Detecting the Early Stage of Phaeosphaeria Leaf Spot Infestations in Maize Crop Using In Situ Hyperspectral Data and Guided Regularized Random Forest Algorithm

Phaeosphaeria leaf spot (PLS) is considered one of the major diseases that threaten the stability of maize production in tropical and subtropical African regions. The objective of the present study was to investigate the use of hyperspectral data in detecting the early stage of PLS in tropical maize. Field data were collected from healthy and the early stage of PLS over two years (2013 and 2014) using a handheld spectroradiometer. An integration of a newly developed guided regularized random forest (GRRF) and a traditional random forest (RF) was used for feature selection and classification, respectively. The 2013 dataset was used to train the model, while the 2014 dataset was used as independent test dataset. Results showed that there were statistically significant differences in biochemical concentration between the healthy leaves and leaves that were at an early stage of PLS infestation. The newly developed GRRF was able to reduce the high dimensionality of hyperspectral data by selecting key wavelengths with less autocorrelation. These wavelengths are located at 420 nm, 795 nm, 779 nm, 1543 nm, 1747 nm, and 1010 nm. Using these variables (), a random forest classifier was able to discriminate between the healthy maize and maize at an early stage of PLS infestation with an overall accuracy of 88% and a kappa value of 0.75. Overall, our study showed potential application of hyperspectral data, GRRF feature selection, and RF classifiers in detecting the early stage of PLS infestation in tropical maize.

[1]  S. Chakraborty,et al.  Climate change: potential impact on plant diseases. , 2000, Environmental pollution.

[2]  P. Tongoona,et al.  Combining ability analysis for Phaeosphaeria leaf spot resistance and grain yield in tropical advanced maize inbred lines , 2011 .

[3]  A. S. Ferreira,et al.  Reaction of maize inbred lines to the bacterium Pantoea ananas isolated from Phaeosphaeria leaf spot lesions , 2002 .

[4]  Mahesh Pal,et al.  Random forest classifier for remote sensing classification , 2005 .

[5]  Pablo J. Zarco-Tejada,et al.  Early Detection and Quantification of Verticillium Wilt in Olive Using Hyperspectral and Thermal Imagery over Large Areas , 2015, Remote. Sens..

[6]  M. Carson.  Yield Loss Potential of Phaeosphaeria Leaf Spot of Maize Caused by Phaeosphaeria maydis in the United States. , 2005, Plant disease.

[7]  M. S. Moran,et al.  Opportunities and limitations for image-based remote sensing in precision crop management , 1997 .

[8]  J. E. F. Figueiredo,et al.  ETIOLOGY OF PHAEOSPHAERIA LEAF SPOT DISEASE OF MAIZE , 2013 .

[9]  George P. Petropoulos,et al.  Hyperion hyperspectral imagery analysis combined with machine learning classifiers for land use/cover mapping , 2012, Expert Syst. Appl..

[10]  M. Cho,et al.  A new technique for extracting the red edge position from hyperspectral data: The linear extrapolation method , 2006 .

[11]  Andrew K. Skidmore,et al.  Predicting foliar biochemistry of tea (Camellia sinensis) using reflectance spectra measured at powder, leaf and canopy levels , 2013 .

[12]  C.A.J.M. de Bie,et al.  Comparative performance analysis of agro-ecosystems , 2000 .

[13]  Onisimo Mutanga,et al.  Discriminating the early stages of Sirex noctilio infestation using classification tree ensembles and shortwave infrared bands , 2011 .

[14]  D. Raes,et al.  Agro-climatic suitability mapping for crop production in the Bolivian Altiplano: A case study for quinoa , 2006 .

[15]  Houtao Deng,et al.  Guided Random Forest in the RRF Package , 2013, ArXiv.

[16]  John J. Read,et al.  Canopy reflectance in cotton for growth assessment and lint yield prediction , 2007 .

[17]  D. Mulla.  Twenty five years of remote sensing in precision agriculture: Key advances and remaining knowledge gaps , 2013 .

[18]  J. Benhin,et al.  South African crop farming and climate change: An economic assessment of impacts , 2008 .

[19]  O. Mutanga,et al.  Discriminating indicator grass species for rangeland degradation assessment using hyperspectral data resampled to AISA Eagle resolution , 2012 .

[20]  Bhupinder Singh,et al.  Potential applications of remote sensing in horticulture—A review , 2013 .

[21]  Elfriede Penz,et al.  Zeileis Conditional Variable Importance for Random Forests , 2015 .

[22]  M. S. Moran,et al.  Remote Sensing for Crop Management , 2003 .

