In September 2021 a new doctoral project started at Linnaeus University, with Softwerk as partner and co-funder. The project is called Advanced identification methods for the forest industry through CV/AI and is also funded by the Knowledge Foundation (KK-stiftelsen).
Doctoral student Dag Björnberg works on it in the research group DISTA, part of the Linnaeus University Centre for Data Intensive Sciences and Applications (DISA). The supervisor is Morgan Ericsson.
What the project does
The aim is to support further digitalisation in the forest industry, from traceability to quality assessment. It has two tracks:
- Following logs: an AI model that recognises the same log at different stations in a sawmill. This lays the groundwork for tracing logs through the whole production chain.
- Training data on demand: using generative models (GANs) to create images of logs with a chosen pith position and age. This cuts down the slow manual labelling needed to train models for quality assessment.
The project is part of our long collaboration with Linnaeus University.