Digital Library of Gesellschaft für Informatik e.V.
Not a member yet
40758 research outputs found
Sort by
GANDiffKids: Benchmarking Face Recognition for Children on Synthetic Data
No full textFace recognition systems are primarily tailored to adult faces, and numerous studies have shown that their performance significantly declines when applied to children. The development of child-specific recognition models is further constrained by the scarcity of publicly available datasets and the ethical and privacy concerns surrounding the collection of children's biometric data. To overcome these limitations, we propose GANDiffKids, a novel framework for synthesizing realistic child face images with diverse intra-class variations. Our approach combines a generative adversarial network (GAN) with a diffusion model - a strategy that has proven effective for adult face synthesis in prior work. We demonstrate that GANDiffKids can generate a large-scale dataset of synthetic child faces across a range of image qualities. The resulting dataset is benchmarked using a state-of-the-art face recognition model, establishing its potential as a valuable resource for advancing child-specific face recognition research
Single-Instance Multi-Sample Fusion in Deep Fingerprint Fuzzy Vault
No full textThis paper investigates the impact of multi-sample fusion at enrolment on the performance and security of biometric cryptosystems, specifically focusing on the fuzzy vault scheme. Biometric template protection is critical for safeguarding sensitive biometric data, and we propose enhancing robustness by averaging multiple biometric samples from the same instance of a modality during the enrolment phase. This approach aims to reduce intra-class variability and stabilize feature representations, thereby improving recognition accuracy without significantly impacting usability. We evaluate a fuzzy vault system based on deep learning-based fingerprint embeddings, using a comprehensive fingerprint dataset (MCYT330) with varying sample qualities. Our experimental results demonstrate that fusing multiple enrolment samples significantly shifts mated score distributions, leading to notable improvements in recognition rates and security in both unprotected and fuzzy vault systems.
This work confirms the practical benefits of multi-sample fusion in developing more reliable and secure privacy-preserving biometric systems
Innovative Verknüpfung von freien Transportkapazitäten im ländlichen ÖPNV
No full textDer öffentliche Personennahverkehr (ÖPNV) ist – da kommunale Pflichtaufgabe – in vielen ländlichen Regionen überwiegend auf den Schülerverkehr ausgerichtet und entsprechend finanziert. Bedienung in Tagesrandzeiten weisen eher eine Schwachlast aus. Darüber hinaus ergibt sich während der Ferienzeit und am Wochenende sowohl für viele Pendler:innen als auch für Fahrgäste in der Region eine Lücke in der umweltfreundlichen Mobilitätskette. Der ÖPNV ist folglich gegenüber dem eigenen Pkw nicht konkurrenzfähig, zumal das eigene Auto weiterhin als preiswertes und attraktives Verkehrsmittel mit einem hohen Maß an Komfort, Autonomie und Geschwindigkeit gilt. Darüber hinaus spiegeln sich auch niedrige Bevölkerungsdichten in rückläufigen Passagierzahlen wider. Im Bemühen um wirtschaftliche Effizienz wird die Haltestellendichte verringert und die Abfahrtszeiten loser getaktet, Maßnahmen, die das ÖPNV-System noch mehr schwächen als fördern. Verschiedene innovative Ansätze werden von Seiten der Wissenschaft und Praxis erprobt, um den Zugang zu Mobilität im ländlichen Raum zu stärken und zukunftsfähig zu gestalten. Dieser Artikel stellt den Ansatz des kombinierten Verkehrs vor, der im Zuge des mFUND-Forschungsprojekts CargoSurfer in Theorie und Praxis in Reallaboren in Hessen, Deutschland untersucht wurde. Die Idee ist es, die Logistiklücke im ländlichen Raum zu schließen und gleichzeitig den ÖPNV zu stärken. Hierfür wurde eine Vermittlungs- und Buchungsplattform entwickelt, auf der freie Ladeflächenkapazitäten mit Transportanfragen von Warenversender:innen zusammengeführt werden. Durch diese zusätzlichen Frachtmitnahmen können bereits geplante Fahrten des Personen- und Güterverkehrs effektiver ausgelastet werden und regionale Betreiber zusätzliche Deckungsbeiträge generieren. In many rural regions, local public transport—as a mandatory municipal task—is primarily geared towards school transport and financed accordingly. Services during off-peak times tend to be underutilized. Consequently, there is a gap in the environmentally friendly mobility chain for many commuters and passengers in the region during the vacation season and at weekends. As a result, public transport is not competitive with the private car, especially as the private car mode is still considered the most attractive means of transportation, offering a high degree of comfort, autonomy and speed. In addition, low population densities are also reflected in declining passenger numbers. In an effort to achieve economic viability, the density of stops is being reduced and departure times become less frequent in rural regions, measures that weaken the public transport system even more than they promote it. Various innovative approaches are being tested in the course of studies and by practitioners aiming to strengthen access to a sustainable mobility service in rural areas. This article presents the combined transport approach, which was investigated in theory and practice in real laboratories as part of the mFUND-research project CargoSurfer in Hesse, Germany. The idea is to close the logistics gap in rural areas and to strengthen local public transport at the same time. To this end, a brokerage and booking platform was developed on which free public and private loading space capacities are brought together with transport requests from customer looking to ship freight. These additional pick-ups mean that already planned passenger and freight transport trips can be utilized more effectively and regional operators can generate additional contribution margins
