepts should be very high and specific; for other areas, concepts should be summarized to keep the complexity of the collection low and thus facilitate the handling and presentation of the collection. A LadeCA vocabulary is an extended controlled vocabulary that is particularly suitable for collections of works of art. LadeCA enables a user to describe, analyze, and explore art collections. LadeCA is also suitable for comparing the contents of different collections or different parts of a collection. The general concepts of a controlled vocabulary/thesaurus, their terms, and defined semantic relations between them can be automatically adopted in a LadeCA vocabulary so that the vocabulary only needs to be adapted for the current collection. The process of creating a LadeCA vocabulary for art collections is supported by three interactive interfaces that ensure that the intention behind a collection is taken into account, and that the creation effort is kept low. This article explains the process of creating LadeCA vocabularies for art collections and describes the initial experiences and results.
  • Thumbnail Image
    ItemOpen Access
    Examining a hybrid pitting detection approach on real data
    (2023) Schmid, Tobias
    We propose a hybrid detection approach for pitting detection in gears. Most research on pitting detection with machine learning is done on supervised data and often on simulated pitting. However, for pitting detection in practice an unsupervised approach is required. The main idea behind the proposed solution is the ability to leverage elements of supervised machine learning models in pitting detection, while operating on unlabeled data. The training data used for this algorithm is taken from gear boxes with actual pitting failure and without prior knowledge about pitting size during different stages of operation, providing a realistic case for an operational scenario. The approach can be seen as two parts. At first we try to detect changes in the underlying structure of the vibration data using fast fourier transform to obtain frequency spectra that are fed to a sparse autoencoder. The encoded reduced feature space is the clustered to look for separability by time. In case there are significant changes in the underlying structure, an Long Short Term Memory model is trained to see if the changes can be validated as actual pitting damage. The LSTM can then further be used to adapt fast to the pitting damage with the goal of prolonging the lifespan of the gear.