Data cleaning
The acquired data is usually noisy and incomplete because of limitations in the acquisition process (hidden parts due to occlusions, windy conditions making the plant move, etc.). Denoising, filtering and completion are thus often necessary.
Terrestrial laser scanner noise analysis, modelling and detection
We propose methods to model and detect two types of noise in 3D point clouds generated by Terrestrial Laser Scanners (TLS): sky noise, which is the noise produced when an Amplitude Modulated Continuous Wave TLS measures range only from background radiation, and mixed point noise, corresponding to points acquired when the TLS receives return signals from several different surfaces. Such noises are frequent in TLS acquisitions of plants and trees.
Context: Romain Rombourg's Master and PhD theses, Digitree project
Collaborators: Romain Rombourg (Master then PhD student), Eric Casella (researcher, Forest Research UK)
Software: Matlab code available upon request