Integration neuer Mobilitätsdienste in klassische Verkehrsbetriebe
No full textNeue Mobilitätsdienste (NMS) wie Carsharing, Citybikes und Elektroroller werden in Großstädten immer beliebter. NMS haben bereits begonnen, die Art und Weise zu beeinflussen, wie sich die Menschen in städtischen Gebieten fortbewegen. Es wird erwartet, dass sie auch Auswirkungen auf die Betreiber des öffentlichen Verkehrs und insbesondere auf ihre IT-Landschaft haben werden. In diesem Zusammenhang sind NMS-Teil des bedarfsgerechten öffentlichen Verkehrs. In diesem Beitrag wird das Thema aus der Perspektive der Unternehmensarchitektur (EA) betrachtet. Obwohl die Bedeutung des EA-Managements für ÖPNV-Organisationen anerkannt ist, gibt es nur wenige Untersuchungen darüber, wie NMS die EA verändern. Das Ziel unserer Forschung ist es, einen Beitrag zu diesem Bereich zu leisten, indem eine mehrstufig evaluierten Referenzarchitektur für traditionelle angebotsorientierte ÖPNV-Unternehmen vorgestellt wird, die New Mobility Services (NMS) entweder als nachfrage- oder bedarfsorientierte Verkehrslösung in ihr Geschäft integrieren wollen. New mobility services (NMS) such as car sharing, city bikes and electric scooters are becoming increasingly popular in large cities. NMS have already started to influence the way people get around in urban areas. They are also expected to have an impact on public transport operators and especially on their IT landscape. In this context, NMS are part of demand responsive public transport. In this paper, the topic is approached from an enterprise architecture (EA) perspective. Although the importance of EA management for public transport organizations is recognized, there is little research on how NMSs change the EA. The aim of our research is to contribute to this area by presenting a multi-level evaluated reference architecture for traditional supply-oriented public transport organizations that want to integrate New Mobility Services (NMS) into their business either as a demand-responsive or demand-responsive transport solution
A Portfolio Management Method for Process Mining-Enabled Business Process Improvement Projects
No full textProcess mining has received tremendous attention from research and industry, establishing itself as a highly sought-after technology. Despite the technological maturity of process mining solutions, which has been achieved through extensive investments in research and development, organizations still face the challenge of elusive value when systematically adopting process mining. The authors attribute this dilemma to a lack of support for scaling and managing process mining project portfolios. To address this practical need and research gap, the authors propose a method for managing portfolios of so-called process mining value cases, which are defined as process mining-enabled business process improvement projects, towards an evolutionary roadmap ( mapper ). The method is designed to support organizations identify portfolios of process mining projects that generate value by improving business processes. The method was developed through a combination of design science research and situational method engineering and comprises five activities that each outline techniques, roles, and tools: strategize , identify , select , implement , and monitor . The method has been instantiated as a software prototype and iteratively evaluated for applicability and real-world fidelity by involving an expert panel of academics and practitioners. The usefulness of the artifact was substantiated through a real-world case study in a naturalistic setting
Resilient Governance: Securing Public Services by Implementing Business Continuity Management Systems in German Municipal Administrations
No full textBusiness Continuity Management (BCM) complements information security management by providing reactive capabilities. Establishing a BCM is resource intensive and only few materials are tailored to German municipalities. This paper examines the BSI 200-4 standard, particularly its usability in the public sector. Our results indicate that the standard offers sufficient baseline start implementation but additional support from the state government is needed. In particular, the official supplementary materials are perceived as insufficiently usable, which might also reflect organizational and cultural constraints within the German public administration
Smart Public Displays: Einfluss von Anmeldeverfahren auf Technologieakzeptanz, Gebrauchstauglichkeit und Nutzungserlebnis
