Early Detection of Northern Corn Leaf Blight Using Hyperspectral Images Combined With One-Dimensional Convolutional Neural Networks
LU Yang1*, GU Fu-qian1, GU Ying-nan2*, XU Si-yuan1, WANG Peng3, 4
1. College of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, China
2. Institute of Agricultural Remote Sensing and Information, Heilongjiang Academy of Agricultural Sciences/Postdoctoral Research Workstation of Heilongjiang Academy of Agricultural Sciences, Harbin 150086, China
3. Heilongjiang Provincial Key Laboratory of Networking and Intelligent Control, Northeast Petroleum University, Daqing 163318, China
4. Sanya Research Institute of Offshore Oil and Gas, Northeast Petroleum University, Sanya 572024, China
Abstract：Northern corn leaf blight (NCLB) occurs in major maize-producing regions globally, leading to a reduction in both maize quality and yield. Disease identification typically occurs when lesions are more obvious, making it challenging to prevent and control the disease promptly. This study proposes a one-dimensional convolutional neural network (1DCNN) model for early disease detection using hyperspectral imaging. In this research, NCLB was selected as the target disease. After manual inoculation, maize leaves at the silking stage were used for experiments, when lesions had just begun to appear, but the disease could not yet be visually identified. First, hyperspectral images were captured using the SOC710E spectrometer, and spectral data of both healthy and NCLB-infected maize leaves were obtained by selecting regions of interest. Four spectral preprocessing methods Savitzky-Golay smoothing (SG), multiplicative scatter correction (MSC), standard normal variate transformation (SNV), and detrending (DT) were applied to remove noise from the spectral data. Supervised learning algorithms, random forest (RF) and K-nearest neighbors (KNN), were employed for hyperspectral image classification, with accuracy as the evaluation metric. The results indicated that MSC was the optimal preprocessing method, achieving prediction accuracies of 88.13% and 86.26% for the RF and KNN models, respectively. Next, a competitive adaptive reweighted sampling (CARS) algorithm was applied to extract characteristic wavenumbers from the maize leaf spectral data, reducing the original 260 wavenumbers to 48 selected features. Finally, a 1DCNN deep learning model was developed for classification, achieving an accuracy of 99.61%. Compared with traditional classification models such as KNN, RF, partial least squares discriminant analysis (PLS-DA), backpropagation neural network (BP), and support vector machines (SVM), the proposed model improved recognition accuracy by 5.94%, 6.88%, 6.48%, 8.27%, 12.12%, respectively. These findings demonstrate that combining hyperspectral technology with deep learning models provides a new approach and method for early detection of maize diseases, enhancing the accuracy and timeliness of disease recognition.
[1] DAI Rui-xi, XU Wei-ping（代瑞熙，徐伟平）. Agricultural Outlook（农业展望）, 2022, 18(3): 41.
[2] SHI Jie, LIN Shuang-shuang, LUO Jian-gang, et al（施 杰，林双双，罗建刚，等）. Jiangsu Agricultural Sciences（江苏农业科学）, 2023, 51(24): 175.
[3] LIU Tan, LI Zi-mo, FENG Shuai, et al（刘 潭，李子默，冯 帅, 等）. Transactions of the Chinese Society for Agricultural Machinery（农业机械学报）, 2023, 54(11): 208.
[4] SHEN Yan-yan, ZHAO Yu-tao, CHEN Geng-shen, et al（沈艳艳，赵玉涛，陈庚申，等）. Smart Agriculture（智慧农业）, 2024, 6(2): 28.
[5] Bai Y, Nie C, Liu S, et al. Computers and Electronics in Agriculture, 2024, 225: 109350.
[6] Liu X, Meng K, Zhang K, et al. Frontiers in Plant Science, 2024, 15: 1434163.
[7] Cen Y, Huang Y, Hu S, et al. Remote Sensing, 2022, 14(12): 2882.
[8] LIANG Wan-jie, FENG Hui, JIANG Dong, et al（梁万杰，冯 辉，江 东，等）. Spectroscopy and Spectral Analysis（光谱学与光谱分析）, 2023, 43(7): 2220.
[9] WANG Yuan-yuan, LIN Jian, WANG Shan（王圆圆，林 建，王 姗）. Jiangsu Agricultural Sciences（江苏农业科学）, 2023, 51(16): 147.
[10] ZHAN Shu-lin, HUANG Ming-qing, CHEN Lin, et al（詹术霖，黄明清，陈 霖，等）. Journal of Fuzhou University(Natural Science Edition)[福州大学学报（自然科学版）], 2023, 51(6): 879.
[11] WANG Zhen, LI Ying-xue, WU Fang, et al（王 震，李映雪，吴 芳，等）. Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2024, 40(4): 166.
[12] LI Miao-yu, FU Ying（李妙宇，付 莹）. Journal of Image and Graphics（中国图象图形学报），2023, 28(12): 3922.
[13] XING Yu-qi, ZHOU Jia, HUANG Wei-li, et al（邢瑜琪，周 佳，黄卫利，等）. Acta Agriculturae Boreali-occidentalis Sinica（西北农业学报）, 2023, 32(11): 1842.
[14] Nayak S, Panda M, Palai G. Optik，2020, 212: 164675.
[15] ZHOU Min, CUI Zhi-hang, SHAN Lei（周 敏，崔志航，单 蕾）. Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2023, 39(14): 200.
[16] ZHU Zhi-hui, YE Zi-fan, HE Yu-ting, et al（祝志慧，叶子凡，何昱廷，等）. Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2022, 38(18): 308.
[17] GUO Jun-xian, ZHANG Zhen-zhen, HAN Jing, et al（郭俊先，张振振，韩 景, 等）. Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2023, 39(21): 152.
[18] YANG Qing, ZHANG Ya-jing, ZHENG Dan, et al（杨 青，张雅静，郑 丹，等）. Journal of Food Safety & Quality（食品安全质量检测学报）, 2023, 14(17): 95.
[19] SUN Xiao-rong, ZHANG Chen-guang, LIU Cui-ling, et al（孙晓荣，张晨光，刘翠玲，等）. Food Science and Technology（食品科技）, 2023, 48(11): 246.
[20] YUAN Shi-bo, ZHAO Yao-dong, LI Peng, et al（袁士博，赵耀东，李 鹏，等）. Electronic Information Warfare Technology（电子信息对抗技术）, 2023, 38(5): 21.
