The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples

Research output: Contribution to conference › Paper › Research › peer-review

Standard

The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples. / Belongie, Serge; Wang, Qianqian; Snavely, Noah; Khan, Awais.

2021. Paper presented at Plant Pathology 2021 Challenge - FGVC8.

Research output: Contribution to conference › Paper › Research › peer-review

Harvard

Belongie, S, Wang, Q, Snavely, N & Khan, A 2021, 'The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples', Paper presented at Plant Pathology 2021 Challenge - FGVC8, 15/03/2021 - 26/05/2021. <https://vision.cornell.edu/se3/wp-content/uploads/2021/09/029.pdf>

APA

Belongie, S., Wang, Q., Snavely, N., & Khan, A. (2021). The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples. Paper presented at Plant Pathology 2021 Challenge - FGVC8. https://vision.cornell.edu/se3/wp-content/uploads/2021/09/029.pdf

Vancouver

Belongie S, Wang Q, Snavely N, Khan A. The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples. 2021. Paper presented at Plant Pathology 2021 Challenge - FGVC8.

Author

Belongie, Serge ; Wang, Qianqian ; Snavely, Noah ; Khan, Awais. / The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples. Paper presented at Plant Pathology 2021 Challenge - FGVC8.3 p.

Bibtex

@conference{08f42118349e45e2868daa21d2bff6b7,
title = "The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples",
abstract = "Here, we describe the details and motivation behind the new dataset collected for the FGVC8 Plant Pathology 2021 Challenge competition. A total of 3,561 RGB images from four foliar disease categories of apples: apple scab, cedar apple rust, complex and healthy were used to launch the Plant Pathology 2020 Challenge competition on Kaggle.These images were captured with different distances from the leaves, angles, and sunlight to represent real world scenarios for disease symptoms on leaves of cultivated apples.Within 70 days of competition, 1,317 teams participated, and a machine learning based model with highest AUC (Area Under ROC Curve) value of 0.98 was reported. This year, we have significantly increased the dataset to 23,249RGB images, including two additional disease categories, powdery mildew and frogeye leaf spot. We have also introduced a new dataset for multi-label classification representing occurrence of multiple disease symptoms in single leaf. The dataset and competition was launched at Kaggle on March 15, 2021 and was open until May 26, 2021. A total of 657 teams participated in the competition this year and the highest mean F1-score reported on private leaderboard was 0.883.",
author = "Serge Belongie and Qianqian Wang and Noah Snavely and Awais Khan",
year = "2021",
month = may,
day = "26",
language = "English",
note = "Plant Pathology 2021 Challenge - FGVC8 ; Conference date: 15-03-2021 Through 26-05-2021",
url = "https://www.kaggle.com/competitions/plant-pathology-2021-fgvc8/data",

}

RIS

TY - CONF

T1 - The Plant Pathology 2021 Challenge dataset to classify foliar disease of apples

AU - Belongie, Serge

AU - Wang, Qianqian

AU - Snavely, Noah

AU - Khan, Awais

PY - 2021/5/26

Y1 - 2021/5/26

N2 - Here, we describe the details and motivation behind the new dataset collected for the FGVC8 Plant Pathology 2021 Challenge competition. A total of 3,561 RGB images from four foliar disease categories of apples: apple scab, cedar apple rust, complex and healthy were used to launch the Plant Pathology 2020 Challenge competition on Kaggle.These images were captured with different distances from the leaves, angles, and sunlight to represent real world scenarios for disease symptoms on leaves of cultivated apples.Within 70 days of competition, 1,317 teams participated, and a machine learning based model with highest AUC (Area Under ROC Curve) value of 0.98 was reported. This year, we have significantly increased the dataset to 23,249RGB images, including two additional disease categories, powdery mildew and frogeye leaf spot. We have also introduced a new dataset for multi-label classification representing occurrence of multiple disease symptoms in single leaf. The dataset and competition was launched at Kaggle on March 15, 2021 and was open until May 26, 2021. A total of 657 teams participated in the competition this year and the highest mean F1-score reported on private leaderboard was 0.883.

AB - Here, we describe the details and motivation behind the new dataset collected for the FGVC8 Plant Pathology 2021 Challenge competition. A total of 3,561 RGB images from four foliar disease categories of apples: apple scab, cedar apple rust, complex and healthy were used to launch the Plant Pathology 2020 Challenge competition on Kaggle.These images were captured with different distances from the leaves, angles, and sunlight to represent real world scenarios for disease symptoms on leaves of cultivated apples.Within 70 days of competition, 1,317 teams participated, and a machine learning based model with highest AUC (Area Under ROC Curve) value of 0.98 was reported. This year, we have significantly increased the dataset to 23,249RGB images, including two additional disease categories, powdery mildew and frogeye leaf spot. We have also introduced a new dataset for multi-label classification representing occurrence of multiple disease symptoms in single leaf. The dataset and competition was launched at Kaggle on March 15, 2021 and was open until May 26, 2021. A total of 657 teams participated in the competition this year and the highest mean F1-score reported on private leaderboard was 0.883.

M3 - Paper

T2 - Plant Pathology 2021 Challenge - FGVC8

Y2 - 15 March 2021 through 26 May 2021

ER -

ID: 304508381