Next Generation Woodmeasurement

Important results from the project

We see that it is possible to measure lumber this way and the measurement accuracy is on par with what we predicted in our early simulations. We have been able to show potential customers and stakeholders our results, based on the data we have now, and received a good response from the market. By being able to visualize and show, together with solid sales work, we have received the award as one of Sweden´s hottest startups. The purpose was to get closer to commercialization and with both a working concept and a good response from the market, we feel that we are heading in the right direction.

Expected long term effects

Before project start, we simulated our methodology would provide significantly better measurement accuracy than conventional lumber measurement systems. This was confirmed during the field tests, where we achieved 2-3 times better accuracy, depending on the condition of the wood. After the measurements, we realized that the sensors at the back of the truck require better optics, especially a longer focal length, to see the logs more clearly. For the algorithm, we need to collect more data from different weather conditions and environments to increase the robustness of the system.

Approach and implementation

The project has followed the defined milestones. Subgoal 1 changed from using a 3D camera to a mobile phone on a logging truck for data collection, which turned out to be easier. The main objective of collecting training data was met and is still ongoing. Subgoal 2 is met and worked in field test. Subgoal 3 was achieved by obtaining measurement data, but more training data is needed for more robust algorithm. Subgoal 4 succeeded in visualizing log segmentations, increasing customer interest. The next step is continued data collection, algorithm improvement and customer engagement.