Lodging stress seriously affects the yield, quality, and mechanical harvesting of maize, and is a major natural disaster causing maize yield reduction. The aim of this study was to obtain light detection and ranging (LiDAR) data of lodged maize using an unmanned aerial vehicle (UAV) equipped with a RIEGL VUX-1UAV sensor to analyze changes in the vertical structure of maize plants with different degrees of lodging, and thus to use plant height to quantitatively study maize lodging. Based on the UAV-LiDAR data, the height of the maize canopy was retrieved using a canopy height model to determine the height of the lodged maize canopy at different times. The profiles were analyzed to assess changes in maize plant height with different degrees of lodging. The differences in plant height growth of maize with different degrees of lodging were evaluated to determine the plant height recovery ability of maize with different degrees of lodging. Furthermore, the correlation between plant heights measured on the ground and LiDAR-estimated plant heights was used to verify the accuracy of plant height estimation. The results show that UAV-LiDAR data can be used to achieve maize canopy height estimation, with plant height estimation accuracy parameters of R² = 0.964, RMSE = 0.127, and nRMSE = 7.449%. Thus, it can reflect changes of plant height of lodging maize and the recovery ability of plant height of different lodging types. Plant height can be used to quantitatively evaluate the lodging degree of maize. Studies have shown that the use of UAV-LiDAR data can effectively estimate plant heights and confirm the feasibility of LiDAR data in crop lodging monitoring.

Lodging is one of the main factors affecting the quality and yield of crops. Timely and accurate determination of crop lodging grade is of great significance for the quantitative and objective evaluation of yield losses. The purpose of this study was to analyze the monitoring ability of a multispectral image obtained by an unmanned aerial vehicle (UAV) for determination of the maize lodging grade. A multispectral Parrot Sequoia camera is specially designed for agricultural applications and provides new information that is useful in agricultural decision-making. Indeed, a near-infrared image which cannot be seen with the naked eye can be used to make a highly precise diagnosis of the vegetation condition. The images obtained constitute a highly effective tool for analyzing plant health. Maize samples with different lodging grades were obtained by visual interpretation, and the spectral reflectance, texture feature parameters, and vegetation indices of the training samples were extracted. Different feature transformations were performed, texture features and vegetation indices were combined, and various feature images were classified by maximum likelihood classification (MLC) to extract four lodging grades. Classification accuracy was evaluated using a confusion matrix based on the verification samples, and the features suitable for monitoring the maize lodging grade were screened. The results showed that compared with a multispectral image, the principal components, texture features, and combination of texture features and vegetation indices were improved by varying degrees. The overall accuracy of the combination of texture features and vegetation indices is 86.61, and the Kappa coefficient is 0.8327, which is higher than that of other features. Therefore, the classification result based on the feature combinations of the UAV multispectral image is useful for monitoring of maize lodging grades.

Maize crop lodging is a recurrent phenomenon which results in significant reduction of grain yield and quality in addition to the impediment of mechanical harvesting. The large-scale monitoring of maize crop lodging is important for production policy adjustment and agricultural insurance compensation. In this article, we derived a variety of features from multitemporal GaoFen-1 (GF-1) images before and after maize crop lodging. We screened the most sensitive features of the spectrum, texture, and vegetation index to monitor maize crop lodging. The recursive feature elimination method based on cross-validation and mutual information were compared to obtain the optimal feature combination for monitoring the lodging extents of maize crop. The random forest classifier was used to classify the lodging extents. The results showed that the most sensitive features of the spectrum, texture, and vegetation indices of lodging extents included the difference of reflectance in blue, green, and red bands, the difference of normalized difference vegetation index, the difference of ratio vegetation index, the difference of enhanced vegetation index difference, the difference of mean value of blue band, the difference of mean value of green band, and the difference of mean value of red band. The total accuracy of lodging extents classification was 87.50%, and the Kappa coefficient was 0.83 for testing samples. Based on multiple features derived from GF-1 images before and after lodging, the lodging extents of maize crop can be monitored on a large scale.

