EVALUATING DATA AUGMENTATION FOR GRAPEVINE VARIETIES IDENTIFICATION

被引：0
|
作者：
Carneiro, Gabriel [1 ,2 ]
Neto, Alexandre [1 ,2 ]
Teixeira, Ana [1 ,2 ]
Cunha, Antonio [1 ,2 ]
Sousa, Joaquim [1 ,2 ]
机构：
[1] Univ Tras Os Montes & Alto Douro, Vila Real, Portugal
[2] INESC TEC, Porto, Portugal
来源：
IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM | 2023年
基金：
欧盟地平线“2020”;
关键词：
deep learning; grapevine variety identification; convolutional neural networks; data augmentation; precision viticulture;
D O I：
10.1109/IGARSS52108.2023.10283128
中图分类号：
P [天文学、地球科学];
学科分类号：
07 ;
摘要：
The grapevine variety identification is important in the wine's production chain since it is related to its quality, authenticity and singularity. In this study, we addressed the data augmentation approach to identify grape varieties with images acquired in-field. We tested the static transformations, RandAugment, and Cutmix methods. Our results showed that the best result was achieved by the Static method generating 5 images per sample (F1 = 0.89), however without a significative difference if compared with RandAugment generating 2 images. The worst performance was achieved by CutMix (F1 = 0.86).
引用
收藏
页码：3566 / 3569
页数：4
相关论文
共 50 条
  • [1] GRAPEVINE VARIETIES IDENTIFICATION USING VISION TRANSFORMERS
    Carneiro, Gabriel Antonio
    Padua, Luis
    Peres, Emanuel
    Morais, Raul
    Sousa, Joaquim J.
    Cunha, Antonio
    2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022), 2022, : 5866 - 5869
  • [2] Prospection and identification of grapevine varieties cultivated in north Portugal and northwest Spain
    Martin, J. P.
    Arranz, C.
    Castro, I. D.
    Yuste, J.
    Rubio, J. A.
    Pinto-Carnide, O.
    Ortiz, J. M.
    VITIS, 2011, 50 (01) : 29 - 33
  • [3] Identification of grapevine varieties using leaf spectroscopy and partial least squares
    Diago, Maria P.
    Fernandes, A. M.
    Millan, B.
    Tardaguila, J.
    Melo-Pinto, P.
    COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2013, 99 : 7 - 13
  • [4] Characterization of main grapevine varieties of Albania and Kosovo based on molecular data
    Bacu, A.
    Bajraktari, Y.
    Papa, S.
    Thomaj, F.
    Michailidou, S.
    Argyriou, A.
    VITIS, 2015, 54 : 91 - 92
  • [5] Identification of Moldy Peanuts under Different Varieties and Moisture Content Using Hyperspectral Imaging and Data Augmentation Technologies
    Liu, Ziwei
    Jiang, Jinbao
    Li, Mengquan
    Yuan, Deshuai
    Nie, Cheng
    Sun, Yilin
    Zheng, Peng
    FOODS, 2022, 11 (08)
  • [6] Recovering old grapevine varieties
    Garcia, J.
    Peiro, R.
    Martinez-Gil, F.
    Soler, J. X.
    Jimenez, C.
    Yuste, A.
    Xirivella, C.
    Gisbert, C.
    VITIS, 2020, 59 (03) : 101 - 103
  • [7] Looking for old grapevine varieties
    Jimenez, C.
    Peiro, R.
    Yuste, A.
    Garcia, J.
    Martinez-Gil, F.
    Gisbert, C.
    VITIS, 2019, 58 (02) : 59 - 60
  • [8] Recovery and identification of grapevine varieties cultivated in old vineyards from Navarre (Northeastern Spain)
    Urrestarazu, Jorge
    Miranda, Carlos
    Santesteban, Luis G.
    Royo, Jose B.
    SCIENTIA HORTICULTURAE, 2015, 191 : 65 - 73
  • [9] Nonlinear system identification via data augmentation
    Formentin, Simone
    Mazzoleni, Mirko
    Scandella, Matteo
    Previdi, Fabio
    SYSTEMS & CONTROL LETTERS, 2019, 128 : 56 - 63
  • [10] Ampelographical description of new grapevine varieties
    Pavlousek, P
    PROCEEDINGS OF THE 8TH INTERNATIONAL CONFERENCE ON GRAPE GENETICS AND BREEDING, VOLS 1 AND 2, 2003, (603): : 633 - 640
12345