Publikationen

Die Entwicklung von Softwaresystemen ist eine komplexe Aufgabe. Qualitätssicherung versucht auftretenden Softwarefehler (bugs) in Systemen zu vermeiden, jedoch können Fehler nie ausgeschlossen werden. Sobald ein Softwarefehler entdeckt wird, wird typischerweise ein Fehlerbericht (bug report) erstellt. Dieser dient als Ausgangspunkt für den Entwickler den Fehler im Quellcode der Software zu finden und zu beheben (bug fixing). Fehlerberichte sowie weitere Softwareartefakte, z.B. Anforderungen und der Quellcode selbst, werden in Software Repositories abgelegt. Diese erlauben die Artefakte mit trace links zur Nachvollziehbarkeit (traceability) zu verknüpfen. Oftmals ist die Erstellung der trace links im Entwicklungsprozess vorgeschrieben. Dazu zählen u.a. die Luftfahrt- und Automobilindustrie, sowie die Entwicklung von medizinischen Geräten. Das Auffinden von Softwarefehlern in großen Systemen mit tausenden Artefakten ist eine anspruchsvolle, zeitintensive und fehleranfällige Aufgabe, welche eine umfangreiche Projektkenntnis erfordert. Deswegen wird seit Jahren aktiv an der Automatisierung dieses Prozesses geforscht. Weiterhin wird die manuelle Erstellung und Pflege von trace links als Belastung empfunden und sollte weitgehend automatisiert werden. In dieser Arbeit wird ein neuartiger Algorithmus zum Auffinden von Softwarefehlern vorgestellt, der aktiv die erstellten trace links ausnutzt. Die Artefakte und deren Beziehungen dienen zur Erstellung eines Nachvollziehbarkeitsgraphen, welcher analysiert wird um fehlerhafte Quellcodedateien anhand eines Fehlerberichtes zu finden. Jedoch muss angenommen werden, dass nicht alle notwendigen trace links zwischen den Softwareartefakten eines Projektes erstellt wurden. Deswegen wird ein vollautomatisierter, projektunabhängiger Ansatz vorgestellt, der diese fehlenden trace links erstellt (augmentation). Die Grundlage zur Entwicklung dieses Algorithmus ist der typische Entwicklungsprozess eines Softwareprojektes. Die entwickelten Ansätze wurden mit mehr als 32.000 Fehlerberichten von 27 Open-Source Projekten evaluiert und die Ergebnisse zeigen, dass die Einbeziehung von traceability signifikant das Auffinden von Fehlern im Quellcode verbessert. Weiterhin kann der entwickelte Augmentation Algorithmus zuverlässig fehlende trace links erstellen.

Every moment, countless meteoroids enter our atmosphere unseen. The detection and measurement of meteors offer the unique opportunity to gain insights into the composition of our solar systems’ celestial bodies. Researchers therefore carry out a wide-area-sky-monitoring to secure 360-degree video material, saving every single entry of a meteor. Existing machine intelligence cannot accurately recognize events of meteors intersecting the earth’s atmosphere due to a lack of high-quality training data publicly available. This work presents four reusable open source solutions for researchers trained on data we collected due to the lack of available labelled high-quality training data. We refer to the proposed data set as the NightSkyUCP data set, consisting of a balanced set of 10 000 meteor- and 10 000 non-meteor-events. Our solutions apply various machine-learning techniques, namely classification, feature learning, anomaly detection, and extrapolation. For the classification task, a mean accuracy of 99.1 per cent is achieved. The code and data are made public at figshare with DOI 10.6084/m9.figshare.16451625.

