capcelera

 

M-UdL_200

 

 

Logo Agrotecnio

 

 

 Centres

ETSEA


 

 

Departaments

EAGROF

 

 

 

 

 

 

Inici > English version > GRAP publications > Apple size estimation using SfM-MVS point clouds

Apple size estimation using SfM-MVS point clouds

 

 

Fruit size estimation results related to the research article entitled “In-field apple size estimation using photogrammetry-derived 3D point clouds: comparison of 4 different methods considering fruit occlusions” [1].

First the captured scene (three Fuji apple trees) and the corresponding fruit detections are shown. Then, four examples of apple estimated diameters using MSAC algorithm are illustrated and compared with ground truth measurements. Fruit visibility results are also reported. The reader is referred to the corresponding research article [1] for a detailed description of the methodology and a complete representation of results, including an evaluation of four different tested methods assessed using the PFuji-Size dataset [2].

 

PFuji-Size datset

The PFuji-Size dataset [2] includes a total of 615 Fuji apples scanned in-field conditions and an additional 25 apples scanned in laboratory conditions. Structure-from-motion and multi-view stereo techniques were used to generate the 3D point clouds of the captured scene. Apple locations and ground truth diameter annotations are provided for assessing fruit detection and size estimation algorithms.

The dataset can be downloaded from the following links: PFuji-Size_dataset (600 MB); PFuji-Size_dataset_raw_images (25.3 GB)

This dataset is available only for research and educational purpose and not for any commercial use. If you use the database in any publications or reports, you must refer to the following papers: [1] and [2].

 

Code

The baseline used for apple size estimation in [1] is publicly available at apple_size_estimation_in_3D_point_clouds

 

References

[1] Gené-Mola J, Sanz-Cortiella R, Rosell-Polo JR, Escolà A, Gregorio E. 2021. In-field apple size estimation using photogrammetry-derived 3D point clouds: comparison of 4 different methods considering fruit occlusions. (Submitted)

[2] Gené-Mola J, Sanz-Cortiella R, Rosell-Polo JR, Escolà A, Gregorio E. 2021. PFuji-Size dataset: a collection of photogrammetry-derived 3D point clouds with ground truth annotations for Fuji apple detection and size estimation in field conditions. (Submitted)

 

Last modified: 22/03/2021
Imprimir Enrera Pujar


Home | Accessibility | About the web | Site map | Legal notice | Contact
© 2014 Universitat de Lleida - Pl. de Víctor Siurana 1, 25003 Lleida - Tel. (+34) 973 702 000 - All rights reserved
Ombra