DOI: 10.17586/1023-5086-2020-87-04-28-35

Classification of maize leaf diseases based on hyperspectral imaging technology

Deng, Hanbing, Miao, Teng, Song, Kai, Song, Ping, Xiao, Yang, Xu, Jing, Zhou, Yuncheng

Publication in Journal of Optical Technology

For Russian citation (Opticheskii Zhurnal):

Jing Xu, Teng Miao, Yuncheng Zhou, Yang Xiao, Hanbing Deng, Ping Song, Kai Song Classification of maize leaf diseases based on hyperspectral imaging technology (Классификация заболеваний листьев маиса с использованием технологии гиперспектральных изображений) [на англ. яз.] // Оптический журнал. 2020. Т. 87. № 4. С. 28–35. http://doi.org/10.17586/1023-5086-2020-87-04-28-35

Jing Xu, Teng Miao, Yuncheng Zhou, Yang Xiao, Hanbing Deng, Ping Song, Kai Song Classification of maize leaf diseases based on hyperspectral imaging technology (Классификация заболеваний листьев маиса с использованием технологии гиперспектральных изображений) [in English] // Opticheskii Zhurnal. 2020. V. 87. № 4. P. 28–35. http://doi.org/10.17586/1023-5086-2020-87-04-28-35

For citation (Journal of Optical Technology):

Jing Xu, Teng Miao, Yuncheng Zhou, Yang Xiao, Hanbing Deng, Ping Song, and Kai Song, "Classification of maize leaf diseases based on hyperspectral imaging technology," Journal of Optical Technology. 87(4), 212-217 (2020). https://doi.org/10.1364/JOT.87.000212

Abstract:

Curvularia lunata and aureobasidium zeae are the main leaf diseases of maize in Northeast China. Meanwhile, the two diseases are similar and difficult to distinguish. In order to diagnose diseases correctly, a diagnostic method based on hyperspectral imaging technology for curvularia lunata and aureobasidium zeae was proposed. The experimental leaves were inoculated in vivo, and hyperspectral imaging system was used to collect the hyperspectral image data of the leaves of the Curvularia lunata, the leaves of aureobasidium zeae and the normal leaves at the near-infrared band. By analyzing the spectral characteristics of chlorotic spots and normal leaves, it is found that there is a significant difference in spectral information between chlorotic spots and normal leaves. Then, by means of confidence interval estimation and significance test, the characteristic bands of curvularia lunata and aureobasidium zeae can be distinguished. Finally, the classification model of support vector machine is established based on the characteristic bands. The results show that the 10 characteristic bands of 412.7 nm, 416.3 nm, 421.2 nm, 465.1 nm, 484.8 nm, 580.9 nm, 615 nm, 640.5 nm, 676.2 nm, 880.8 nm can be used to distinguish between curvularia lunata and aureobasidium zeae spectral characteristics of the disease, and the support vector machine classification model established by the above band is used for 288 samples. The accuracy rate was 96.7%. These results provide a theoretical basis and technical method for rapid and nondestructive diagnosis of curvularia lunata and aureobasidium zeae.

Keywords:

curvularia lunata, aureobasidium zeae, confidence interval estimation, significance test, hyperspectral imaging technology, support vector machine

Acknowledgements:

This research was financially supported by Natural Science Foundation of China (31501217, 31601218, 31701318) and Beijing Municipal Natural Science Foundation (Grant No.4162028).

OCIS codes: 100.4145, 150.1135, 100.0100, 300.6280, 300.6550, 000.5490

References:

1. Dai F.C., Wang X.M., Zhu Z.D. et al. Curvularia leaf spot of maize pathogens and varietal resistance // Acta. Phytopathologica. Sinica. 1998. V. 28(2). P. 123–129.
2. Chen L.M., Dou S.H., Lv Z.M. et al. Studies on some epidemic links of maize curvularia leaf spot // J. Maize. Sci. 2006. V. 14(4). P. 148–150, 154.
3. Sun J.Y., Xiao S.Q., Lu Y.Y. Isolation, identification and biological characteristics of Aureobasidium zeae in Liaoning Province // Journal of Plant Protection. 2015. V. 42(6). P. 927–934.
4. Meng L.M., Su Q.F., Jia J. et al. Identification of the Pathogen of corn Northern Anthrax and its growth affecting factors // J. Maize. Sci. 2016. V. 24(4). P. 160–165.
5. Kirshnan P., Alexander J.D., Butler B.J. et al. Reflectance technique for predicting soil organic matte // Soil sci. soc. Am. J. 1980. V. 44(6). P. 1282–1285.
6. Dalal R.C., Herry R.J. Simultanous determination of moisture, organic carbon, and total nitrogen by near infrared reflectance spectrophotometry // Soil sci. soc. Am. J. 1986. V. 50(1). P. 120–123.
7. Sudduth K.A., Hummel L.W. Evaluation of reflectance methods for soil organic matter sensing // Transactions of the ASAE. 1991. V. 34(4). P. 1900–1909.
8. Sudduth K.A., Hummel L.W. Portable near-infrared spectrophoto meter for rapid soil analysis // Transactions of the ASAE. 1993. V. 36(1). P. 185–193.
9. Sudduth K.A., Hummel L.W. Soil organic matter, CEC, an moisture sensing with a portable NIR spectrophotometer // Transactions of the ASAE. 1993. V. 36(6). P. 1571–1582.
10. Sudduth K.A., Hummel L.W. Geographic operating range evaluation of a NIR soil sensor // Transactions of the ASAE. 1996. V. 39(5). P. 1599–1604.
11. Chen P.F., Liu L.Y., Wang J.H. et al. Real-time analysis of soil N and P with near infrared diffuse reflectance spectroscop // Spectrosc. Spect. Anal. 2008. V. 28(2). P. 295–298.
12. Lichtenthaler H.K., Langsdorf G., Buschmann C. Multicolor fluorescence images and fluorescence ratio images of green apples at harvest and during storage // Israel Journal of Plant Sciences. 2012. V. 60(1–2). P. 97–106.
13. He Y., Li X.L., Shao Y.N. Discrimination of varieties of apple using near infrared spectra based on principal component analysis and artificial neural network model // Spectrosc. Spect. Anal. 2006. V. 26(5). P. 850–853.
14. Liu Y.D., Chen X.M., Sun X.D. Nondestructive measurement of vitamin C in Nanfeng tangerine by visible/near-infrared diffuse reflectance spectroscopy // Spectrosc. Spect. Anal. 2008. V. 28(10). P. 2318–2320.
15. Yao X., Tian Y.C., Ni J. et al. Estimation of leaf pigment concentration in rice by near infrared reflectance spectroscopy // Chinese Journal of Analytical Chemistry. 2012. V. 40(4). P. 589–595.
16. Wang F., Li Y.Y., Peng Y.K. et al. Determination of Tomato’s SSC and TS based on diffuse transmittance spectroscopy // Spectrosc. Spect. Anal. 2016. V. 36(10). P. 3185–3189.
17. Li H.Q., Sun H., Li M.Z. Detection of vitamin C content in head cabbage based on visible/near-infrared spectroscopy // Transaction of the CSAE. 2018. V. 34(8). P. 269–275.
18. Liu Z.Y., Wu H.F., Huang J.F. Application of neural networks to discriminate fungal infection levels in rice panicles using hyperspectral reflectance and principal components analysis // Computers and Electronics in Agriculture. 2010. V. 72(2). P. 99–106.
19. Jones C.D., Jones J.B., Lee W.S. Diagnosis of bacterial spot of tomato using spectral signatures // Computers and Electronics in Agriculture. 2010. V. 74(2). P. 329–335.
20. Zhang J.C., Pu R.L., Wang J.H. et al. Detecting powdery mildew of winter wheat using leaf level hyperspectral measurements // Computer and Electronics in Agriculture. 2012. V. 85(1). P. 13–23.
21. Feng L., Chai R.Y., Sun G.M. et al. Identification and classification of rice leaf blast based on multi-spectral imaging sensor // Spectrosc. Spect. Anal. 2009. V. 29(10). P. 2730–2733.
22. Sun G.M., Yang K.S., Zhang C.Q. et al. Identification of barley scab based on multi-spectral imaging technology // Transaction of the CSAE. 2009. V. 25 (S2). P. 204–207.
23. Feng W., Wang X.Y., Song X. et al. Hyperspectral estimation of canopy chlorophyll density in winter wheat under stress of powdery mildew // Transaction of the CSAE. 2013. V. 29(13). P. 114–123.
24. Xu J., Miao T., Zhou Y.C. et al. Early identification of Curvularia lunata based on hyperspectral imaging // Journal of Optical Technology. 2018. V. 85(7). P. 432–436.