[23]  M. Carson.  Vulnerability of U.S. Maize Germ Plasm to Phaeosphaeria Leaf Spot. , 1999, Plant disease.

[24]  Mahesh Pal,et al.  Margin-based feature selection for hyperspectral data , 2009, Int. J. Appl. Earth Obs. Geoinformation.

[25]  Achim Zeileis,et al.  BMC Bioinformatics BioMed Central Methodology article Conditional variable importance for random forests , 2008 .

[26]  M. Senior,et al.  Quantitative Trait Loci Conditioning Resistance to Phaeosphaeria Leaf Spot of Maize Caused by Phaeosphaeria maydis. , 2005, Plant disease.

[27]  George C. Runger,et al.  Gene selection with guided regularized random forest , 2012, Pattern Recognit..

[28]  Mike Crawford,et al.  The Social Functioning Questionnaire: A Rapid and Robust Measure of Perceived Functioning , 2005, The International journal of social psychiatry.

[29]  Meirelles,et al.  Detection of a Bacterium Associated with a Leaf Spot Disease of Maize in Brazil , 2001 .

[30]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[31]  Lee-Ann Jaykus,et al.  Climate change and food safety: A review , 2010 .

[32]  Rongling Wu,et al.  Mapping genes for plant structure, development and evolution: functional mapping meets ontology. , 2010, Trends in genetics : TIG.

[33]  S. Ustin,et al.  Detection of stress in tomatoes induced by late blight disease in California, USA, using hyperspectral remote sensing , 2003 .

[34]  P. Groves,et al.  Methodology For Hyperspectral Band Selection , 2004 .

[35]  George C. Runger,et al.  Feature selection via regularized trees , 2012, The 2012 International Joint Conference on Neural Networks (IJCNN).

[36]  K. Dybkær,et al.  Primary testicular diffuse large B-cell lymphoma displays distinct clinical and biological features for treatment failure in rituximab era: a report from the International PTL Consortium , 2016, Leukemia.

[37]  M. Bänziger,et al.  Challenges of the maize seed industry in eastern and southern Africa: A compelling case for private–public intervention to promote growth , 2010 .

[38]  R. Schulze,et al.  An assessment of sustainable maize production under different management and climate scenarios for smallholder agro-ecosystems in KwaZulu-Natal, South Africa , 2006 .

[39]  O. Mutanga,et al.  Discriminating the papyrus vegetation (Cyperus papyrus L.) and its co-existent species using random forest and hyperspectral data resampled to HYMAP , 2012 .

[40]  Georg Bareth,et al.  Investigation of Leaf Diseases and Estimation of Chlorophyll Concentration in Seven Barley Varieties Using Fluorescence and Hyperspectral Indices , 2013, Remote. Sens..

[41]  C. L. Souza,et al.  QTL mapping for reaction to Phaeosphaeria leaf spot in a tropical maize population , 2009, Theoretical and Applied Genetics.

[42]  Sandra Lowe,et al.  Classification Methods For Remotely Sensed Data , 2016 .

[43]  Pai-Hui Hsu,et al.  Feature extraction of hyperspectral images using wavelet and matching pursuit , 2007 .

[44]  P. Zimba,et al.  Detecting Sugarcane yellow leaf virus infection in asymptomatic leaves with hyperspectral remote sensing and associated leaf pigment changes. , 2010, Journal of virological methods.

[45]  Russell G. Congalton,et al.  Assessing the accuracy of remotely sensed data : principles and practices , 1998 .

[46]  Hao Wu,et al.  An effective feature selection method for hyperspectral image classification based on genetic algorithm and support vector machine , 2011, Knowl. Based Syst..

[47]  Onisimo Mutanga,et al.  Spectral resampling based on user-defined inter-band correlation filter: C3 and C4 grass species classification , 2013, Int. J. Appl. Earth Obs. Geoinformation.

[48]  Onisimo Mutanga,et al.  Potential utility of the spectral red-edge region of SumbandilaSat imagery for assessing indigenous forest structure and health , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[49]  B. Vivek,et al.  Gene action determining Phaeosphaeria leaf spot disease resistance in experimental maize hybrids , 2007 .

[50]  André Stumpf,et al.  Combining Random Forests and object-oriented analysis for landslide mapping from very high resolution imagery , 2011 .

[51]  Jingfeng Huang,et al.  Characterizing and Estimating Fungal Disease Severity of Rice Brown Spot with Hyperspectral Reflectance Data , 2008 .

[52]  Malik Braik,et al.  Fast and Accurate Detection and Classification of Plant Diseases , 2011 .

[53]  R. C. Muchow,et al.  Effect of nitrogen on the time-course of sucrose accumulation in sugarcane , 1996 .