Abstract

With the popularization of close planting and high fertilizer technology, the occurrence of Curvularia lunata has been increasing year after year and has caused very serious harm. Early identification of the disease is of great significance to reduce environmental pollution and improve the quality of crops. To identify the disease quickly and without damaging the crops, we studied maize leaves and propose a method of early identification of Curvularia lunata based on hyperspectral imaging technology. The leaves used in the experiment were inoculated in vitro, and at 11 points in time after inoculation (2, 4, 6, 8, 12, 24, 32, 40, 48, 60, and 72 h), we collected hyperspectral images of normal leaves and inoculated leaves in the visible and near-infrared bands. By comparative analysis between the two types of leaves, small chlorotic spots were found after inoculation within 48 hours; the spectral information has obvious differences. Based on the mixed distance method, we find that the best bands for identification of Curvularia lunata are 465.1, 550.7, and 681.4 nm. We used these bands as the input quantity and built a backpropagation neural network detection model. We used the model to experiment on the samples (160 samples for each point in time mentioned above), and the accuracy rate for identification of inoculated leaves is above 97.5%. The results show that the backpropagation neural network detection model with characteristic bands can identify Curvularia lunata quickly and nondestructively. It provides a new method for early detection of maize diseases.

© 2018 Optical Society of America

Using continous wavelet analysis for monitoring wheat yellow rust in different infestation stages based on unmanned aerial vehicle hyperspectral images

Qiong Zheng, Wenjiang Huang, Huichun Ye, Yingying Dong, Yue Shi, and Shuisen Chen
Appl. Opt. 59(26) 8003-8013 (2020)

Classification, identification, and growth stage estimation of microalgae based on transmission hyperspectral microscopic imaging and machine learning

Zhanpeng Xu, Yiming Jiang, Jiali Ji, Erik Forsberg, Yuanpeng Li, and Sailing He
Opt. Express 28(21) 30686-30700 (2020)

Spectral-spatial feature-based neural network method for acute lymphoblastic leukemia cell identification via microscopic hyperspectral imaging technology

Qian Wang, Jianbiao Wang, Mei Zhou, Qingli Li, and Yiting Wang
Biomed. Opt. Express 8(6) 3017-3028 (2017)

Identification of cancerous gastric cells based on common features extracted from hyperspectral microscopic images

Siqi Zhu, Kang Su, Yumeng Liu, Hao Yin, Zhen Li, Furong Huang, Zhenqiang Chen, Weidong Chen, Ge Zhang, and Yihong Chen
Biomed. Opt. Express 6(4) 1135-1145 (2015)

Dual-type hyperspectral microscopic imaging for the identification and analysis of intestinal fungi

Sifan Lin, Xiaolin Bi, Siqi Zhu, Hao Yin, Zhen Li, and Chenqiang Chen
Biomed. Opt. Express 9(9) 4496-4508 (2018)