54 Informatik
Refine
Document Type
- Doctoral Thesis (3)
- Preprint (3)
- Article (2)
- Master's Thesis (1)
Institute
- Institut für Strukturmechanik (ISM) (3)
- Bauhaus-Institut für zukunftsweisende Infrastruktursysteme (b.is) (1)
- Professur Content Management und Webtechnologien (1)
- Professur Graphische Datenverarbeitung (1)
- Professur Informatik im Bauwesen (1)
- Professur Medieninformatik (1)
- Professur Systeme der Virtuellen Realität (1)
Keywords
- machine learning (3)
- OA-Publikationsfonds2018 (2)
- extreme learning machine (2)
- Abtastung (1)
- Adaptive sampling method (1)
- Amazon Alexa (1)
- Approximation (1)
- Argument (1)
- Assistent (1)
- BTF (1)
Year of publication
- 2018 (9) (remove)
As one of its primary objectives, Computer Graphics aims at the simulation of fabrics’ complex reflection behaviour. Characteristic surface reflectance of fabrics, such as highlights, anisotropy or retro-reflection arise the difficulty of synthesizing. This problem can be solved by using Bidirectional Texture Functions (BTFs), a 2D-texture under various light and view direction. But the acquisition of Bidirectional Texture Functions requires an expensive setup and the measurement process is very time-consuming. Moreover, the size of BTF data can range from hundreds of megabytes to several gigabytes, as a large number of high resolution pictures have to be used in any ideal cases. Furthermore, the three-dimensional textured models rendered through BTF rendering method are subject to various types of distortion during acquisition, synthesis, compression, and processing. An appropriate image quality assessment scheme is a useful tool for evaluating image processing algorithms, especially algorithms designed to leave the image visually unchanged. In this contribution, we present and conduct an investigation aimed at locating a robust threshold for downsampling BTF images without loosing perceptual quality. To this end, an experimental study on how decreasing the texture resolution influences perceived quality of the rendered images has been presented and discussed.
Next, two basic improvements to the use of BTFs for rendering are presented: firstly, the study addresses the cost of BTF acquisition by introducing a flexible low-cost step motor setup for BTF acquisition allowing to generate a high quality BTF database taken at user-defined arbitrary angles. Secondly, the number of acquired textures to the perceptual quality of renderings is adapted so that the database size is not overloaded and can fit better in memory when rendered.
Although visual attention is one of the essential attributes of HVS, it is neglected in most existing quality metrics. In this thesis an appropriate objective quality metric based on extracting visual attention regions from images and adequate investigation of the influence of visual attention on perceived image quality assessment, called Visual Attention Based Image Quality Metric (VABIQM), has been proposed. The novel metric indicates that considering visual saliency can offer significant benefits with regard to constructing objective quality metrics to predict the visible quality differences in images rendered by compressed and non-compressed BTFs and also outperforms straightforward existing image quality metrics at detecting perceivable differences.
This cumulative dissertation discusses - by the example of four subsequent publications - the various layers of a tangible interaction framework, which has been developed in conjunction with an electronic musical instrument with a tabletop tangible user interface. Based on the experiences that have been collected during the design and implementation of that particular musical application, this research mainly concentrates on the definition of a general-purpose abstraction model for the encapsulation of physical interface components that are commonly employed in the context of an interactive surface environment. Along with a detailed description of the underlying abstraction model, this dissertation also describes an actual implementation in the form of a detailed protocol syntax, which constitutes the common element of a distributed architecture for the construction of surface-based tangible user interfaces. The initial implementation of the presented abstraction model within an actual application toolkit is comprised of the TUIO protocol and the related computer-vision based object and multi-touch tracking software reacTIVision, along with its principal application within the Reactable synthesizer. The dissertation concludes with an evaluation and extension of the initial TUIO model, by presenting TUIO2 - a next generation abstraction model designed for a more comprehensive range of tangible interaction platforms and related application scenarios.
Management strategies for sustainable sugarcane production need to deal with the increasing complexity and variability of the whole sugar system. Moreover, they need to accommodate the multiple goals of different industry sectors and the wider community. Traditional disciplinary approaches are unable to provide integrated management solutions, and an approach based on whole systems analysis is essential to bring about beneficial change to industry and the community. The application of this approach to water management, environmental management and cane supply management is outlined, where the literature indicates that the application of extreme learning machine (ELM) has never been explored in this realm. Consequently, the leading objective of the current research was set to filling this gap by applying ELM to launch swift and accurate model for crop production data-driven. The key learning has been the need for innovation both in the technical aspects of system function underpinned by modelling of sugarcane growth. Therefore, the current study is an attempt to establish an integrate model using ELM to predict the concluding growth amount of sugarcane. Prediction results were evaluated and further compared with artificial neural network (ANN) and genetic programming models. Accuracy of the ELM model is calculated using the statistics indicators of Root Means Square Error (RMSE), Pearson Coefficient (r), and Coefficient of Determination (R2) with promising results of 0.8, 0.47, and 0.89, respectively. The results also show better generalization ability in addition to faster learning curve. Thus, proficiency of the ELM for supplementary work on advancement of prediction model for sugarcane growth was approved with promising results.
