Publications
NIBIOs employees contribute to several hundred scientific articles and research reports every year. You can browse or search in our collection which contains references and links to these publications as well as other research and dissemination activities. The collection is continously updated with new and historical material.
2025
Authors
Heidi Udnes Aamot Adnan Šišić Lars Olav Brandsæter Silje Kvist Simonsen Birgitte Henriksen Jörg Peter BareselAbstract
Over the past decades, significant efforts have been made to promote the cultivation of legumes. Cultivation of legumes, particularly grain legumes, can reduce the use of mineral nitrogen fertilizers, enhance biodiversity, reduce dependence on imported feed proteins, and improve soil biological properties and humus content. Despite these efforts, grain legumes are still not widely grown. One major obstacle to legume cultivation is "legume fatigue". Legume fatigue limits the expansion of legume cultivation in many European regions. The exact causes of legume fatigue are not fully understood, but soil-borne diseases interacting with abiotic factors are believed to play a key role. Recent findings suggest that the balance between pathogen load and soil suppressiveness is critical. Some farms and regions do not report legume fatigue as a problem, while others experience severe limitations in legume production. Identifying the causes of this variation is urgent and requires a collaborative effort that covers different environments and includes comprehensive assessments of both biotic and abiotic factors. In a recently launched project, LeFaSus, a network of farms and long-term experiments has been established to identify the primary factors contributing to legume fatigue. This network spans a transect from southern to northern Europe, including Italy, Germany, Luxembourg, and Norway. The project aims to deliver a reliable set of indicators for both legume fatigue and disease-suppressive soils, linking these indicators to the management practices that likely influenced them. The background and plans for the project will be presented.
Authors
Daniel Flø Johan A. Stenberg Lawrence Richard Kirkendall Kjetil Klaveness Melby Anders Nielsen Selamawit Tekle Gobena Beatrix Alsanius Jorunn Børve Paal Krokene Christer Magnusson Mogens Nicolaisen Line Nybakken May-Guri Sæthre Iben Margrete ThomsenAbstract
No abstract has been registered
Authors
Daniel Flø Johan A. Stenberg Lawrence Richard Kirkendall Kjetil Klaveness Melby Anders Nielsen Selamawit Tekle Gobena Beatrix Alsanius Jorunn Børve Paal Krokene Christer Magnusson Mogens Nicolaisen Line Nybakken May-Guri Sæthre Iben M. Thomsen Sandra WrightAbstract
No abstract has been registered
Authors
Daniel Flø Johan A. Stenberg Kjetil Klaveness Melby Selamawit Tekle Gobena Beatrix Alsanius Jorunn Børve Paal Krokene Christer Magnusson Mogens Nicolaisen Line Nybakken May-Guri Sæthre Iben M. Thomsen Sandra WrightAbstract
No abstract has been registered
Authors
Monica Sanden Eirill Ager-Wick Johanna Eva Bodin Nur Duale Anne-Marthe Ganes Jevnaker Kristian Prydz Volha Shapaval Ville Erling Sipinen Tage ThorstensenAbstract
No abstract has been registered
Abstract
No abstract has been registered
Authors
David Chludil Curt Almqvist Mats Berlin Arne Steffenrem Steven E. McKeand Jiří Korecký Jan Stejskal Jaroslav Čepl Fikret Isik Debojyoti Chakraborty Silvio Schueler Torsten Pook Christi Sagariya Milan LstibůrekAbstract
No abstract has been registered
Authors
Sean P Healey Zhiqiang Yang Angela M Erb Ryan Bright Grant M Domke Tracey S Frescino Crystal B SchaafAbstract
No abstract has been registered
Abstract
Accurately predicting whether pedestrians will cross in front of an autonomous vehicle is essential for ensuring safe and comfortable maneuvers. However, developing models for this task remains challenging due to the limited availability of diverse datasets containing both crossing (C) and non-crossing (NC) scenarios. Therefore, we propose a procedure that leverages synthetic videos with C/NC labels and an untrained model whose architecture is designed for C/NC prediction to automatically produce C/NC labels for a set of real-world videos. Thus, this procedure performs a synth-to-real unsupervised domain adaptation for C/NC prediction, so we term it S2R-UDA-CP. To assess the effectiveness of S2R-UDA-CP in self-labeling, we utilize two state-of-the-art models, PedGNN and ST-CrossingPose, and we rely on the publicly-available PedSynth dataset, which consists of synthetic videos with C/NC labels. Notably, once the real-world videos are self-labeled, they can be used to train models different from those used in S2R-UDA-CP. These models are designed to operate onboard a vehicle, whereas S2R-UDA-CP is an offline procedure. To evaluate the quality of the C/NC labels generated by S2R-UDA-CP, we also employ PedGraph+ (another literature referent) as it is not used in S2R-UDA-CP. Overall, the results show that training models to predict C/NC using videos labeled by S2R-UDA-CP achieves performance even better than models trained on human-labeled data. Our study also highlights different discrepancies between automatic and human labeling. To the best of our knowledge, this is the first study to evaluate synth-to-real self-labeling for C/NC prediction.
Authors
Daniel Moreno-Fernández Johannes Breidenbach Isabel Cañellas Gherardo Chirici Giovanni D’amico Marco Ferretti Francesca Giannetti Stefano Puliti Sebastian Schnell Ross Shackleton Mitja Skudnik Iciar AlberdiAbstract
No abstract has been registered