Name
Abstract | ||
|---|---|---|
•An approach is proposed to detect whole grape bunches in vineyard scenes.•The approach relies on shape (FRST, HOG), texture (LBP) descriptors and a novel bunch separation strategy.•The method does not use color features as most of the existing methods do.•Applicable to white or dark grapes with natural or artificial illumination and different occlusion levels.•Precision and recall detection rates are respectively 88.6% and 80.3% for grape bunches, while 99% and 84.0 to 92.5% for single berries. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1016/j.compag.2018.05.019 | Computers and Electronics in Agriculture |
Keywords | Field | DocType |
Grape bunch detection,Grape recognition,Precision viticulture,Histogram of oriented gradients,Local binary pattern,Support vector machine | Vineyard,Computer vision,Histogram,Feature vector,Pattern recognition,Thinning,Bunches,Local binary patterns,Support vector machine,Artificial intelligence,Pixel,Engineering | Journal |
Volume | ISSN | Citations |
151 | 0168-1699 | 0 |
PageRank | References | Authors |
0.34 | 12 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Rodrigo Pérez-Zavala | 1 | 0 | 0.34 |
| Miguel Torres-Torriti | 2 | 84 | 10.61 |
| Fernando Auat | 3 | 26 | 8.39 |
| Giancarlo Troni | 4 | 21 | 4.17 |