A 360˚ perception of scene geometry is essential for automated driving, notably for parking and urban driving scenarios. Typically, it is achieved using surround-view fisheye cameras, focusing on the near-field area around the vehicle. The majority of current depth estimation approaches focus on employing just a single camera, which cannot be straightforwardly generalized to multiple cameras. The depth estimation model must be tested on a variety of cameras equipped to millions of cars with varying camera geometries. Even within a single car, intrinsics vary due to manufacturing tolerances. Deep learning models are sensitive to these changes, and it is practically infeasible to train and test on each camera variant. As a result, we present novel camera-geometry adaptive multi-scale convolutions which utilize the camera parameters as a conditional input, enabling the model to generalize to previously unseen fisheye cameras. Additionally, we improve the distance estimation by pairwise and patchwise vector-based self-attention encoder networks. We evaluate our approach on the Fisheye WoodScape surround-view dataset, significantly improving over previous approaches. We also show a generalization of our approach across different camera viewing angles and perform extensive experiments to support our contributions. To enable comparison with other approaches, we evaluate the front camera data on the KITTI dataset (pinhole camera images) and achieve state-of-the-art performance among self-supervised monocular methods. An overview video with qualitative results is provided at https://youtu.be/bmX0UcU9wtA. Baseline code and dataset will be made public.

Having a specific understanding of the actual ingredient composition of products helps to calculate additional nutritional information, such as containing fatty and amino acids, minerals and vitamins, as well as to determine its environmental impacts. Unfortunately, producers rarely provide information on how much of each ingredient is in a product. Food manufacturers are, however, required to declare their products in terms of a label comprising an ingredient list (in descending order) and Big7 nutrient values. In this paper, we propose an automated approach for estimating ingredient contents in food products. First, we parse product labels to extract declared ingredients. Next, we exert mathematical formulations on the assumption that the weighted sum of Big7 ingredients as available from food compositional tables should resemble the product’s declared overall Big7 composition. We apply mathematical optimization techniques to find the best fitting ingredient composition estimate. We apply the proposed method to a dataset of 1804 food products spanning 11 product categories. We find that 76% of these products could be analyzed by our approach, and a composition within the prescribed nutrient tolerances could be calculated, using 20% of the allowed tolerances per Big7 ingredient on average. The remaining 24% of the food products could still be estimated when relaxing one or multiple nutrient tolerances. A study with known ingredient compositions shows that estimates are within a 0.9% difference of products’ actual recipes. Hence, the automated approach presented here allows for further analysis of large product quantities and provides possibilities for more intensive nutritional and ecological evaluations of food.

Climate change represents one of the most critical threats to biodiversity with far-reaching consequences for species interactions, the functioning of ecosystems, or the assembly of biotic communities. Plant phenology research has gained increasing attention as the timing of periodic events in plants is strongly affected by seasonal and interannual climate variation. Recent technological development allowed us to gather invaluable data at a variety of spatial and ecological scales. The feasibility of phenological monitoring today and in the future depends heavily on developing tools capable of efficiently analyzing these enormous amounts of data. Deep Neural Networks learn representations from data with impressive accuracy and lead to significant breakthroughs in, e.g., image processing. This article is the first systematic literature review aiming to thoroughly analyze all primary studies on deep learning approaches in plant phenology research. In a multi-stage process, we selected 24 peer-reviewed studies published in the last five years (2016-2021). After carefully analyzing these studies, we describe the applied methods categorized according to the studied phenological stages, vegetation type, spatial scale, data acquisition- and deep learning methods. Furthermore, we identify and discuss research trends and highlight promising future directions. We present a systematic overview of previously applied methods on different tasks that can guide this emerging complex research field.

Detailed knowledge of frequently recurring code changes can be beneficial for a variety of software engineering activities. For example, it is a key step to understand the process of software evolution, but is also necessary when developing more sophisticated code completion features predicting likely changes. Previous attempts on automatically finding such code change patterns were mainly based on frequent itemset mining, which essentially finds sets of edits occurring in close proximity. However, these approaches do not analyze the interplay among code elements, e.g., two code objects being named similarly, and thereby neglect great potential in identifying a number of meaningful patterns. We present a novel method for the automated mining of code change patterns from Git repositories that captures these context relations between individual edits. Our approach relies on a transformation of source code into a graph representation, while keeping relevant relations present. We then apply graph mining techniques to extract frequent subgraphs, which can be used for further analysis of development projects. We suggest multiple usage scenarios for the resulting pattern type. Additionally, we propose a transformation into complex event processing (CEP) rules which allows for easier application, especially for event-based auto-completion recommenders or similar tools. For evaluation, we mined seven open-source code repositories. We present 25 frequent change patterns occurring across these projects. We found these patterns to be meaningful, easy to interpret and mostly persistent across project borders. On average, a pattern from our set appeared in 45 percent of the analyzed code changes.