No full textIn der vorgestellten Studie werden zwei Varianten eines Smart Public Displays mit personalisierten Fahrgastinformationen des öffentlichen Personennahverkehrs verglichen, die Gesichtserkennung und eine RFID-fähige Karte als Anmeldeverfahren zu einem interaktiven persönlichen Bereich nutzen. Zwei unterschiedliche Abstände einer umstehenden Person wurden ebenfalls berücksichtigt. Die Studie mit zwei Gruppen aus je 30 Probanden untersucht, ob sich das Anmeldeverfahren auf die Technologieakzeptanz, die Gebrauchstauglichkeit und das Nutzungserlebnis auswirkt. Die Technologieakzeptanz, die Gebrauchstauglichkeit und das Nutzungserlebnis schneiden beim Smart Public Display mit Anmeldung via Karte signifikant besser ab. Selbes gilt für die Teilskalen Neugierde, Nützlichkeit, Zugänglichkeit und Skepsis des verwendeten TUI-Fragebogens. Das Anmeldeverfahren via Karte scheint das geeignetere für den persönlichen Bereich eines Smart Public Displays mit personalisierten Fahrgastinformationen zu sein. In the study presented, two variants of a smart public display with personalized passenger information for local public transport are compared, which use an RFID-enabled card and facial recognition as identification methods to an interactive personal area. Two different distances of a person standing nearby were also considered. The study with two groups of 30 subjects each tests the hypotheses whether the identification procedure has an impact on technology acceptance, usability and user experience. Technology acceptance, usability and user experience were significantly better for the smart public display with identification via card. The same applies to the subscales curiosity, usefulness, accessibility and skepticism. The identification via card appears to be more suitable for the personal area of a smart public display
Workarounds as a Cause of Mismatches in Business Processes—
No full textProcess mining has been established as a data-driven approach to analyze and improve business processes based on event data documented in event logs. A core assumption for meaningful analyses is that event data accurately represent the real-world execution of business processes in an organization. However, anecdotal evidence and recent case studies show that these aspects do not always align, and the business process management community is only beginning to investigate the mechanisms generating mismatches between process execution and event data. This study aims to identify the role of workarounds—goal-directed deviations from standard processes performed by process participants to overcome obstacles—in this context. Through an inductive multiple case study of 13 workarounds in four organizations, three mismatch categories between event logs and real-world process execution related to workarounds are identified and explored. This study contributes to the literature by describing how workarounds can act as mechanisms that cause mismatches between process execution and event data, adding to the discussion on process drift and workaround mining. Furthermore, exploring the mismatch categories offers insights for practitioners and researchers on how to handle and interpret data quality issues in event data
Does touch have impact on free style dance interaction in the VR?
No full textDancing is a basic natural, spontaneous activity that can bring people together and create a sense of connection. A key difference between dancing in physical environments and dancing in virtual reality is the lack of tactile and haptic stimuli through touch. Here we investigate whether the integration of haptic feedback positively influences interactive pair dance improvisation in virtual reality environments. We describe the design of stimuli for a virtual dance environment and investigate whether they have an impact on user presence, co-presence and unpleasent feelings. To answer these questions, two studies are conducted using focus groups, quantitative questionnaires and semi-structured interviews. The analysis of these studies have led to a design for the stimuli and a recommendation for the use of haptic feedback in the virtual dance environments
Single-image-based georeferencing for unmanned aerial vehicles
No full textGeoreferencing is important for many applications in precision farming, in particular those based on unmanned aerial vehicles (UAVs). In this context, georeferencing typically relates the optical features of UAV images to their actual position in the 3D world, creating a grid map of the area of interest. Although state-of-the-art georeferencing methods are very accurate, these methods rely on multiple-view geometry reconstruction, which requires largely overlapping images of high quality. Acquiring such images can be difficult in practice, given the low-cost requirements for precision farming. In this paper, we study the practical applications and challenges of a simple, computationally inexpensive and fast method for georeferencing that is solely based on single images. Our method only uses an affine transformation where the UAV’s height is adjusted by a digital terrain model and does not require overlapping images. We find that our single-image-based method can be used for smart farming applications, where spatial accuracies of around 25 cm are sufficient