Lodging stress influences the yield, quality, and mechanical harvesting ability of rice (Oryza sativa L.). It is of great significance for the quantitative and objective evaluation of rice yield loss to obtain the rice lodging distribution and grade information rapidly and nondestructively. The purpose of this paper was to establish a remote sensing monitoring method for the grade of rice lodging based on change vector analysis via Sentinel-2A images before and after lodging. Taking the lodging ratio (LR) of rice in the parcel as the characterization index, the changes in spectral reflectance and vegetation index of the rice canopy under different lodging grade stresses were analyzed. A 7-dimensional space vector was composed of a canopy spectral reflectance and vegetation index. A new angle vector was formed by the angle between each vector object and the axis of the canopy spectrum (vegetation index). Based on the difference in feature vectors before and after lodging, a lodging grade monitoring model based on rice parcels was constructed, and the accuracy of the model was verified using field observations. The results showed that the spectral reflectance of the rice canopy after lodging increased with the increase in lodging stress. The difference in vegetation index before and after lodging increased with the increase in lodging severity. Based on the change in the magnitude of the canopy spectral reflectance and its angle vector, and the vegetation index and its angle vector, the R² of the rice lodging model was 0.66, 0.66, 0.60, and 0.59, respectively, and the RMSE values were 21.03%, 20.37%, 22.13%, and 22.84%, respectively. Therefore, the accuracy of the model constructed using the vegetation index was the highest. The canopy spectral reflectance and difference in the vegetation index of rice changed regularly with lodging grade stress. The model based on the change magnitude of the vegetation index could effectively realize remote sensing monitoring of the grade of rice lodging in Wuchang.

Lodging depresses the grain yield and quality of maize crop. Previous machine learning methods are used to classify crop lodging extents through visual interpretation and sensitive features extraction manually, which are cost-intensive, subjective and inefficient. The analysis on the accuracy of subdivision categories is insufficient for multi-grade crop lodging. In this study, a classification method of maize lodging extents was proposed based on deep learning algorithms and unmanned aerial vehicle (UAV) RGB and multispectral images. The characteristic variation of three lodging extents in RGB and multispectral images were analyzed. The VGG-16, Inception-V3 and ResNet-50 algorithms were trained and compared depending on classification accuracy and Kappa coefficient. The results showed that the more severe the lodging, the higher the intensity value and spectral reflectance of RGB and multispectral image. The reflectance variation in red edge band were more evident than that in visible band with different lodging extents. The classification performance using multispectral images was better than that of RGB images in various lodging extents. The test accuracies of three deep learning algorithms in non-lodging based on RGB images were high, i.e., over 90%, but the classification performance between moderate lodging and severe lodging needed to be improved. The test accuracy of ResNet-50 was 96.32% with Kappa coefficients of 0.9551 by using multispectral images, which was superior to VGG-16 and Inception-V3, and the accuracies of ResNet-50 on each lodging subdivision category all reached 96%. The ResNet-50 algorithm of deep learning combined with multispectral images can realize accurate lodging classification to promote post-stress field management and production assessment.

<!-- *** Custom HTML *** --> Lodging stress affects the yield, quality, and mechanical harvesting capacity of maize, thereby resulting in a reduction in maize production. The timely monitoring of maize lodging assists in determining the extent and degree of the effects. This study proposed a method for this purpose using Gaofen-1 (GF-1) optical satellite images, which can be significant for lodging field management, production assessments, and implementation of the related remedial measures. The evaluation index of the lodging grade was constructed by measuring the proportion and angle of maize lodging after the lodging incident. Variations in spectral reflectance and vegetation index were calculated before and after lodging. The competitive adaptive reweighted sampling algorithm was used to screen the optimal combination of variables sensitive to the evaluation index of the maize lodging grade. A remote sensing monitoring model was established by using random forest (RF) and partial least squares (PLS). Results show that the maize lodging monitoring model established by RF is better than that of PLS. The accuracy of monitoring the lodging damage using the GF-1 images reaches 79%. Meanwhile, the classification map of the lodging grade we constructed using an RF model is consistent with the actual lodging area. Hence, this method can effectively characterize the intensity of lodging stress and can also be used to assess the range of damages caused by regional-scale lodging.