Cost-sensitive stacked sparse auto-encoder models to detect striped stem borer infestation on rice based on hyperspectral imaging

Abstract Striped stem borer (SSB) usually causes serious damage to rice, and timely detection of SSB infestation is crucial in rice production. Hyperspectral imaging technology has been employed in to detect abiotic and biotic stresses on plant. In this study, the cost-sensitive stacked sparse auto-encoder (SSAE) was proposed to detect the early infestation stage on rice combined with visible/near-infrared hyperspectral imaging technology. The Fisher linear discriminate algorithm (LDA) was modified to quantify the distribution of feature representation of each layer. Multiple structures of the cost-sensitive SSAE were compared and the optimal structure was two depths with width of 6–6 and spares constraint of 0.1. The cost-sensitive SSAE acquired highest total accuracy of 93.44% with full input variables and satisfying total accuracy of 90.98% with reduced input variables, which were superior and stable in comparison with other state-of-art feature extraction and selection methods. These results indicated that the cost-sensitive SSAE has great potential in detecting early SSB infestation based on hyperspectral imaging technology.

[1]  Yi-Zeng Liang,et al.  Application of Competitive Adaptive Reweighted Sampling Method to Determine Effective Wavelengths for Prediction of Total Acid of Vinegar , 2012, Food Analytical Methods.

[2]  S. Z. Hosseini,et al.  Study of Silicon Effects on Plant Growth and Resistance to Stem Borer in Rice , 2012 .

[3]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[4]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[5]  R. A. Leibler,et al.  On Information and Sufficiency , 1951 .

[6]  Ludmila I. Kuncheva,et al.  PCA Feature Extraction for Change Detection in Multidimensional Unlabeled Data , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[7]  Zhang Yi,et al.  Learning a good representation with unsymmetrical auto-encoder , 2015, Neural Computing and Applications.

[8]  Jing Lu,et al.  The Rice Transcription Factor WRKY53 Suppresses Herbivore-Induced Defenses by Acting as a Negative Feedback Modulator of Mitogen-Activated Protein Kinase Activity1 , 2015, Plant Physiology.

[9]  Chu Zhang,et al.  Rice Seed Cultivar Identification Using Near-Infrared Hyperspectral Imaging and Multivariate Data Analysis , 2013, Sensors.

[10]  B. San Segundo,et al.  Expression of the maize proteinase inhibitor (mpi) gene in rice plants enhances resistance against the striped stem borer (Chilo suppressalis): effects on larval growth and insect gut proteinases. , 2005, Plant biotechnology journal.

[11]  Duncan Fyfe Gillies,et al.  A Review of Feature Selection and Feature Extraction Methods Applied on Microarray Data , 2015, Adv. Bioinformatics.

[12]  Jianzhong Wu,et al.  Stacked Sparse Autoencoder (SSAE) for Nuclei Detection on Breast Cancer Histopathology Images , 2016, IEEE Transactions on Medical Imaging.

[13]  Martin Herold,et al.  Spectral resolution requirements for mapping urban areas , 2003, IEEE Trans. Geosci. Remote. Sens..

[14]  W. Cai,et al.  A variable selection method based on uninformative variable elimination for multivariate calibration of near-infrared spectra , 2008 .

[15]  R. Yu,et al.  An ensemble of Monte Carlo uninformative variable elimination for wavelength selection. , 2008, Analytica chimica acta.

[16]  S. Chander,et al.  Simulation of rice brown planthopper [ Nilaparvata lugens (Stal)] damage for determining economic injury levels , 2011 .

[17]  Chu Zhang,et al.  Hyperspectral imaging analysis for ripeness evaluation of strawberry with support vector machine , 2016 .

[18]  M. D. Pathak.  Ecology of Common Insect Pests of Rice , 1968 .

[19]  Y. G. Prasad,et al.  Use of ground based hyperspectral remote sensing for detection of stress in cotton caused by leafhopper (Hemiptera: Cicadellidae) , 2011 .

[20]  Fei Liu,et al.  Non-destructive determination of Malondialdehyde (MDA) distribution in oilseed rape leaves by laboratory scale NIR hyperspectral imaging , 2016, Scientific Reports.

[21]  Josep Peñuelas,et al.  Visible and near-infrared reflectance techniques for diagnosing plant physiological status , 1998 .

[22]  Hui Ye,et al.  Determination and Visualization of pH Values in Anaerobic Digestion of Water Hyacinth and Rice Straw Mixtures Using Hyperspectral Imaging with Wavelet Transform Denoising and Variable Selection , 2016, Sensors.

[23]  N. Aronszajn.  Theory of Reproducing Kernels. , 1950 .

[24]  Yaozong Gao,et al.  Deformable MR Prostate Segmentation via Deep Feature Learning and Sparse Patch Matching , 2016, IEEE Transactions on Medical Imaging.

[25]  Hongxun Yao,et al.  Auto-encoder based dimensionality reduction , 2016, Neurocomputing.

[26]  Wen-Hao Su,et al.  Evaluation of spectral imaging for inspection of adulterants in terms of common wheat flour, cassava flour and corn flour in organic Avatar wheat (Triticum spp.) flour , 2017 .

[27]  Seong-Whan Lee,et al.  Latent feature representation with stacked auto-encoder for AD/MCI diagnosis , 2013, Brain Structure and Function.

[28]  Ning Zhang,et al.  Hyperspectral Image Classification Based on Nonlinear Spectral–Spatial Network , 2016, IEEE Geoscience and Remote Sensing Letters.

[29]  U. Knauer,et al.  Improved classification accuracy of powdery mildew infection levels of wine grapes by spatial-spectral analysis of hyperspectral images , 2017, Plant Methods.

[30]  Qing-Song Xu,et al.  Using variable combination population analysis for variable selection in multivariate calibration. , 2015, Analytica chimica acta.

[31]  Abdul Ahad Buhroo,et al.  Mechanisms of plant defense against insect herbivores , 2012, Plant signaling & behavior.

[32]  Roberto Kawakami Harrop Galvão,et al.  The successive projections algorithm for spectral variable selection in classification problems , 2005 .