Collaborative training of neural networks leverages distributed data by exchanging gradient information between different clients. Although training data entirely resides with the clients, recent work shows that training data can be reconstructed from such exchanged gradient information. To enhance privacy, gradient perturbation techniques have been proposed. However, they come at the cost of reduced model performance, increased convergence time, or increased data demand. In this paper, we introduce PRECODE, a PRivacy EnhanCing mODulE that can be used as generic extension for arbitrary model architectures. We propose a simple yet effective realization of PRECODE using variational modeling. The stochastic sampling induced by variational modeling effectively prevents privacy leakage from gradients and in turn preserves privacy of data owners. We evaluate PRECODE using state of the art gradient inversion attacks on two different model architectures trained on three datasets. In contrast to commonly used defense mechanisms, we find that our proposed modification consistently reduces the attack success rate to 0% while having almost no negative impact on model training and final performance. As a result, PRECODE reveals a promising path towards privacy enhancing model extensions.

Traceability recovery captures trace links among different software artifacts (e.g., requirements and code) when two artifacts cover the same part of system functionalities. These trace links provide important support for developers in software maintenance and evolution tasks. Information Retrieval (IR) is now the mainstream technique for semi-automatic approaches to recover candidate trace links based on textual similarities among artifacts. The performance of IR-based traceability recovery is evaluated by the ranking of relevant traces in the generated lists of candidate links. Unfortunately, this performance is greatly hindered by the vocabulary mismatch problem between different software artifacts. To address this issue, a growing body of enhancing strategies based on user feedback is proposed to adjust the calculated IR values of candidate links after the user verifies part of these links. However, the improvement brought by this kind of strategies requires a large amount of user feedback, which could be infeasible in practice. In this paper, we propose to improve IR-based traceability recovery by propagating a small amount of user feedback through the closeness analysis on call and data dependencies in the code. Specifically, our approach first iteratively asks users to verify a small set of candidate links. The collected frugal feedback is then composed with the quantified functional similarity for each code dependency (called closeness) and the generated IR values to improve the ranking of unverified links. An empirical evaluation based on nine real-world systems with three mainstream IR models shows that our approach can outperform five baseline approaches by using only a small amount of user feedback.

Nature has always inspired the human spirit and scientists frequently developed new methods based on observations from nature. Recent advances in imaging and sensing technology allow fascinating insights into biological neural processes. With the objective of finding new strategies to enhance the learning capabilities of neural networks, we focus on a phenomenon that is closely related to learning tasks and neural stability in biological neural networks, called homeostatic plasticity. Among the theories that have been developed to describe homeostatic plasticity, synaptic scaling has been found to be the most mature and applicable. We systematically discuss previous studies on the synaptic scaling theory and how they could be applied to artificial neural networks. Therefore, we utilize information theory to analytically evaluate how mutual information is affected by synaptic scaling. Based on these analytic findings, we propose two flavors in which synaptic scaling can be applied in the training process of simple and complex, feedforward, and recurrent neural networks. We compare our approach with state-of-the-art regularization techniques on standard benchmarks. We found that the proposed method yields the lowest error in both regression and classification tasks compared to previous regularization approaches in our experiments across a wide range of network feedforward and recurrent topologies and data sets.

