Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung IOSB

Filters
Reset filters

Settings
Now showing 1 - 10 of 182
No Thumbnail Available
Publication

Regression via causally informed Neural Networks

2024 , Youssef, Shahenda , Doehner, Frank , Beyerer, Jürgen

Neural Networks have been successful in solving complex problems across various fields. However, they require significant data to learn effectively, and their decision-making process is often not transparent. To overcome these limitations, causal prior knowledge can be incorporated into neural network models. This knowledge improves the learning process and enhances the robustness and generalizability of the models. We propose a novel framework RCINN that involves calculating the inverse probability of treatment weights given a causal graph model alongside the training dataset. These weights are then concatenated as additional features in the neural network model. Then incorporating the estimated conditional average treatment effect as a regularization term to the model loss function, the potential influence of confounding variables can be mitigated, leading to bias minimization and improving the neural network model. Experiments conducted on synthetic and benchmark datasets using the framework show promising results.

View
No Thumbnail Available
Publication

Domänenadaptation für feingranulare Fahrzeugklassifikation mittels Domain-Adversarial-Learning

2024 , Wolf, Stefan , Beyerer, Jürgen

View
No Thumbnail Available
Publication

Simulation study and experimental validation of a neural network-based predictive tracking system for sensor-based sorting

2023 , Maier, Georg , Reith-Braun, Marcel , Bauer, Albert , Gruna, Robin , Pfaff, Florian , Kruggel-Emden, Harald , Längle, Thomas , Hanebeck, Uwe D. , Beyerer, Jürgen

Sensor-based sorting offers cutting-edge solutions for separating granular materials. The line-scanning sensors currently in use in such systems only produce a single observation of each object and no data on its movement. According to recent studies, using an area-scan camera has the potential to reduce both characterization and separation error in a sorting process. A predictive tracking approach based on Kalman filters makes it possible to estimate the followed paths and parametrize a unique motion model for each object using a multiobject tracking system. While earlier studies concentrated on physically-motivated motion models, it has been demonstrated that novel machine learning techniques produce predictions that are more accurate. In this paper, we describe the creation of a predictive tracking system based on neural networks. The new algorithm is applied to an experimental sorting system and to a numerical model of the sorter. Although the new approach does not yet fully reach the achieved sorting quality of the existing approaches, it allows the use of the general method without requiring expert knowledge or a fundamental understanding of the parameterization of the particle motion model.

View
No Thumbnail Available
Publication

Optimizing Fine-Grained Fungi Classification for Diverse Application-Oriented Open-Set Metrics

2023 , Wolf, Stefan , Beyerer, Jürgen

Fine-grained fungi species classification is an important task to support distinguishing edible and poisonous fungi and thus, reducing the risk of accidental poisoning. Therefore, the FungiCLEF 2023 challenge seeks to find the best solution for this task considering multiple metrics with each having a different application in focus like e.g., a low confusion of edible and poisonous fungi. We propose a method to approach the different metrics by exploiting modern deep learning networks, strong data augmentation and class-balanced training. The challenge assumes an open-set scenario which includes unknown classes during evaluation which we identify by a confidence thresholding approach. With our method, we achieved the 2nd place in the challenge with good scores across all metrics. Code is available at: https://github.com/wolfstefan/fungi2023.

View
No Thumbnail Available
Publication

Metrics for the evaluation of learned causal graphs based on ground truth

2024 , Rehak, Josephine , Falkenstein, Alexander , Doehner, Frank , Beyerer, Jürgen

The self-guided learning of causal relations may contribute to the general maturity of artificial intelligence in the future. To develop such learning algorithms, powerful metrics are required to track advances. In contrast to learning algorithms, little has been done in regards to developing suitable metrics. In this work, we evaluate current state of the art metrics by inspecting their discovery properties and their considered graphs. We also introduce a new combination of graph notation and metric, which allows for benchmarking given a variety of learned causal graphs. It also allows the use of maximal ancestral graphs as ground truth.

View
No Thumbnail Available
Publication

Sensitivity enhanced glucose sensing by return-path Mueller matrix ellipsometry

2023 , Chen, Chia Wei , Hartrumpf, Matthias , Längle, Thomas , Beyerer, Jürgen

Diabetes is a worldwide public health problem. According to the survey of the Robert Koch Institute, in Germany, at least 7.2 percent population (aged between 18 to 79 years) have diabetes. Therefore, the demand for glucose monitoring is increasing, especially for non-invasive glucose monitoring technology. In this work, we proposed a novel method to enhance the sensitivity of glucose monitoring by return-path ellipsometry with a quarter-wave plate and mirror. The coaxial design improves the sensitivity and reduces the complexity of optical system alignment by means of a fixed quarter-wave plate. The proposed system showed higher sensitivity compared to the transmission configuration.

