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
Abstract
No abstract has been registered
Abstract
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
Matti Koivula Adam Felton Mari Jönsson Therese Löfroth Fride Høistad Schei Juha Siitonen Jörgen SjögrenAbstract
• This chapter summarises biodiversity responses to continuous cover forestry (CCF). The comparator throughout this chapter is rotation forestry (RF) and its main harvesting method—clearcutting—unless otherwise stated. • Research on the biodiversity effects of logging methods applied in CCF (mostly selection or gap cutting) mainly concerns the short-term effects of measures taken in mature, originally fairly even-aged forests, at best 10–15 years after cutting. Thus far, no surveys or chronosequences cover the whole rotation period (60–100 years). • Continuous cover forestry is likely to beneft species that suffer when the tree cover is removed, such as bilberry and its associated species. Species requiring spatial continuity in host trees or canopy cover may also benefit. • Selection cutting may preserve the majority of species in the mature forest, but the most sensitive species may decline or even disappear. Gap cutting (diameter 20–50 m) affects forest-interior species relatively little, but species’ abundances in gaps change with increasing gap size. Shelterwood cutting seems to closely resemble selection cutting in terms of species responses. In the long term, however, shelterwood cutting results in an even-aged and sparse overstorey, which does not produce the biodiversity benefits of CCF. • Species that have declined due to forestry mostly require large living and dead trees. The preservation of these species is not ensured by CCF alone, but requires deliberately maintaining these structural features. • A mosaic of different forest-management practices within landscapes may provide complementary ways to maintain rich biodiversity.
Abstract
This article presents a novel, ultralight tree planting mechanism for use on an aerial vehicle. Current tree planting operations are typically performed manually, and existing automated solutions use large land-based vehicles or excavators which cause significant site damage and are limited to open, clear-cut plots. Our device uses a high-pressure compressed air power system and a novel double-telescoping design to achieve a weight of only 8 kg: well within the payload capacity of medium to large drones. This article describes the functionality and key components of the device and validates its feasibility through experimental testing. We propose this mechanism as a cost-effective, highly scalable solution that avoids ground damage, produces minimal emissions, and can operate equally well on open clear-cut sites as in denser, selectively-harvested forests.
Abstract
No abstract has been registered
Authors
Heikki Korpunen Yrjö Nuutinen Paula Jylhä Lars Eliasson Aksel Granhus Juha Laitila Stephan Hoffmann Timo MuhonenAbstract
• Overall forest management objectives and stand properties set the requirements and possibilities for harvesting in continuous cover forestry (CCF). • Harvester and forwarder operators play a key role in successful CCF harvesting, as both productivity and quality of work are essential factors in harvesting operations. • Optimal stand conditions improve work productivity on selection harvesting sites; harvested stem volume correlates well with work productivity in cutting, and density of remaining trees does not significantly reduce work productivity in forwarding. • Carefully executed group cutting and shelterwood harvesting can reduce the number of damaged remaining trees, which is beneficial for future tree generations. • Research-based information is needed about work productivity in harvesting, damage caused by harvesting, and optimisation of strip road and forest road networks for CCF.
Abstract
No abstract has been registered
Authors
Mostafa Hoseini Helle Ross Gobakken Stephan Hoffmann Csongor Horvath Johannes Rahlf Jan Bjerketvedt Stefano Puliti Rasmus AstrupAbstract
No abstract has been registered
2024
Abstract
No abstract has been registered