Rapid Diagnosis of Sound, Yellow and Citrus Greening Leaves with Hyperspectral Imaging
SUN Xu-dong1, LIU Yan-de1*, XIAO Huai-chun1, ZHANG Zhi-cheng1, LI Ze-min1, Lü Qiang2
1. School of Mechatronics & Vehicle Engineering, East China Jiaotong University, Nanchang 330013, China 2. Citrus Research Institute, Chinese Academy of Agricultural Sciences, Chongqing 400712, China
Abstract：The feasibility was investigated for identifying sound, yellow and citrus greening leaves of navel orange trees based on hyperspectral imaging combined with correlation analysis and discriminant partial least square (DPLS) methods. The hyperspectral data of sound, yellow and citrus greening leaves were recorded in the wavelength range of 374.28～1 016.89 nm. Two regions of interest (ROI) were marked symmetrically on both sides along main veins with an ellipse of major axis of 60 pixels and minor axis of 30 pixels. The average reflectance spectrum was extracted from ROI regions. A pair wavelengths of 502.79 and 374.28 nm were chosen with correlation analysis method in the wavelength range of 374.28～1 016.89 nm. The classification model was developed with the peak ratio of the pair wavelengths. This model was effective to sound leaves with the recognition accuracy of 1.7% but yellow and citrus greening leaves. The DPLS model was employed with the preprocessing spectra of second derivative and Savitzky-Golay smoothing. The recognition accuracy of this model was 100% for citrus greening leaves and yellow ones. The number of latent variables (LVs) was optimized with the leave one out cross validation method. The optimal LVs, correlation coefficient and standard error of calibration of the DPLS model were 17, 0.96 and 0.13, respectively. The correction classification rate of the DPLS model was 100% for yellow leaves and citrus greening ones. Two-step method of the peak ratio models combination with the DPLS was proposed for identifying sound, yellow and citrus greening leaves. The new samples were applied to evaluation the classification ability of the two-step method, which included sound leaves of 10, citrus greening leaves of 10 and yellow leaves of 10. The correction classification rate reached 96.7%. Experimental results showed that it was feasible to identify sound, yellow and citrus greening leaves by hyperspectral imaging coupled with the peak ratio and DPLS models.
Key words：Hyperspectral;Citrus greening;Partial least square;Correlation analysis;Yellow
[1] BOVé J M. Phytoparasitica, 2014, 42: 579. [2] Deng X L, Gao Y D, Chen J C, et al. Journal of Integrative Agriculture, 2012, 11(3): 424. [3] Bove J M. Journal of Plant Pathology, 2006, 88(1): 7. [4] LI Xiu-hua, LI Min-zan, LEE W S, et al(李修华，李民赞，LEE W S，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2014，34(6)：1553. [5] LI X H, LEE W S, LI M Z, et al. Computers and Electronics in Agriculture, 2012, 83: 32. [6] MA Hao, JI Hai-yan, LEE W S(马 淏，吉海彦，LEE W S). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2014, 34(10): 2713. [7] MEI Hui-lan, DENG Xiao-ling, HONG Tian-sheng, et al(梅慧兰，邓小玲，洪添胜，等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2014, 30(9): 140. [8] Garcia-Ruiz F, Sankaran S, Maja J M, et al. Computers and Electronics in Agriculture, 2013, 91: 106. [9] Sankaran S, Ehsani R. Crop Protection, 2011, 30: 1508. [10] Mishra A, Karimi D, Ehsani R, et al. Biosystems Engineering, 2011, 110: 302. [11] Pourreza A, Lee W S, Etxeberria E, et al. Biosystems Engineering, 2015, 130: 13. [12] Sankaran S, Ehsani R, Etxeberria E. Talanta, 2010, 83: 574. [13] Lee K, Kang S, Delwiche S R, et al. Sens. & Instrument. Food. Qual., 2008, 2: 90.