The production of a desired product needs an effective use of the experimental model. The present study proposes an extreme learning machine (ELM) and a support vector machine (SVM) integrated with the response surface methodology (RSM) to solve the complexity in optimization and prediction of the ethyl ester and methyl ester production process. The novel hybrid models of ELM-RSM and ELM-SVM are further used as a case study to estimate the yield of methyl and ethyl esters through a trans-esterification process from waste cooking oil (WCO) based on American Society for Testing and Materials (ASTM) standards. The results of the prediction phase were also compared with artificial neural networks (ANNs) and adaptive neuro-fuzzy inference system (ANFIS), which were recently developed by the second author of this study. Based on the results, an ELM with a correlation coefficient of 0.9815 and 0.9863 for methyl and ethyl esters, respectively, had a high estimation capability compared with that for SVM, ANNs, and ANFIS. Accordingly, the maximum production yield was obtained in the case of using ELM-RSM of 96.86% for ethyl ester at a temperature of 68.48 °C, a catalyst value of 1.15 wt. %, mixing intensity of 650.07 rpm, and an alcohol to oil molar ratio (A/O) of 5.77; for methyl ester, the production yield was 98.46% at a temperature of 67.62 °C, a catalyst value of 1.1 wt. %, mixing intensity of 709.42 rpm, and an A/O of 6.09. Therefore, ELM-RSM increased the production yield by 3.6% for ethyl ester and 3.1% for methyl ester, compared with those for the experimental data.
In computer-aided design (CAD), industrial products are designed using a virtual 3D model. A CAD model typically consists of curves and surfaces in a parametric representation, in most cases, non-uniform rational B-splines (NURBS). The same representation is also used for the analysis, optimization and presentation of the model. In each phase of this process, different visualizations are required to provide an appropriate user feedback. Designers work with illustrative and realistic renderings, engineers need a
comprehensible visualization of the simulation results, and usability studies or product presentations benefit from using a 3D display. However, the interactive visualization of NURBS models and corresponding physical simulations is a challenging task because of the computational complexity and the limited graphics hardware support.
This thesis proposes four novel rendering approaches that improve the interactive visualization of CAD models and their analysis. The presented algorithms exploit latest graphics hardware capabilities to advance the state-of-the-art in terms of quality, efficiency and performance. In particular, two approaches describe the direct rendering of the parametric representation without precomputed approximations and timeconsuming pre-processing steps. New data structures and algorithms are presented for the efficient partition, classification, tessellation, and rendering of trimmed NURBS surfaces as well as the first direct isosurface ray-casting approach for NURBS-based isogeometric analysis. The other two approaches introduce the versatile concept of programmable order-independent semi-transparency for the illustrative and comprehensible visualization of depth-complex CAD models, and a novel method for the hybrid reprojection of opaque and semi-transparent image information to accelerate stereoscopic rendering. Both approaches are also applicable to standard polygonal geometry which contributes to the computer graphics and virtual reality research communities.
The evaluation is based on real-world NURBS-based models and simulation data. The results show that rendering can be performed directly on the underlying parametric representation with interactive frame rates and subpixel-precise image results. The computational costs of additional visualization effects, such as semi-transparency and stereoscopic rendering, are reduced to maintain interactive frame rates. The benefit of this performance gain was confirmed by quantitative measurements and a pilot user study.
In the field of engineering, surrogate models are commonly used for approximating the behavior of a physical phenomenon in order to reduce the computational costs. Generally, a surrogate model is created based on a set of training data, where a typical method for the statistical design is the Latin hypercube sampling (LHS). Even though a space filling distribution of the training data is reached, the sampling process takes no information on the underlying behavior of the physical phenomenon into account and new data cannot be sampled in the same distribution if the approximation quality is not sufficient. Therefore, in this study we present a novel adaptive sampling method based on a specific surrogate model, the least-squares support vector regresson. The adaptive sampling method generates training data based on the uncertainty in local prognosis capabilities of the surrogate model - areas of higher uncertainty require more sample data. The approach offers a cost efficient calculation due to the properties of the least-squares support vector regression. The opportunities of the adaptive sampling method are proven in comparison with the LHS on different analytical examples. Furthermore, the adaptive sampling method is applied to the calculation of global sensitivity values according to Sobol, where it shows faster convergence than the LHS method. With the applications in this paper it is shown that the presented adaptive sampling method improves the estimation of global sensitivity values, hence reducing the overall computational costs visibly.
