Heikki Korpunen
Forsker
Biografi
Jeg er en forsker i driftsteknikk med spesiell vekt på treteknologi. Arbeidet mitt omfatter de tradisjonelle skogteknologiske spørsmål, som produktivitet- og tidsstudier, forskning innen treets verdikjede og apteringsoptimalisering, samt studier innen logistikk og bioenergi. Jeg fokuserer også på produksjon og kostnadsmodellering innen skogindustrien, samt analyser av treets egenskaper.
Utdannelse: Doktorgrad i skogfag, med spesialisering i treteknologi ved Universitetet i Helsinki, Finland (2015). Mastergrad i skogfag, med spesialisering i driftsteknikk og treteknologi ved Universitetet i Joensuu, Finland (2006).
Sammendrag
Time and motion studies in forest operations benefit from video-based analysis, but manual annotation is time consuming. This pilot study aims to reduce analysis time by developing a deep-learning framework that classifies dashcam video into four work elements: crane out, cutting and processing, driving, and processing. Using a 3D ResNet-50 (PyTorchVideo) trained on manually annotated clips, the model achieved validation F1 = 0.88 and precision = 0.90, showing that spatiotemporal CNNs can capture rele-vant motion and appearance cues in forest environments. Overfitting indicates that more diverse data and better class balance are needed, but the approach shows clear potential to scale automated work-element monitoring and efficiency analysis.
Forfattere
Katrin Zimmer Ulf Lemke Irena Fundova Henrik Heräjärvi Mateusz Liziniewicz Marlene Cramer Stefanie Wieland Lukas Emmerich Thomas Uibel Anti Rohumaa Heikki Korpunen Grace Jones Ekaterina Nikolskaya Tom Jansen Anneli Adler Tom Mortelmans Ian Short Vincent Defays Karol TomczakSammendrag
Birch is the third most abundant tree species in northern Europe and the Baltic region, but remains underutilized in several countries despite wood properties that support a broad range of applications, including pulp, veneer, plywood, furniture, flooring, joinery, and potentially structural products. Constraints on higher-value utilization include insufficient logistics for sorting and transport, the lack of standardized grading methods, and limited data infrastructure for systematic quality assessment across supply chains. The Nordic Forest Research (SNS) network VALUE:BIRCH was established to examine these constraints through transnational collaboration among researchers, industry actors, and students across northern Europe. In 2025 and 2026, the network held workshops in Borås, Sweden, and North Rhine-Westphalia, Germany, with emphasis on strength properties, grading, and quality assessment of birch wood. The activities integrated technical presentations, laboratory and field visits, and student contributions, enabling comparison of birch value chains across countries with differing levels of industrial development. The network identified shared technical and organizational bottlenecks related to birch silviculture and management, grading, mobilization, and market formation, while also strengthening inter-institutional co-creation and collaboration. The results indicate that coordinated work on grading systems, quality data, logistics, and market development is essential to support more efficient and value-adapted utilization of birch.
Sammendrag
Det er ikke registrert sammendrag