FisheyeDistanceNet [1] proposed a self-supervised monocular depth estimation method for fisheye cameras with a large field of view (> 180˚). To achieve scale-invariant depth estimation, FisheyeDistanceNet supervises depth map predictions over multiple scales during training. To overcome this bottleneck, we incorporate self-attention layers and robust loss function [2] to FisheyeDistanceNet. A general adaptive robust loss function helps obtain sharp depth maps without a need to train over multiple scales and allows us to learn hyperparameters in loss function to aid in better optimization in terms of convergence speed and accuracy. We also ablate the importance of Instance Normalization over Batch Normalization in the network architecture. Finally, we generalize the network to be invariant to camera views by training multiple perspectives using front, rear, and side cameras. Proposed algorithm improvements, FisheyeDistanceNet++, result in 30% relative improvement in RMSE while reducing the training time by 25% on the WoodScape dataset. We also obtain state-of-the-art results on the KITTI dataset, in comparison to other self-supervised monocular methods.

3D object detection based on LiDAR point clouds is a crucial module in autonomous driving particularly for long range sensing. Most of the research is focused on achieving higher accuracy and these models are not optimized for deployment on embedded systems from the perspective of latency and power efficiency. For high speed driving scenarios, latency is a crucial parameter as it provides more time to react to dangerous situations. Typically a voxel or point-cloud based 3D convolution approach is utilized for this module. Firstly, they are inefficient on embedded platforms as they are not suitable for efficient parallelization. Secondly, they have a variable runtime due to level of sparsity of the scene which is against the determinism needed in a safety system. In this work, we aim to develop a very low latency algorithm with fixed runtime. We propose a novel semantic segmentation architecture as a single unified model for object center detection using key points, box predictions and orientation prediction using binned classification in a simpler Bird's Eye View (BEV) 2D representation. The proposed architecture can be trivially extended to include semantic segmentation classes like road without any additional computation. The proposed model has a latency of 4 ms on the embedded Nvidia Xavier platform. The model is 5X faster than other top accuracy models with a minimal accuracy degradation of 2% in Average Precision at IoU = 0.5 on KITTI dataset.

Deep learning algorithms classify plant species with high accuracy, and smartphone applications leverage this technology to enable users to identify plant species in the field. The question we address here is whether such crowd-sourced data contain substantial macroecological information. In particular, we aim to understand if we can detect known environmental gradients shaping plant co-occurrences. In this study we analysed 1 million data points collected through the use of the mobile app Flora Incognita between 2018 and 2019 in Germany and compared them with Florkart, containing plant occurrence data collected by more than 5000 floristic experts over a 70-year period. The direct comparison of the two data sets reveals that the crowd-sourced data particularly undersample areas of low population density. However, using nonlinear dimensionality reduction we were able to uncover macroecological patterns in both data sets that correspond well to each other. Mean annual temperature, temperature seasonality and wind dynamics as well as soil water content and soil texture represent the most important gradients shaping species composition in both data collections. Our analysis describes one way of how automated species identification could soon enable near real-time monitoring of macroecological patterns and their changes, but also discusses biases that must be carefully considered before crowd-sourced biodiversity data can effectively guide conservation measures.

Assisting and automating software engineering tasks is a state-of-the-art way to support stakeholders of development projects. A common assistance function of IDEs is the auto-completion of source code. Assistance functions, such as auto-completion, are almost entirely missing in modeling tools though auto-completion in general gains continuously more importance in software development. We analyze a user’s performed editing operations in order to anticipate modeling activities and to recommend appropriate auto-completions for them. Editing operations are captured as events and modeling activities are defined as complex event patterns, facilitating the matching by complex-event-processing. The approach provides adapted auto-completions reactively upon each editing operation of the user. We implemented the RapMOD prototype as add-in for the modeling tool Sparx Enterprise Architect™ . A controlled user experiment with 37 participants performing modeling tasks demonstrated the approach's potential to reduce modeling effort significantly. Users having auto-completions available for a modeling scenario performed the task 27 percent faster, needed to perform 56 percent less actions, and perceived the task 29 percent less difficult.