View
No Thumbnail Available
Publication

Usability for Data Sovereignty - Evaluation of Privacy Risk Quantification Interfaces

2023 , Appenzeller, Arno , Balduf, Falk , Beyerer, Jürgen

Digital medical data is becoming widely available through the ongoing digitization efforts in the medical sector. This also leads to more personal health data available for secondary usage like medical research. A common way to collect medical data in a privacy compliant way is through the informed consent of the affected person. While consent forms are typically paper-based, in the last years the concept of digital consent is becoming more and more common. Still, such consent forms can be very complex and overwhelming for the patient. For example, it can be hard to estimate the personal privacy impact when sharing data of rare medical conditions and user interfaces for data sharing need to be designed carefully. Privacy Risk Quantifications (PRQ) and Dynamic Consent (DC) are two tools to help patients to make a consent decision and choose the data for sharing. However, interfaces for those two technologies are not trivial to design. This paper develops two new interface variants with a focus on design guidelines and best practices for usability. To evaluate the new designs, a user study was conducted which shows improved usability in comparison to an existing interface. The Base variant is a prototype that was previously developed as a technical demonstrator for DC and PRQ without requirements for usability. The three interfaces are evaluated in a user study, which shows that the usability focus of the newer variants leads to a better rating by the test subjects compared to the basic prototype. It can also be seen that interfaces with more detailed explanations and a focus on visual comprehensive and compelling interfaces get a better usability score compared to the pure technical versions. Finally, it can be seen that interfaces for those technologies are ideally developed in an iterative design development cycle.

View
No Thumbnail Available
Publication

Root cause analysis using anomaly detection and temporal informed causal graphs

2024 , Rehak, Josephine , Youssef, Shahenda , Beyerer, Jürgen

In industrial processes, anomalies in the production equipment may lead to expensive failures. To avoid and avert such failures, the identification of the right root cause is crucial. Ideally, the search for a root cause is backed by causal information such as causal graphs. We have extended a framework that fuses causal graphs with anomaly detection to infer likely root causes. In this work, we add the use of temporal information to draw temporal valid conclusions about the potential propagation of anomalous information in causal graphs. The use of the framework is demonstrated on a robotic gripping process.

View
No Thumbnail Available
Publication

SWaTEval: An Evaluation Framework for Stateful Web Application Testing

2023 , Borcherding, Anne , Penkov, Nikolay , Giraud, Mark Leon , Beyerer, Jürgen

Web applications are an easily accessible and valuable target for attackers. Therefore, web applications need to be examined for vulnerabilities. Modern web applications usually behave in a stateful manner and hence have an underlying state machine that determines their behavior based on the current state. To thoroughly test a web application, it is necessary to consider all aspects of a web application, including its internal states. In a blackbox setting, which we presuppose for this work, however, the internal state machine must be inferred before it can be used for testing. For state machine inference it is necessary to choose a similarity measure for web pages. Some approaches for automated blackbox stateful testing for web applications have already been proposed. It is, however, unclear how these approaches perform in comparison. We therefore present our evaluation framework for stateful web application testing, SWaTEval. In our evaluation, we show that SWaTEval is able to repr oduce evaluation results from literature, demonstrating that SWaTEval is suitable for conducting meaningful evaluations. Further, we use SWaTEval to evaluate various approaches to similarity measures for web pages, including a new method based on the euclidean distance that we propose in this paper. These similarity measures are an important part of the automated state machine inference necessary for stateful blackbox testing. We show that the choice of similarity measure has an impact on the performance of the state machine inference regarding the number of correctly identified states, and that our newly proposed similarity measure leads to the highest number of correctly identified states.

View
No Thumbnail Available
Publication

Detecting Tar Contaminated Samples in Road-rubble using Hyperspectral Imaging and Texture Analysis

2023 , Bäcker, Paul , Maier, Georg , Gruna, Robin , Längle, Thomas , Beyerer, Jürgen

Polycyclic aromatic hydrocarbons (PAH) containing tar-mixtures pose a challenge for recycling road rubble, as the tar containing elements have to be extracted and decontaminated for recycling. In this preliminary study, tar, bitumen and minerals are discriminated using a combination of color (RGB) and Hyperspectral Short Wave Infrared (SWIR) cameras. Further, the use of an autoencoder for detecting minerals embedded inside tar- and bitumen mixtures is proposed. Features are extracted from the spectra of the SWIR camera and the texture of the RGB images. For classification, linear discriminant analysis combined with a k-nearest neighbor classification is used. First results show a reliable detection of minerals and positive signs for separability of tar and bitumen. This work is a foundation for developing a sensor-based sorting system for physical separation of tar contaminated samples in road rubble.

View