Detection of Black Spot of Rose Based on Hyperspectral Imaging and Convolutional Neural Network

Black spot is one of the seriously damaging plant diseases in China, especially in rose production. Hyperspectral technology reflects both external features and internal structure information of measured samples, which can be used to identify the disease. In this research, both the spectral and image features of two infected roses with black spot were used to train a convolutional neural network (CNN) model. Multiple scattering correction (MSC) and standard normal variable (SNV) methods were applied to preprocess the spectral data. Cropping, median filtering and binarization were pretreatments used on the hyperspectral images. Three CNN models based on Alexnet, VGG16 and neural discriminative dimensionality reduction (NDDR) were evaluated by analyzing the classification accuracy and loss function. The results show that the CNN model based on the fusion of features has higher accuracy. The highest accuracies of detection of blackspot in different roses are 12–26 (100%) and 13–54 (99.95%), applying the NDDR-CNN model. Therefore, this research indicates that the spectral analysis based on CNN can detect black spot of roses, which provides a reference for the detection of other plant diseases, and has favorable research significance as well as prospect for development.

[1]  Changfeng Li,et al.  Rapid and visual detection of Verticillium dahliae using recombinase polymerase amplification combined with lateral flow dipstick , 2020 .

[2]  Reza Ehsani,et al.  Detection and Differentiation between Laurel Wilt Disease, Phytophthora Disease, and Salinity Damage Using a Hyperspectral Sensing Technique , 2016 .

[3]  T. Debener,et al.  The Beast and the Beauty: What Do we know about Black Spot in Roses? , 2019, Critical Reviews in Plant Sciences.

[4]  Hongbin Pu,et al.  Pathogenetic process monitoring and early detection of pear black spot disease caused by Alternaria alternata using hyperspectral imaging , 2019, Postharvest Biology and Technology.

[5]  Yubin Lan,et al.  Early detection of bacterial wilt in peanut plants through leaf-level hyperspectral and unmanned aerial vehicle data , 2020, Comput. Electron. Agric..

[6]  Yiannis Ampatzidis,et al.  Finite Difference Analysis and Bivariate Correlation of Hyperspectral Data for Detecting Laurel Wilt Disease and Nutritional Deficiency in Avocado , 2019, Remote. Sens..

[7]  Yong He,et al.  Hyperspectral imaging for classification of healthy and gray mold diseased tomato leaves with different infection severities , 2017, Comput. Electron. Agric..

[8]  Beth Fallon,et al.  Spectral differentiation of oak wilt from foliar fungal disease and drought is correlated with physiological changes. , 2020, Tree physiology.

[9]  Richard Archer,et al.  Hyperspectral imaging for identification of Zebra Chip disease in potatoes , 2020, Biosystems Engineering.

[10]  Christian Bauckhage,et al.  Metro Maps of Plant Disease Dynamics—Automated Mining of Differences Using Hyperspectral Images , 2015, PloS one.

[11]  A. Nayak,et al.  Simultaneous detection of downy mildew and powdery mildew pathogens on Cucumis sativus and other cucurbits using duplex-qPCR and HRM analysis , 2020, AMB Express.

[12]  Stan C. Hokanson,et al.  Mapping a Novel Black Spot Resistance Locus in the Climbing Rose Brite Eyes™ (‘RADbrite’) , 2018, Front. Plant Sci..

[13]  Thomas Debener,et al.  Morphological characterization of the interaction between Diplocarpon rosae and various rose species , 2005 .

[14]  Shattri Mansor,et al.  Early Detection of Ganoderma Basal Stem Rot of Oil Palms Using Artificial Neural Network Spectral Analysis. , 2017, Plant disease.

[15]  Koushik Nagasubramanian,et al.  Plant disease identification using explainable 3D deep learning on hyperspectral images , 2019, Plant Methods.

[16]  Yanbo Huang,et al.  Detection of Wheat Powdery Mildew by Differentiating Background Factors using Hyperspectral Imaging , 2016 .

[17]  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.

[18]  Xiang-long Meng,et al.  Development and Evaluation of a Loop-mediated Isothermal Amplification Assay for the Rapid Detection and Identification of Pectobacterium carotovorum on Celery in the Field , 2020 .

[19]  Yubin Lan,et al.  Assessment of rice leaf blast severity using hyperspectral imaging during late vegetative growth , 2020, Australasian Plant Pathology.

[20]  J. Behmann,et al.  Hyperspectral signal decomposition and symptom detection of wheat rust disease at the leaf scale using pure fungal spore spectra as reference , 2019, Plant Pathology.

[21]  L. Palou,et al.  First report of Alternaria alternata causing postharvest black spot of persimmon in Spain , 2012, Australasian Plant Disease Notes.

[22]  J. Woodhall,et al.  Development of a TaqMan PCR assay for specific detection and quantification of Pectobacterium brasiliense in potato tubers and soil , 2020, European Journal of Plant Pathology.

[23]  Anne-Katrin Mahlein,et al.  Hyperspectral Sensors and Imaging Technologies in Phytopathology: State of the Art. , 2018, Annual review of phytopathology.