Abstract:Taking the leaves of Acanthopanax acanthopanax infected with black spot disease as an example, the study of plant disease detection by spectral technology provides a research basis for early screening and precise treatment of medicinal plant diseases. The experiment aimed to realize the supervised classification and identification of plant diseases by hyperspectral imaging technology. The experimental procedure is as follows: First, the leaf samples of Acanthopanax japonicus were collected using the hyperspectral imaging technique. After the spectral data were preprocessed by removing light and dark noise and smoothing, the data dimension was reduced using principal component analysis. Then, a support vector machine (SVM) based on different kernel functions was used to establish a classification model separately. Finally, the overall classification accuracy, Kappa coefficient and other factors were used to evaluate different kernel functions’ influence on the classifier performance. According to the leaf’s surface characteristics, the leaf was divided into four kinds of samples: healthy bright part, healthy dark part, mild disease and severe disease. It can be seen that the healthy sample of Acanthopanax senticosus had a significant peak at 540 nm, and the spectral curve rose sharply at 620~680 nm; while the spectral reflectance of disease samples showed a slow and steady rising trend. The above features could completely distinguish healthy samples with close reflection intensity from serious disease samples on the image. After comparison, it was found that the first four principal components (PC1, PC2, PC3, PC4) have certain differences in the classification results. The main differences were that PC1 contains much information and can better distinguish various samples; PC2 showed a cross-confusion between bright healthy samples and seriously diseased samples; PC3 was a supplement to PC2, which can find mild diseased areas; PC4 contribution rate was only 0.19%, and it could still accurately identify serious diseased areas. The differences of principal component components in showing various sample characteristics can be used to reference complex sample classification. Compared with the classification accuracy of SVM modeling based on different kernel functions, the results showed that the linear kernel function’s recognition process was greatly affected by light intensity reflection. The training accuracy of the Sigmoid kernel function was easily affected by the size of the data set, and there were certain errors in recognition of healthy light or dark and minor diseases. The effect of polynomial kernel function and radial basis kernel function was good, and the accuracy of the polynomial kernel was higher, which was 92.77%. Research showed that the hyperspectral imaging technology could accurately identify the healthy and diseased leaves of Acanthopanax senticosus and provided a new method for the automatic diagnosis of diseases of medicinal plant leaves.
Key words:Visible spectrum; Plant diseases; Support Vector Machine; Kernel function
赵 森,付 芸,崔江南,鲁 烨,杜旭东,李永亮. 高光谱的刺五加黑斑病的早期检测研究[J]. 光谱学与光谱分析, 2021, 41(06): 1898-1904.
ZHAO Sen, FU Yun, CUI Jiang-nan, LU Ye, DU Xu-dong, LI Yong-liang. Application of Hyperspectral Imaging in the Diagnosis of Acanthopanax Senticosus Black Spot Disease. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(06): 1898-1904.
[1] LÜ Dong-mei, YUAN Yuan, ZHAN Zhi-lai(吕冬梅,袁 媛,詹志来). China Journal of Chinese Materia Medica(中国中药杂志), 2014, 39(17): 3413.
[2] Kong Wenwen, Zhang Chu, Huang Weihao, et al. Sensors, 2018, 18(1): 123.
[3] ZHANG Jing-yi, CHEN Jin-chao, FU Xia-ping, et al(张静宜,陈锦超,傅霞萍,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2019, 39(10): 3184.
[4] Zhou Ruiqing, Jin Juanjuan, Li Qingmian, et al. Frontiers in Plant Science, 2019. 9: https://doi.org/10.3389/fpls.2018.01962.
[5] Wang Yujie, Hu Xin, Hou Zhiwei, et al. Journal of the Science of Food and Agriculture, 2018, 98(12): 4659.
[6] Zhang S W, Zhang S B, Zhang C L, et al. Computers and Electronics in Agriculture, 2019, 162: 422.
[7] Li Y H, Luo Z H, Wang F J, et al. Sensors, 2020, 20: 4045.
[8] Zhao Yanru, Yu Keqiang, Li Xiaoli, et al. Scientific Reports, 2016, 6(1): 38878.
[9] Mishra P, Asaari M S M, Herrero-Langreo A, et al. Biosystems Engineering, 2017, 164: 49.
[10] Kant Rama, Joshi Pooja, Bhandari Maneesh S, et al. Forest Pathology, 2020, 50(2):e12584.
[11] GAO Sheng, WANG Qiao-hua, FU Dan-dan, et al(高 升, 王巧华, 付丹丹,等). Acta Optica Sinica(光学学报), 2019, 39(10): 355.
[12] DENG Nai-yang, TIAN Ying-jie(邓乃扬, 田英杰). Support Vector Machines: Theoretical Algorithms and Extensions(支持向量机:理论算法与拓展). Beijing: Science Press(科学出版社), 2016. 92.
[13] Prey L, Von Bloh M, Schmidhalter U, et al. Sensors, 2018, 18(9): 2931.
[14] Foody G M. Remote Sensing of Environment, 2020, 239: 111630.