How can we trace timber from its harvest site to the end product?
Our SmartForest PhD candidate Yohann Jacob Sandvik, together with Mostafa Hoseini og Carolin Fischer recently published their work
A precise deep learning approach for timber traceability along the forest value chain
Congratulations!
Their study proposes a cost-effective, image-based method for tracing logs using alphabetic codes printed onto logs at the harvest site. Theses codes are detected and interpreted through a two stage system that uses deep learning models.
The detection stage uses YOLOv8 to locate tracking codes in images of log piles. It is trained and evaluated on a dataset of 125 images, achieving an F1-score of 0.811 on unseen images.
The recognition stage, trained on 1 020 images, uses YOLOv8 models to detect individual charactersand their position within each code. On a set of unseen images, the interpretation stage is able to identify 98.2% of the individual logs despite the limited quality of the printer and degradation of the codes caused by wet stems.
Analysis indicates that errors predominantly arise in the character detection step. Compared to existing tracability approaches, this method is more cost-effective than RFID tags and attains higher accuracy than image-based biomarker tracking methods.
Read the article here: https://doi.org/10.1080/14942119.2025.2572938