Defocus particle tracking (DPT) has gained increasing importance for its use to determine particle trajectories in all three dimensions with a single-camera system, as typical for a standard microscope, the workhorse of todays ongoing biomedical revolution. DPT methods derive the depth coordinates of particle images from the different defocusing patterns that they show when observed in a volume much larger than the respective depth of field. Therefore it has become common for state-of-the-art methods to apply image recognition techniques. Two of the most commonly and widely used DPT approaches are the application of (astigmatism) particle image model functions (MF methods) and the normalized cross-correlations between measured particle images and reference templates (CC methods). Though still young in the field, the use of neural networks (NN methods) is expected to play a significant role in future and more complex defocus tracking applications. To assess the different strengths of such defocus tracking approaches, we present in this work a general and objective assessment of their performances when applied to synthetic and experimental images of different degrees of astigmatism, noise levels, and particle image overlapping. We show that MF methods work very well in low-concentration cases, while CC methods are more robust and provide better performance in cases of larger particle concentration and thus stronger particle image overlap. The tested NN methods generally showed the lowest performance, however, in comparison to the MF and CC methods, they are yet in an early stage and have still great potential to develop within the field of DPT.

Surround View fisheye cameras are commonly deployed in automated driving for 360˚ near-field sensing around the vehicle. This work presents a multi-task visual perception network on unrectified fisheye images to enable the vehicle to sense its surrounding environment. It consists of six primary tasks necessary for an autonomous driving system: depth estimation, visual odometry, semantic segmentation, motion segmentation, object detection, and lens soiling detection. We demonstrate that the jointly trained model performs better than the respective single task versions. Our multi-task model has a shared encoder providing a significant computational advantage and has synergized decoders where tasks support each other. We propose a novel camera geometry based adaptation mechanism to encode the fisheye distortion model both at training and inference. This was crucial to enable training on the WoodScape dataset, comprised of data from different parts of the world collected by 12 different cameras mounted on three different cars with different intrinsics and viewpoints. Given that bounding boxes is not a good representation for distorted fisheye images, we also extend object detection to use a polygon with non-uniformly sampled vertices. We additionally evaluate our model on standard automotive datasets, namely KITTI and Cityscapes. We obtain the state-of-the-art results on KITTI for depth estimation and pose estimation tasks and competitive performance on the other tasks. We perform extensive ablation studies on various architecture choices and task weighting methodologies. A short video at https://youtu.be/xbSjZ5OfPes provides qualitative results.

Pollen identification and quantification are crucial but challenging tasks in addressing a variety of evolutionary and ecological questions (pollination, paleobotany), but also for other fields of research (e.g. allergology, honey analysis or forensics). Researchers are exploring alternative methods to automate these tasks but, for several reasons, manual microscopy is still the gold standard. In this study, we present a new method for pollen analysis using multispectral imaging flow cytometry in combination with deep learning. We demonstrate that our method allows fast measurement while delivering high accuracy pollen identification. A dataset of 426 876 images depicting pollen from 35 plant species was used to train a convolutional neural network classifier. We found the best-performing classifier to yield a species-averaged accuracy of 96%. Even species that are difficult to differentiate using microscopy could be clearly separated. Our approach also allows a detailed determination of morphological pollen traits, such as size, symmetry or structure. Our phylogenetic analyses suggest phylogenetic conservatism in some of these traits. Given a comprehensive pollen reference database, we provide a powerful tool to be used in any pollen study with a need for rapid and accurate species identification, pollen grain quantification and trait extraction of recent pollen.

