Plant phenotyping attempts to objectively measure a plant's reaction to its environment as encoded by its genotype. It has become an essential tool for deepening our understanding of plant responses to environmental stimuli. Understanding the plant's reaction to, for example, a warmer climate is crucial to ensure food production for future generations. Breeders and researchers rely on automated high-throughput phenotyping for optimizing crops. Ideally, above- and below-ground traits are observed simultaneously. The newly established controlled environment facility at the Technology Center for Phenotyping of the Fraunhofer IIS in Merkendorf provides several climate chambers with individually controllable conditions for up to 400 individual plants to allow simulation of even extreme climatic conditions all year around. Comprehensive measurement of plant structures using X-ray as well as optical cameras provide highly detailed 2D and 3D information to researchers and breeders worldwide. In combination with automated data pipelines, distinct plant traits can be extracted from the sensor data. By bridging above- and below-ground phenotyping, this facility not only advances plant science but also contributes to the breeding of more resilient and productive crops. Collaborators are welcome to unlock the transformative potential of these unique phenotyping capabilities, exploring traits such as root length, leaf area, biomass, and more. Unter Pflanzenph & auml;notypisierung versteht man die m & ouml;glichst objektive Erfassung der genotypabh & auml;ngigen Reaktion einer Pflanze auf ihre Umwelt. Sie ist zu einem wichtigen Instrument geworden, um unser Verst & auml;ndnis der Reaktionen von Pflanzen auf Umweltreize zu vertiefen. Das Verst & auml;ndnis der Reaktion der Pflanze auf z. B. ein w & auml;rmeres Klima ist der Schl & uuml;ssel zur Sicherung der Nahrungsmittelproduktion f & uuml;r k & uuml;nftige Generationen. Z & uuml;chter und Forscher verlassen sich bei der Optimierung von Kulturpflanzen auf die automatisierte Ph & auml;notypisierung im Hochdurchsatzverfahren. Im Idealfall werden ober- und unterirdische Merkmale gleichzeitig beobachtet. Das neu er & ouml;ffnete Technologiezentrum f & uuml;r Ph & auml;notypisierung des Fraunhofer IIS in Merkendorf bietet in kontrollierter Umgebung mehrere Klimakammern mit individuell steuerbaren Bedingungen f & uuml;r bis zu 400 Einzelpflanzen, um selbst extreme Klimabedingungen ganzj & auml;hrig simulieren zu k & ouml;nnen. Die ganzheitliche Erfassung von Pflanzenstrukturen mit R & ouml;ntgen- und optischen Kameras liefert hoch detaillierte 2D- und 3D-Informationen f & uuml;r Forscher und Z & uuml;chter weltweit. In Kombination mit automatisierten Datenpipelines k & ouml;nnen aus den Sensordaten Pflanzeneigenschaften extrahiert werden. Durch die Verkn & uuml;pfung von ober- und unterirdischer Ph & auml;notypisierung bringt diese Einrichtung nicht nur die Pflanzenforschung voran, sondern tr & auml;gt auch zur Z & uuml;chtung widerstandsf & auml;higerer und produktiverer Nutzpflanzen bei. Partner sind willkommen, um das transformative Potenzial dieser einzigartigen Ph & auml;notypisierungskapazit & auml;ten zu erschlie ss en und Merkmale wie Wurzell & auml;nge, Blattfl & auml;che und Biomasse zu erforschen.
Subjective evaluation remains the most reliable way of testing speech and audio coding techniques. Crowdsourcing the listening task is a cost-efficient and fast way of conducting this evaluation, but the quality of the results tends to be inferior to that of conventional listening tests done in the controlled environment of a laboratory. In this paper, classical and neural speech codecs are evaluated to compare P.808 against P.800 DCR tests. A statistical analysis is conducted to investigate the effectiveness of selected screening methods. The analysis shows that the crowdsourced evaluation can be improved by employing postscreening methods based on anchor ordering and rating span, and continuous screening methods like traps and gold standard questions, thus giving more value to the ratings obtained for the codecs under test. Based on these outcomes, a set of suitable screenings is proposed, for cost-effective, simplified, and bias-free enhancement of listening results.
Enabling efficient deep neural network (DNN) inference on edge devices with different hardware constraints is a challenging task that typically requires DNN architectures to be specialized for each device separately. To avoid the huge manual effort, one can use neural architecture search (NAS). However, many existing NAS methods are resource-intensive and time-consuming because they require the training of many different DNNs from scratch. Furthermore, they do not take the resource constraints of the target system into account. To address these shortcomings, we propose PrototypeNAS, a zero-shot NAS method to accelerate and automate the selection, compression, and specialization of DNNs to different target microcontroller units (MCUs). We propose a novel three-step search method that decouples DNN design and specialization from DNN training for a given target platform. First, we present a novel search space that not only cuts out smaller DNNs from a single large architecture, but instead combines the structural optimization of multiple architecture types, as well as optimization of their pruning and quantization configurations. Second, we explore the use of an ensemble of zero-shot proxies during optimization instead of a single one. Third, we propose the use of Hypervolume subset selection (HSS) to distill DNN architectures from the Pareto front of the multi-objective optimization (MOO) that represent the most meaningful tradeoffs between accuracy and floating-point operations (FLOPs). We evaluate the effectiveness of PrototypeNAS on 12 different datasets in three different tasks: image classification, time series classification, and object detection. Our results demonstrate that PrototypeNAS is able to identify DNNs within minutes that are small enough to be deployed on off-the-shelf MCUs and still achieve accuracies comparable to the performance of large DNN architectures.
Interference poses a significant challenge to satellite-based positioning systems, making it essential to accurately quantify the effects of specific interference types on receiver performance and the resulting reliability of position computation. In current practice, interference effects are often quantified using receiver-independent metrics, with receiver-specific front-end characteristics either idealized or only implicitly considered. In this paper, we address this limitation by explicitly incorporating receiver-specific front-end characteristics into the computation of interference effects and validating the resulting receiver-dependent analysis experimentally. Therefore, we record a real-world open-field dataset comprising 210 distinct interference scenarios and compute the receiver-dependent spectral separation coefficient (SSC) and interference impact for a specific receiver module. Furthermore, we verify the computation using a controlled dataset generated with a radio frequency constellation simulator (RFCS), employing the same receiver module and replaying similar interferences classes. The comparison of results obtained in both environments demonstrates the robustness of the interference impact computation.
In this study, we conduct a comprehensive comparative analysis of generative and discriminative deep learning-based speech enhancement methods, specifically in noise reduction tasks. Our investigation focuses on evaluating their effectiveness under high and low signal-to-noise ratio conditions, considering both matched and mismatched training scenarios. We further investigate the impact of training data volume, model convergence speed, and interpret the performance differences in terms of objective results for the considered training paradigms. Additionally, we compare the complexity-performance trade-off and the practical viability of these approaches. To further strengthen the evaluation, we study the hallucination characteristics of generative approaches in terms of word error rate and phoneme similarity. The insights derived from this study provide empirical evidence to assist researchers and practitioners in understanding whether the perceptual gains of different approaches justify their computational cost in practical applications.
