Automated Identification of Northern Leaf Blight-Infected Maize Plants from Field Imagery Using Deep Learning.

Northern leaf blight (NLB) can cause severe yield loss in maize; however, scouting large areas to accurately diagnose the disease is time consuming and difficult. We demonstrate a system capable of automatically identifying NLB lesions in field-acquired images of maize plants with high reliability. This approach uses a computational pipeline of convolutional neural networks (CNNs) that addresses the challenges of limited data and the myriad irregularities that appear in images of field-grown plants. Several CNNs were trained to classify small regions of images as containing NLB lesions or not; their predictions were combined into separate heat maps, then fed into a final CNN trained to classify the entire image as containing diseased plants or not. The system achieved 96.7% accuracy on test set images not used in training. We suggest that such systems mounted on aerial- or ground-based vehicles can help in automated high-throughput plant phenotyping, precision breeding for disease resistance, and reduced pesticide use through targeted application across a variety of plant and disease categories.

[1]  A. Hooker,et al.  Measuring the relationship between northern corn leaf blight and yield losses , 1981 .

[2]  R. T. Sherwood.  Illusions in Visual Assessment of Stagonospora Leaf Spot of Orchardgrass , 1983 .

[3]  J. M. Perkins,et al.  Disease development and yield losses associated with northern leaf blight on corn. , 1987 .

[4]  Z. Wicks,et al.  Mapping Components of Partial Resistance to Northern Leaf Blight of Maize Using Reciprocal Translocations , 1992 .

[5]  H. Geiger,et al.  Genes for resistance to northern corn leaf blight in diverse maize populations. , 2000 .

[6]  Erich-Christian Oerke,et al.  Safeguarding production-losses in major crops and the role of crop protection , 2004 .

[7]  T R Gottwald,et al.  Visual Rating and the Use of Image Analysis for Assessing Different Symptoms of Citrus Canker on Grapefruit Leaves. , 2008, Plant disease.

[8]  K.-H. Dammer,et al.  Variable-rate fungicide spraying in real time by combining a plant cover sensor and a decision support system , 2008, Precision Agriculture.

[9]  T R Gottwald,et al.  Comparison of Assessment of Citrus Canker Foliar Symptoms by Experienced and Inexperienced Raters. , 2009, Plant disease.

[10]  R. Nelson,et al.  Resistance loci affecting distinct stages of fungal pathogenesis: use of introgression lines for QTL mapping and characterization in the maize - Setosphaeria turcica pathosystem , 2010, BMC Plant Biology.

[11]  C. Sato,et al.  Influences of Northern Leaf Blight on corn silage fermentation quality, nutritive value and feed intake by sheep. , 2010, Animal science journal = Nihon chikusan Gakkaiho.

[12]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[13]  B. S. Manjunath,et al.  The iPlant Collaborative: Cyberinfrastructure for Plant Biology , 2011, Front. Plant Sci..

[14]  R. Nelson,et al.  In the eye of the beholder: the effect of rater variability and different rating scales on QTL mapping. , 2011, Phytopathology.

[15]  Assessment of yield loss due to turcicum leaf blight of maize caused by Exserohilum turcicum. , 2011 .

[16]  K. P. Pauls,et al.  Application of image analysis in studies of quantitative disease resistance, exemplified using common bacterial blight-common bean pathosystem. , 2012, Phytopathology.

[17]  Jayme Garcia Arnal Barbedo,et al.  Digital image processing techniques for detecting, quantifying and classifying plant diseases , 2013, SpringerPlus.

[18]  Jeffrey W. White,et al.  Development and evaluation of a field-based high-throughput phenotyping platform. , 2013, Functional plant biology : FPB.

[19]  Daniel L. K. Yamins,et al.  Deep Neural Networks Rival the Representation of Primate IT Cortex for Core Visual Object Recognition , 2014, PLoS Comput. Biol..

[20]  Jose A. Jiménez-Berni,et al.  Pheno-Copter: A Low-Altitude, Autonomous Remote-Sensing Robotic Helicopter for High-Throughput Field-Based Phenotyping , 2014 .

[21]  Ethan L. Stewart,et al.  Measuring quantitative virulence in the wheat pathogen Zymoseptoria tritici using high-throughput automated image analysis. , 2014, Phytopathology.

[22]  Jose A. Jiménez-Berni,et al.  Proximal Remote Sensing Buggies and Potential Applications for Field-Based Phenotyping , 2014 .

[23]  S. Sankaran,et al.  Low-altitude, high-resolution aerial imaging systems for row and field crop phenotyping: A review , 2015 .

[24]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[25]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[26]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[27]  Carolyn J. Lawrence-Dill,et al.  The Quest for Understanding Phenotypic Variation via Integrated Approaches in the Field Environment1[OPEN] , 2016, Plant Physiology.

[28]  Damon L. Smith,et al.  Corn yield loss estimates due to diseases in the United States and Ontario, Canada from 2012 to 2015. , 2016 .

[29]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[30]  Marcel Salathé,et al.  Inference of Plant Diseases from Leaf Images through Deep Learning , 2016 .

[31]  John Salvatier,et al.  Theano: A Python framework for fast computation of mathematical expressions , 2016, ArXiv.

[32]  K. Dammer,et al.  Sensor-based variable-rate fungicide application in winter wheat. , 2016, Pest management science.

[33]  Ashutosh Kumar Singh,et al.  Machine Learning for High-Throughput Stress Phenotyping in Plants. , 2016, Trends in plant science.

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