As a new low volume application technology, unmanned aerial vehicle (UAV) application is developing quickly in China. The aim of this study was to compare the droplet deposition, control efficacy and working efficiency of a six-rotor UAV with a self-propelled boom sprayer and two conventional knapsack sprayers on the wheat crop. The total deposition of UAV and other sprayers were not statistically significant, but significantly lower for run-off. The deposition uniformity and droplets penetrability of the UAV were poor. The deposition variation coefficient of the UAV was 87.2%, which was higher than the boom sprayer of 31.2%. The deposition on the third top leaf was only 50.0% compared to the boom sprayer. The area of coverage of the UAV was 2.2% under the spray volume of 10 L/ha. The control efficacy on wheat aphids of UAV was 70.9%, which was comparable to other sprayers. The working efficiency of UAV was 4.11 ha/h, which was roughly 1.7–20.0 times higher than the three other sprayers. Comparable control efficacy results suggest that UAV application could be a viable strategy to control pests with higher efficiency. Further improvement on deposition uniformity and penetrability are needed.

Previous studies have confirmed that choosing nozzles that produce coarser droplets could reduce the risk of pesticide spray drift, but this conclusion is based on a large volume of application, and it is easy to ignore how this impacts the control effect. The difference from the conventional spray is that the carrier volume of Unmanned Aerial Vehicle (UAV) is very limited. Little was known about how to choose suitable nozzles with UAV’s limited volume to ensure appropriate pest control. Droplet deposition with the addition of adjuvant and the LU110-010, LU110-015, and LU110-020 nozzles and control of planthoppers within nozzles treatments were studied by a quadrotor UAV in rice (Tillering and Flowering stages). Allura Red (10 g/L) was used as a tracer and Kromekote cards were used to collect droplet deposits. The results indicate that the density of the droplets covered by the LU110-01 nozzle is well above other treatments, while the differences in droplet deposition and coverage are not significant. The deposition and coverage were improved with the addition of adjuvant, especially in LU110-01 nozzles’ treatment. The control effects of rice planthoppers treated by LU110-01 nozzle were 89.4% and 90.8% respectively, which were much higher than 67.6% and 58.5% of LU110-020 nozzle at 7 days in the Tillering and Flowering stage. The results suggest that selecting a nozzle with a small atomizing particle size for UAV could improve the control effect of planthoppers.

Plant diseases cause significant economic losses and food security in agriculture each year, with the critical path to reducing losses being accurate identification and timely diagnosis of plant diseases. Currently, deep neural networks have been extensively applied in plant disease identification, but such approaches still suffer from low identification accuracy and numerous parameters. Hence, this paper proposes a model combining channel attention and channel pruning called CACPNET, suitable for disease identification of common species. The channel attention mechanism adopts a local cross-channel strategy without dimensionality reduction, which is inserted into a ResNet-18-based model that combines global average pooling with global max pooling to effectively improve the features’ extracting ability of plant leaf diseases. Based on the model’s optimum feature extraction condition, unimportant channels are removed to reduce the model’s parameters and complexity via the L1-norm channel weight and local compression ratio. The accuracy of CACPNET on the public dataset PlantVillage reaches 99.7% and achieves 97.7% on the local peanut leaf disease dataset. Compared with the base ResNet-18 model, the floating point operations (FLOPs) decreased by 30.35%, the parameters by 57.97%, the model size by 57.85%, and the GPU RAM requirements by 8.3%. Additionally, CACPNET outperforms current models considering inference time and throughput, reaching 22.8 ms/frame and 75.5 frames/s, respectively. The results outline that CACPNET is appealing for deployment on edge devices to improve the efficiency of precision agriculture in plant disease detection.

