P. Rönnholm*, S. Wittke, M. Ingman, P. Putkiranta, H. Kauhanen, H. Kaartinen, M. T. Vaaja

*Corresponding author for this work

Research output: Contribution to journalConference articleScientificpeer-review

47 Downloads (Pure)


The aim of our research was to examine whether simulated forest data can be utilized for training supervised classifiers. We included two classifiers namely the random forest classifier and the novel convolutional neural network classifier that utilizes feature images. We simulated tree parameters and created a feature vector for each tree. The original feature vector was utilised with random forest classifier. However, these feature vectors were also converted into feature images suitable for input into a YOLO (You Only Look Once) convolutional neural network classifier. The selected features were red colour, green colour, near-infrared colour, tree height divided by canopy diameter, and NDVI. The random forest classifier and convolutional neural network classifier performed similarly both with simulated data and field-measured reference data. As a result, both methods were able to identify correctly 97.5 % of the field-measured reference trees. Simulated data allows much larger training data than what could be feasible from field measurements.

Original languageEnglish
Pages (from-to)633-639
Number of pages7
JournalInternational Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
Issue numberB2-2022
Publication statusPublished - 30 May 2022
MoE publication typeA4 Conference publication
EventISPRS Congress “Imaging today, foreseeing tomorrow” - Nice, France
Duration: 6 Jun 202211 Jun 2022
Conference number: 24


  • classification
  • convolutional neural network
  • feature image
  • simulation
  • YOLO


Dive into the research topics of 'UTILISING SIMULATED TREE DATA TO TRAIN SUPERVISED CLASSIFIERS'. Together they form a unique fingerprint.

Cite this