1 Detection of Fusarium Head Blight in Wheat Using a 2 Deep Neural Network and Color Imaging 3

Fusarium head blight (FHB) disease is extensively distributed worldwide. This disease 13 damages grain quality and reduces yield. The detection of this disease in a high throughput way is 14 crucial to planters and breeders. Our study focused on developing a method for processing wheat 15 color images and accurately detecting disease areas using deep learning and image processing 16 techniques. The color images of wheat at the milky stage were collected and processed to construct 17 datasets, which were used to retrain a deep convolutional neural network model using transfer 18 learning. Testing results showed that the model can detect spikes, and the coefficient of 19 determination of the number of spikes between the manual count and the detection was 0.80. The 20 model was assessed, and the mean average precision for the testing dataset was 0.9201. On the 21 basis of the results of spike detection, a new color feature was applied to obtain the gray image of 22 each spike. Then, a modified region growing algorithm was implemented to segment and detect 23 the diseased areas of each spike. Results show that the region growing algorithm performs better 24 than K-means and Otsu’s method in segmenting the FHB disease. Overall, this study demonstrates 25 that deep learning techniques enable the accurate detection of FHB in wheat using color images, 26 and the proposed method can effectively detect spikes and diseased areas, thereby improving the 27 efficiency of FHB detection. 28

[1]  Paul Chen,et al.  Estimation of Fusarium scab in wheat using machine vision and a neural network , 1998 .

[2]  M. Mcmullen,et al.  A Visual Scale to Estimate Severity of Fusarium Head Blight in Wheat , 1998 .

[3]  Z. Kang,et al.  Cytology and ultrastructure of the infection of wheat spikes by Fusarium culmorum , 2000 .

[4]  G. Bai,et al.  Management and resistance in wheat and barley to fusarium head blight. , 2003, Annual review of phytopathology.

[5]  Antonio Torralba,et al.  LabelMe: A Database and Web-Based Tool for Image Annotation , 2008, International Journal of Computer Vision.

[6]  Pierre Gouton,et al.  Texture analysis with statistical methods for wheat ear extraction , 2007, International Conference on Quality Control by Artificial Vision.

[7]  C. Griffey,et al.  Meta-analysis of QTL associated with Fusarium head blight resistance in wheat. , 2009 .

[8]  A. Giebel,et al.  Early detection of Fusarium infection in wheat using hyper-spectral imaging , 2011 .

[9]  J. Murphy,et al.  Digital Image Analysis Method for Estimation of Fusarium‐Damaged Kernels in Wheat , 2014 .

[10]  Jayme Garcia Arnal Barbedo,et al.  Detecting wheat Fusarium head blight using hyperspectral imaging. , 2015 .

[11]  Yong He,et al.  Detection of early blight and late blight diseases on tomato leaves using hyperspectral imaging , 2015, Scientific Reports.

[12]  Zhiguo Cao,et al.  In-field automatic observation of wheat heading stage using computer vision , 2016 .

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

[14]  C. Steinberg,et al.  Visual assessment and computer–assisted image analysis of Fusarium head blight in the field to predict mycotoxin accumulation in wheat grains , 2018, European Journal of Plant Pathology.

[15]  S. Popovski,et al.  Fusarium spp. incidence and DON Contamination in different wheat varieties correlated with the environmental factors , 2017 .

[16]  J. West,et al.  Novel Technologies for the detection of Fusarium head blight disease and airborne inoculum , 2017, Tropical Plant Pathology.

[17]  J. Cai,et al.  Detecting spikes of wheat plants using neural networks with Laws texture energy , 2017, Plant Methods.

[18]  Tony P. Pridmore,et al.  Deep Learning for Multi-task Plant Phenotyping , 2017 .

[19]  Kaiming He,et al.  Mask R-CNN , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[20]  Rebecca L. Whetton,et al.  Hyperspectral measurements of yellow rust and fusarium head blight in cereal crops: Part 1: Laboratory study , 2018 .

[21]  Hong Sun,et al.  Sensors for measuring plant phenotyping: A review , 2018 .

[22]  Rebecca L. Whetton,et al.  Hyperspectral measurements of yellow rust and fusarium head blight in cereal crops: Part 2: On-line field measurement , 2018 .

[23]  P. Zapotoczny,et al.  Classification of Fusarium-infected and healthy wheat kernels based on features from hyperspectral images and flatbed scanner images: a comparative analysis , 2018, European Food Research and Technology.

[24]  Anne-Katrin Mahlein,et al.  Benefits of hyperspectral imaging for plant disease detection and plant protection: a technical perspective , 2018 .

[25]  Hamid Laga,et al.  Detection and analysis of wheat spikes using Convolutional Neural Networks , 2018, Plant Methods.

[26]  Dong Liang,et al.  Wheat Ears Counting in Field Conditions Based on Multi-Feature Optimization and TWSVM , 2018, Front. Plant Sci..

[27]  Shuai Wang,et al.  Classifying Wheat Hyperspectral Pixels of Healthy Heads and Fusarium Head Blight Disease Using a Deep Neural Network in the Wild Field , 2018, Remote. Sens..

[28]  Stefan Thomas,et al.  Quantitative assessment of disease severity and rating of barley cultivars based on hyperspectral imaging in a non-invasive, automated phenotyping platform , 2018, Plant Methods.

[29]  Frédéric Baret,et al.  Ear density estimation from high resolution RGB imagery using deep learning technique , 2019, Agricultural and Forest Meteorology.

[30]  Sergio Benini,et al.  Deep Learning Meets Hyperspectral Image Analysis: A Multidisciplinary Review , 2019, J. Imaging.

[31]  Itamar F. Salazar-Reque,et al.  An Algorithm for Plant Disease Visual Symptom Detection in Digital Images Based on Superpixels , 2019 .

[32]  Hao Zhang,et al.  Identification of grape diseases using image analysis and BP neural networks , 2019, Multimedia Tools and Applications.

[33]  Wei Dong,et al.  SLIC_SVM based leaf diseases saliency map extraction of tea plant , 2019, Comput. Electron. Agric..

[34]  S. Kawamura,et al.  Why RGB Imaging Should be Used to Analyze Fusarium Graminearum Growth and Estimate Deoxynivalenol Contamination , 2019, Methods and protocols.

[35]  Yu Lei,et al.  Early Visual Detection of Wheat Stripe Rust Using Visible/Near-Infrared Hyperspectral Imaging , 2019, Sensors.