Identification of Wheat Yellow Rust Using Spectral and Texture Features of Hyperspectral Images

Abstract

Wheat yellow rust is one of the most destructive diseases in wheat production and significantly affects wheat quality and yield. Accurate and non-destructive identification of yellow rust is critical to wheat production management. Hyperspectral imaging technology has proven to be effective in identifying plant diseases. We investigated the feasibility of identifying yellow rust on wheat leaves using spectral features and textural features (TFs) of hyperspectral images. First, the hyperspectral images were preprocessed, and healthy and yellow rust-infected samples were obtained by creating regions of interest. Second, the extraction of spectral reflectance characteristics and vegetation indices (VIs) were performed from the preprocessed hyperspectral images, and the TFs were extracted using the grey-level co-occurrence matrix from the images transformed by principal component analysis. Third, the successive projections algorithm was employed to choose the optimum wavebands (OWs), and correlation-based feature selection was employed to select the optimal VIs and TFs (those most sensitive to yellow rust and having minimal redundancy between features). Finally, identification models of wheat yellow rust were established using a support vector machine and different features. Six OWs (538, 598, 689, 702, 751, and 895 nm), four VIs (nitrogen reflectance index, photochemical reflectance index, greenness index, and anthocyanin reflectance index), and four TFs (correlation 1, correlation 2, entropy 2, and second moment 3) were selected. The identification models based on the OWs, VIs, and TFs provided overall accuracies of 83.3%, 89.5%, and 86.5%, respectively. The TF results were especially encouraging. The models with the combination of spectral features and TFs exhibited better performance than those using the spectral features or TFs alone. The accuracies of the models with the combined features (OWs and TFs, Vis, and TFs) were 90.6% and 95.8%, respectively. These values were 7.3% and 6.3% higher, respectively, than those of the models using only the OWs or VIs. The model with the combined feature (VIs and TFs) had the highest accuracy (95.8%) and was used to map the yellow rust lesions on wheat leaves with different damage levels. The results showed that the yellow rust lesions on the leaves could be identified accurately. Overall, the combination of spectral features and TFs of hyperspectral images significantly improved the identification accuracy of wheat yellow rust.

A Sustainability Assessment of the Greenseeker N Management Tool: A Lysimetric Experiment on Barley

“…Randelović et al [5] used a machine learning model and vegetation indices extracted by a Unmanned Aerial Vehicles (UAV) device to predict soybean plant density. Guo et al [6] used hyperspectral images to detect wheat Yellow Rust infection, while Sandino et al [7] used remote sensing to detect deterioration by fungal pathogens in forests. In addition, a lot of research has been focused on the use of precision devices to predict crop parameters (biomass, plant nutrient content, yield) [8][9][10][11][12] and nitrogen (N) management [13][14][15][16][17].…”

Detection of Leek Rust Disease under Field Conditions Using Hyperspectral Proximal Sensing and Machine Learning

“…Comparing the leek rust disease detection protocol in this work to that of wheat rust disease reported in the literature, some key differences become apparent [29,35]. First, one of the major constraints in wheat experiments was the absence of a measurement setup that could move the hyperspectral pushbroom sensor over the crop canopy at a continuous pace.…”

“…This resulted in a dataset with the high spatial variability and large sample size normally associated with field trials, e.g. with drone equipment, while maintaining a pixel resolution normally associated with laboratory trials [3,29,35,40,44,53,57,58]. The ability to select ROIs directly from field data and build a hyperspectral library with 'pure' spectra for each class is important, because one of the biggest challenges in disease detection is the translation of results from lab to field scale [1,3,4].…”

“…Several authors have used hyperspectral sensing for detecting rust disease in other arable crops such as wheat and sugarcane [1,3,18,[29][30][31][32][33][34][35]. Looking at recent findings in the literature regarding variable rate crop protection with hyperspectral sensors, it is clear that one of the biggest remaining challenges is disease detection under field conditions.…”

Automatic System for Crop Pest and Disease Dynamic Monitoring and Early Forecasting

“…Then Wheat Rust Index (WRI) is constructed based on Plant Senescence Reflectance Index (PSRI) (1) and Red-edge Vegetation Stress Index (RVSI) (2) to monitor wheat yellow rust, for which WRI (3) could consider wheat growth, chlorophyll content and their variation characteristics. Then, integrated with disease habitat information including land surface temperature (LST, MODIS product), rainfall and wind (meteorological data), also historical data, Disease Index (DI) (4) is constructed for wheat yellow rust habitat monitoring based on our previous teamwork [19][20][21][22][23][24].…”