Peanut is an important food crop, and diseases of its leaves can directly reduce its yield and quality. In order to solve the problem of automatic identification of peanut-leaf diseases, this paper uses a traditional machine-learning method to ensemble the output of a deep learning model to identify diseases of peanut leaves. The identification of peanut-leaf diseases included healthy leaves, rust disease on a single leaf, leaf-spot disease on a single leaf, scorch disease on a single leaf, and both rust disease and scorch disease on a single leaf. Three types of data-augmentation methods were used: image flipping, rotation, and scaling. In this experiment, the deep-learning model had a higher accuracy than the traditional machine-learning methods. Moreover, the deep-learning model achieved better performance when using data augmentation and a stacking ensemble. After ensemble by logistic regression, the accuracy of residual network with 50 layers (ResNet50) was as high as 97.59%, and the F1 score of dense convolutional network with 121 layers (DenseNet121) was as high as 90.50. The deep-learning model used in this experiment had the greatest improvement in F1 score after the logistic regression ensemble. Deep-learning networks with deeper network layers like ResNet50 and DenseNet121 performed better in this experiment. This study can provide a reference for the identification of peanut-leaf diseases.

Leaf area index (LAI) is used to predict crop yield, and unmanned aerial vehicles (UAVs) provide new ways to monitor LAI. In this study, we used a fixed-wing UAV with multispectral cameras for remote sensing monitoring. We conducted field experiments with two peanut varieties at different planting densities to estimate LAI from multispectral images and establish a high-precision LAI prediction model. We used eight vegetation indices (VIs) and developed simple regression and artificial neural network (BPN) models for LAI and spectral VIs. The empirical model was calibrated to estimate peanut LAI, and the best model was selected from the coefficient of determination and root mean square error. The red (660 nm) and near-infrared (790 nm) bands effectively predicted peanut LAI, and LAI increased with planting density. The predictive accuracy of the multiple regression model was higher than that of the single linear regression models, and the correlations between Modified Red-Edge Simple Ratio Index (MSR), Ratio Vegetation Index (RVI), Normalized Difference Vegetation Index (NDVI), and LAI were higher than the other indices. The combined VI BPN model was more accurate than the single VI BPN model, and the BPN model accuracy was higher. Planting density affects peanut LAI, and reflectance-based vegetation indices can help predict LAI.

The purpose of this study is to determine a method for quickly and accurately estimating the chlorophyll content of peanut plants at different plant densities. This was explored using leaf spectral reflectance to monitor peanut chlorophyll content to detect sensitive spectral bands and the optimum spectral indicators to establish a quantitative model. Peanut plants under different plant density conditions were monitored during three consecutive growth periods; single-photon avalanche diode (SPAD) and hyperspectral data derived from the leaves under the different plant density conditions were recorded. By combining arbitrary bands, indices were constructed across the full spectral range (350–2500 nm) based on blade spectra: the normalized difference spectral index (NDSI), ratio spectral index (RSI), difference spectral index (DSI) and soil-adjusted spectral index (SASI). This enabled the best vegetation index reflecting peanut-leaf SPAD values to be screened out by quantifying correlations with chlorophyll content, and the peanut leaf SPAD estimation models established by regression analysis to be compared and analyzed. The results showed that the chlorophyll content of peanut leaves decreased when plant density was either too high or too low, and that it reached its maximum at the appropriate plant density. In addition, differences in the spectral reflectance of peanut leaves under different chlorophyll content levels were highly obvious. Without considering the influence of cell structure as chlorophyll content increased, leaf spectral reflectance in the visible (350–700 nm): near-infrared (700–1300 nm) ranges also increased. The spectral bands sensitive to chlorophyll content were mainly observed in the visible and near-infrared ranges. The study results showed that the best spectral indicators for determining peanut chlorophyll content were NDSI (R₅₂₀, R₅₂₈), RSI (R₇₄₈, R₅₆₁), DSI (R₇₅₈, R₆₀₂) and SASI (R₇₅₃, R₆₂₄). Testing of these regression models showed that coefficient of determination values based on the NDSI, RSI, DSI and SASI estimation models were all greater than 0.65, while root mean square error values were all lower than 2.04. Therefore, the regression model established according to the above spectral indicators was a valid predictor of the chlorophyll content of peanut leaves.