Point-cloud classification is one of the most impor- tant and time consuming stages of airborne LiDAR data process- ing, playing a key role in the generation of cartographic products. This paper describes an innovative algorithm to perform LiDAR point-cloud classification, that relies on Convolutional Neural Networks and takes advantage of full-waveform data registered by modern laser scanners. The proposed method consists of two steps. First, a simple CNN is used to pre-process each waveform, providing a compact representation of the data. Exploiting the coordinates of the points associated to the waveforms, output vectors generated by the first CNN are then mapped into an image, that is subsequently segmented by a Fully Convolutional Network: a label is assigned to each pixel and, consequently, to the point falling in the pixel. In this way, spatial positions and geometrical relationships between neighbouring data are taken into account. These particular architectures allow to accurately identify even challenging classes such as power line and transmission tower.
Full-Waveform Airborne LiDAR Data Classification Using Convolutional Neural Networks
ZORZI, STEFANO;Maset, Eleonora
;Fusiello, Andrea;Crosilla, Fabio
2019-01-01
Abstract
Point-cloud classification is one of the most impor- tant and time consuming stages of airborne LiDAR data process- ing, playing a key role in the generation of cartographic products. This paper describes an innovative algorithm to perform LiDAR point-cloud classification, that relies on Convolutional Neural Networks and takes advantage of full-waveform data registered by modern laser scanners. The proposed method consists of two steps. First, a simple CNN is used to pre-process each waveform, providing a compact representation of the data. Exploiting the coordinates of the points associated to the waveforms, output vectors generated by the first CNN are then mapped into an image, that is subsequently segmented by a Fully Convolutional Network: a label is assigned to each pixel and, consequently, to the point falling in the pixel. In this way, spatial positions and geometrical relationships between neighbouring data are taken into account. These particular architectures allow to accurately identify even challenging classes such as power line and transmission tower.File | Dimensione | Formato | |
---|---|---|---|
tgrs19.pdf
non disponibili
Tipologia:
Documento in Post-print
Licenza:
Non pubblico
Dimensione
1.92 MB
Formato
Adobe PDF
|
1.92 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.