The Plant Pathology Challenge 2020 data set to classify foliar disease of apples

Premise Apple orchards in the United States are under constant threat from a large number of pathogens and insects. Appropriate and timely deployment of disease management depends on early disease detection. Incorrect and delayed diagnosis can result in either excessive or inadequate use of chemicals, with increased production costs and increased environmental and health impacts. Methods and Results We have manually captured 3651 high‐quality, real‐life symptom images of multiple apple foliar diseases, with variable illumination, angles, surfaces, and noise. A subset of images, expert‐annotated to create a pilot data set for apple scab, cedar apple rust, and healthy leaves, was made available to the Kaggle community for the Plant Pathology Challenge as part of the Fine‐Grained Visual Categorization (FGVC) workshop at the 2020 Computer Vision and Pattern Recognition conference (CVPR 2020). Participants were asked to use the image data set to train a machine learning model to classify disease categories and develop an algorithm for disease severity quantification. The top three area under the ROC curve (AUC) values submitted to the private leaderboard were 0.98445, 0.98182, and 0.98089. We also trained an off‐the‐shelf convolutional neural network on this data for disease classification and achieved 97% accuracy on a held‐out test set. Discussion This data set will contribute toward development and deployment of machine learning–based automated plant disease classification algorithms to ultimately realize fast and accurate disease detection. We will continue to add images to the pilot data set for a larger, more comprehensive expert‐annotated data set for future Kaggle competitions and to explore more advanced methods for disease classification and quantification.

[1]  S. Pethybridge,et al.  Leaf Doctor: A New Portable Application for Quantifying Plant Disease Severity. , 2015, Plant disease.

[2]  Yun Zhang,et al.  Identification of Apple Leaf Diseases Based on Deep Convolutional Neural Networks , 2017, Symmetry.

[3]  Marcel Salathé,et al.  Using Deep Learning for Image-Based Plant Disease Detection , 2016, Front. Plant Sci..

[4]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[5]  Andreas Peil,et al.  Improvement of Fire Blight Resistance in Apple and Pear , 2009 .

[6]  Andrey Nechaevskiy,et al.  Disease Detection on the Plant Leaves by Deep Learning , 2018, Advances in Neural Computation, Machine Learning, and Cognitive Research II.

[7]  Robinson Piramuthu,et al.  HD-CNN: Hierarchical Deep Convolutional Neural Networks for Large Scale Visual Recognition , 2014, 2015 IEEE International Conference on Computer Vision (ICCV).

[8]  Chen Hanjie,et al.  Mass trapping of apple leafminer, Phyllonorycter ringoniella with sex pheromone traps in apple orchards , 2017 .

[9]  Anne-Katrin Mahlein Plant Disease Detection by Imaging Sensors - Parallels and Specific Demands for Precision Agriculture and Plant Phenotyping. , 2016, Plant disease.

[10]  Matthias Scheffler,et al.  NOMAD 2018 Kaggle Competition: Solving Materials Science Challenges Through Crowd Sourcing , 2018, 1812.00085.

[11]  Thomas Wöhner,et al.  Apple blotch disease (Marssonina coronaria (Ellis & Davis) Davis) – review and research prospects , 2018, European Journal of Plant Pathology.

[12]  Herb S. Aldwinckle,et al.  Compendium of Apple and Pear Diseases and Pests, Second Edition , 2013 .

[13]  Nima Hatami,et al.  Classification of time-series images using deep convolutional neural networks , 2017, International Conference on Machine Vision.

[14]  Ashraf M. El-Sayed,et al.  Development of kairomone-based lures and traps targeting Spilonota ocellana (Lepidoptera: Tortricidae) in apple orchards treated with sex pheromones , 2017, The Canadian Entomologist.

[15]  B. Porter The Apple Maggot , 1928 .

[16]  D. Gadoury,et al.  Delaying the onset of fungicide programs for control of apple scab in orchards with low potential ascospore dose of Venturia inaequalis , 1993 .

[17]  T. German,et al.  PCR primers that allow intergeneric differentiation of ascomycetes and their application to Verticillium spp , 1994, Applied and environmental microbiology.

[18]  Andreas Kamilaris,et al.  Deep learning in agriculture: A survey , 2018, Comput. Electron. Agric..

[19]  S. Muthumeenakshi,et al.  A PCR primer-specific to Cylindrocarpon heteronema for detection of the pathogen in apple wood. , 1993, FEMS microbiology letters.

[20]  A. Bifet,et al.  Kaggle LSHTC4 Winning Solution , 2014, ArXiv.

[21]  Jayme Garcia Arnal Barbedo,et al.  A review on the main challenges in automatic plant disease identification based on visible range images , 2016 .

[22]  Jayme Garcia Arnal Barbedo,et al.  An Automatic Method to Detect and Measure Leaf Disease Symptoms Using Digital Image Processing. , 2014, Plant disease.

[23]  C. Lane,et al.  A Polymerase Chain Reaction-Based Method to Specifically Detect Alternaria alternata Apple Pathotype (A. mali), the Causal Agent of Alternaria Blotch of Apple. , 2000, Phytopathology.

[24]  D. Gadoury,et al.  Forecasting ascospore dose of Venturia inaequalis in commercial apple orchards , 1986 .

[25]  Vladimir Iglovikov,et al.  Satellite Imagery Feature Detection using Deep Convolutional Neural Network: A Kaggle Competition , 2017, ArXiv.

[26]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[27]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[28]  Cesare Gessler,et al.  Parasitic and Biological Fitness of Venturia inaequalis: Relationship to Disease Management Strategies. , 2001, Plant disease.

[29]  Trevor Darrell,et al.  DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition , 2013, ICML.

[30]  Bill Johnson,et al.  Fire Blight of Apple Rootstocks , 2009 .

[31]  Pol Coppin,et al.  Detection of biotic stress (Venturia inaequalis) in apple trees using hyperspectral data: Non-parametric statistical approaches and physiological implications , 2007 .

[32]  Alsayed Algergawy,et al.  A Deep Learning-based Approach for Banana Leaf Diseases Classification , 2017, BTW.

[33]  Timnit Gebru,et al.  iCassava 2019Fine-Grained Visual Categorization Challenge , 2019, ArXiv.

[34]  Darko Stefanovic,et al.  Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification , 2016, Comput. Intell. Neurosci..

[35]  Jayme Garcia Arnal Barbedo,et al.  Factors influencing the use of deep learning for plant disease recognition , 2018, Biosystems Engineering.

[36]  Reza Ehsani,et al.  Review: A review of advanced techniques for detecting plant diseases , 2010 .

[37]  Xiaoming Liu,et al.  On developing and enhancing plant-level disease rating systems in real fields , 2016, Pattern Recognit..

[38]  D.W.L. Manktelow,et al.  SPRAYCHECK — a model for evaluating grower timing of black spot (Venturia inaequalis) fungicides in apple orchards , 1998 .

[39]  Anand Singh Jalal,et al.  Apple disease classification using color, texture and shape features from images , 2016, Signal Image Video Process..

[40]  Guangyu Sun,et al.  High Humidity and Age-Dependent Fruit Susceptibility Promote Development of Trichothecium Black Spot on Apple. , 2019, Plant disease.