Canopy geometric and structural parameters in a super-intensive almond (Prunus dulcis) orchard were estimated from multispectral vegetation indices from very high-resolution UAV images (0.07 m/pixel). As ground truth data, the canopy geometric and structural parameters were extracted from a 3D LiDAR point cloud. These data were summarized every 0.5 m along the tree rows. The vegetation indices CCCI, NDRE, NDVI, GNDVI and WDRVI were calculated and summarized at the same points as the LiDAR data. The highest correlations (R>0.70) were obtained between the NDVI and the maximum width and the cross-sectional area. The porosity was negatively correlated (R=-0.59) with the NDVI. This research opens the opportunity of using remotely sensed vegetation indices to estimate canopy geometric and structural parameters in addition to vigour for precise canopy management.