In this work, molecular separation of aqueous-organic was simulated by using combined soft computing-mechanistic approaches. The considered separation system was a microporous membrane contactor for separation of benzoic acid from water by contacting with an organic phase containing extractor molecules. Indeed, extractive separation is carried out using membrane technology where complex of solute-organic is formed at the interface. The main focus was to develop a simulation methodology for prediction of concentration distribution of solute (benzoic acid) in the feed side of the membrane system, as the removal efficiency of the system is determined by concentration distribution of the solute in the feed channel. The pattern of Adaptive Neuro-Fuzzy Inference System (ANFIS) was optimized by finding the optimum membership function, learning percentage, and a number of rules. The ANFIS was trained using the extracted data from the CFD simulation of the membrane system. The comparisons between the predicted concentration distribution by ANFIS and CFD data revealed that the optimized ANFIS pattern can be used as a predictive tool for simulation of the process. The R2 of higher than 0.99 was obtained for the optimized ANFIS model. The main privilege of the developed methodology is its very low computational time for simulation of the system and can be used as a rigorous simulation tool for understanding and design of membrane-based systems.
Highlights are, Molecular separation using microporous membranes. Developing hybrid model based on ANFIS-CFD for the separation process, Optimization of ANFIS structure for prediction of separation process
A Hybrid Clustering and Classification Technique for Forecasting Short-Term Energy Consumption
(2018)
Electrical energy distributor companies in Iran have to announce their energy demand at least three 3-day ahead of the market opening. Therefore, an accurate load estimation is highly crucial. This research invoked methodology based on CRISP data mining and used SVM, ANN, and CBA-ANN-SVM (a novel hybrid model of clustering with both widely used ANN and SVM) to predict short-term electrical energy demand of Bandarabbas. In previous studies, researchers introduced few effective parameters with no reasonable error about Bandarabbas power consumption. In this research we tried to recognize all efficient parameters and with the use of CBA-ANN-SVM model, the rate of error has been minimized. After consulting with experts in the field of power consumption and plotting daily power consumption for each week, this research showed that official holidays and weekends have impact on the power consumption. When the weather gets warmer, the consumption of electrical energy increases due to turning on electrical air conditioner. Also, con-sumption patterns in warm and cold months are different. Analyzing power consumption of the same month for different years had shown high similarity in power consumption patterns. Factors with high impact on power consumption were identified and statistical methods were utilized to prove their impacts. Using SVM, ANN and CBA-ANN-SVM, the model was built. Sine the proposed method (CBA-ANN-SVM) has low MAPE 5 1.474 (4 clusters) and MAPE 5 1.297 (3 clusters) in comparison with SVM (MAPE 5 2.015) and ANN (MAPE 5 1.790), this model was selected as the final model. The final model has the benefits from both models and the benefits of clustering. Clustering algorithm with discovering data structure, divides data into several clusters based on similarities and differences between them. Because data inside each cluster are more similar than entire data, modeling in each cluster will present better results. For future research, we suggest using fuzzy methods and genetic algorithm or a hybrid of both to forecast each cluster. It is also possible to use fuzzy methods or genetic algorithms or a hybrid of both without using clustering. It is issued that such models will produce better and more accurate results.
This paper presents a hybrid approach to predict the electric energy usage of weather-sensitive loads. The presented methodutilizes the clustering paradigm along with ANN and SVMapproaches for accurate short-term prediction of electric energyusage, using weather data. Since the methodology beinginvoked in this research is based on CRISP data mining, datapreparation has received a gr eat deal of attention in thisresear ch. Once data pre-processing was done, the underlyingpattern of electric energy consumption was extracted by themeans of machine learning methods to precisely forecast short-term energy consumption. The proposed approach (CBA-ANN-SVM) was applied to real load data and resulting higher accu-racy comparing to the existing models.
2018 American Institute of Chemical Engineers Environ Prog, 2018
https://doi.org/10.1002/ep.12934
The need for finding persuasive arguments can arise in a variety of domains such as politics, finance, marketing or personal entertainment. In these domains, there is a demand to make decisions by oneself or to convince somebody about a specific topic. To obtain a conclusion, one has to search thoroughly different sources in literature and on the web to compare various arguments. Voice interfaces, in form of smartphone applications or smart speakers, present the user with natural conversations in a comfortable way to make search requests in contrast to a traditional search interface with keyboard and display. Benefits and obstacles of such a new interface are analyzed by conducting two studies. The first one consists of a survey for analyzing the target group with questions about situations, motivations, and possible demanding features. The latter one is a wizard-of-oz experiment to investigate possible queries on how a user formulates requests to such a novel system. The results indicate that a search interface with conversational abilities can build a helpful assistant, but to satisfy the demands of a broader audience some additional information retrieval and visualization features need to be implemented.