Many applications in chemistry, biology and medicine use microfluidic devices to separate, detect and analyze samples on a miniaturized size-level. Fluid flows evolving in channels of only several tens to hundreds of micrometers in size are often of a 3D nature, affecting the tailored transport of cells and particles. To analyze flow phenomena and local distributions of particles within those channels, astigmatic particle tracking velocimetry (APTV) has become a valuable tool, on condition that basic requirements like low optical aberrations and particles with a very narrow size distribution are fulfilled. Making use of the progress made in the field of machine vision, deep neural networks may help to overcome these limiting requirements, opening new fields of applications for APTV and allowing them to be used by nonexpert users. To qualify the use of a cascaded deep convolutional neural network (CNN) for particle detection and position regression, a detailed investigation was carried out starting from artificial particle images with known ground truth to real flow measurements inside a microchannel, using particles with uni- and bimodal size distributions. In the case of monodisperse particles, the mean absolute error and standard deviation of particle depth-position of less than and about 1 [my]m were determined, employing the deep neural network and the classical evaluation method based on the minimum Euclidean distance approach. While these values apply to all particle size distributions using the neural network, they continuously increase towards the margins of the measurement volume of about one order of magnitude for the classical method, if nonmonodisperse particles are used. Nevertheless, limiting the depth of measurement volume in between the two focal points of APTV, reliable flow measurements with low uncertainty are also possible with the classical evaluation method and polydisperse tracer particles. The results of the flow measurements presented herein confirm this finding. The source code of the deep neural network used here is available on https://github.com/SECSY-Group/DNN-APTV.

Die Nachvollziehbarkeit von Anforderungen ist wichtiges Qualitätsmerkmal der Softwareentwicklung. Für eine Vielzahl von Softwareentwicklungsaktivitäten ist die Nachvollziehbarkeit von Anforderungen eine notwenige Voraussetzung. Dazu gehören unter anderem die Analyse funktionaler Sicherheit, die Einflussanalyse, die Analyse des Abdeckungsgrades oder die Compliance. Für die Entwicklung sicherheitskritischer Softwaresysteme ist dieses Qualitätsmerkmal von besonderer Bedeutung. Daher wird dieses von entsprechenden Richtlinien zur Entwicklung sicherheitskritischer Software explizit vorgeschrieben. Obwohl die Relevanz der Nachvollziehbarkeit in Softwareprojekten allgemein bekannt ist, findet nur in wenigen Fällen eine systematische Planung zur Erreichung dieses Qualitätsmerkmals Anwendung. Häufig wird Nachvollziehbarkeit erst nachträglich umgesetzt. Daraus resultieren oft unvollständige Implementierungen der Nachvollziehbarkeit, die trotzdem als Grundlage für schwerwiegende Entscheidungen herangezogen werden. Aus diesem Grunde sollten die entsprechenden Implementierungen einer eingehenden Prüfung unterzogen werden, besonders im Rahmen der Entwicklung sicherheitskritischer Systeme. Dazu ist jedoch eine Vielzahl von Herausforderungen zu meistern. Zum einen hängt die Nachvollziehbarkeit von den projektspezifischen Zielen ab. Bei sicherheitskritischen Systemen müssen oft Vorgaben aus Richtlinien erfüllt werden. Auch die Nutzung der Nachvollziehbarkeit ist sehr stark von den jeweiligen Zielen abhängig. In dieser Arbeit wird ein Ansatz zur systematischen Prüfung von Softwareprojekten im Hinblick auf deren Nachvollziehbarkeit der Anforderungen vorgeschlagen. Eine notwendige Voraussetzung für den Prüfansatz ist die präzise Planung und Definition der Nachvollziehbarkeit von Anforderungen in einem Softwareprojekt. Daher wird im Rahmen dieser Arbeit ein entsprechender Planungsansatz präsentiert. Weiterhin wird ein analytisches Modell zur systematischen Prüfung der Nachvollziehbarkeit in Softwareprojekten präsentiert. Dieses Modell umfasst eine vollständige Klassifikation möglicher Fehlertypen. Außerdem werden Kriterien zur systematischen Erkennung dieser Fehler vorgeschlagen. Die Ergebnisse einer Expertenbefragung bestätigen die Vollständigkeit des analytischen Prüfmodells. Zudem wurde der vorgeschlagene Ansatz zur systematischen Prüfung der Nachvollziehbarkeit von Anforderungen in zwei Studien evaluiert. Dabei konnte der Nutzen des Ansatzes für die Entwicklung von sicherheitskritischer und nicht sicherheitskritischer Software nachgewiesen werden.