Tree stem reconstruction from terrestrial laser scanner point cloud using Hough transform and open active contours

Joris A Ravaglia (1, 2, 3) , Alexandra Bac (2, 3) , Richard A Fournier (1)

1 UdeS - Université de Sherbrooke
2 AMU - Aix Marseille Université
3 LSIS - Laboratoire des Sciences de l'Information et des Systèmes

We introduce an innovative methodology for reconstructing tree stems from terrestrial LiDAR data. The method uses a multi-scale combination of an original Hough transform and active contours. Tests were conducted on both simulated data and real LiDAR data. Even though no formal evaluation was completed yet, the procedure seems promising.

Mots clés: terrestrial LiDAR; stem reconstruction; Hough transform; open active contours

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Sylviculture, foresterie

Joris A Ravaglia, Alexandra Bac, Richard A Fournier. Tree stem reconstruction from terrestrial laser scanner point cloud using Hough transform and open active contours. Silvilaser 2015, Sep 2015, La Grande Motte, France. ⟨hal-01376470⟩