Search

In order to improve the accuracy of multispectral image inversion of soil and plant analytical development (SPAD) of the cotton canopy, image segmentation methods were utilized to remove the background interference, such as soil and shadow in UAV multispectral images. UAV multispectral images of cotton bud stage canopies at three different heights (30 m, 50 m, and 80 m) were acquired. Four methods, namely vegetation index thresholding (VIT), supervised classification by support vector machine (SVM), spectral mixture analysis (SMA), and multiple endmember spectral mixture analysis (MESMA), were used to segment cotton, soil, and shadows in the multispectral images of cotton. The segmented UAV multispectral images were used to extract the spectral information of the cotton canopy, and eight vegetation indices were calculated to construct the dataset. Partial least squares regression (PLSR), Random forest (FR), and support vector regression (SVR) algorithms were used to construct the inversion model of cotton SPAD. This study analyzed the effects of different image segmentation methods on the extraction accuracy of spectral information and the accuracy of SPAD modeling in the cotton canopy. The results showed that (1) The accuracy of spectral information extraction can be improved by removing background interference such as soil and shadows using four image segmentation methods. The correlation between the vegetation indices calculated from MESMA segmented images and the SPAD of the cotton canopy was improved the most; (2) At three different flight altitudes, the vegetation indices calculated by the MESMA segmentation method were used as the input variable, and the SVR model had the best accuracy in the inversion of cotton SPAD, with R2 of 0.810, 0.778, and 0.697, respectively; (3) At a flight altitude of 80 m, the R2 of the SVR models constructed using vegetation indices calculated from images segmented by VIT, SVM, SMA, and MESMA methods were improved by 2.2%, 5.8%, 13.7%, and 17.9%, respectively, compared to the original images. Therefore, the MESMA mixed pixel decomposition method can effectively remove soil and shadows in multispectral images, especially to provide a reference for improving the inversion accuracy of crop physiological parameters in low-resolution images with more mixed pixels.

Path planning is the core problem of orchard fertilization robots during their operation. The traditional full-coverage job path planning algorithm has problems, such as being not smooth enough and having a large curvature fluctuation, that lead to unsteady running and low working efficiency of robot trajectory tracking. To solve the above problems, an improved A* path planning algorithm based on a multi-constraint Bessel curve is proposed. First, by improving the traditional A* algorithm, the orchard operation path can be fully covered by adding guide points. Second, according to the differential vehicle kinematics model of the orchard fertilization robot, the robot kinematics constraint is combined with a Bessel curve to smooth the turning path of the A* algorithm, and the global path meeting the driving requirements of the orchard fertilization robot is generated by comprehensively considering multiple constraints such as the minimum turning radius and continuous curvature. Finally, the pure tracking algorithm is used to carry out tracking experiments to verify the robot’s driving accuracy. The simulation and experimental results show that the maximum curvature of the planned trajectory is 0.67, which meets the autonomous operation requirements of the orchard fertilization robot. When tracking the linear path in the fertilization area, the average transverse deviation is 0.0157 m, and the maximum transverse deviation is 0.0457 m. When tracking the U-turn path, the average absolute transverse deviation is 0.1081 m, and the maximum transverse deviation is 0.1768 m, which meets the autonomous operation requirements of orchard fertilization robots.

Obtaining consistent multi-view images of litchis is crucial for various litchi-related studies, such as data augmentation and 3D reconstruction. This paper proposes a two-stage model that integrates the Mask2Former semantic segmentation network with the Wonder3D multi-view generation network. This integration aims to accurately segment and extract litchis from complex backgrounds and generate consistent multi-view images of previously unseen litchis. In the first stage, the Mask2Former model is utilized to predict litchi masks, enabling the extraction of litchis from complex backgrounds. To further enhance the accuracy of litchi branch extraction, we propose a novel method that combines the predicted masks with morphological operations and the HSV color space. This approach ensures accurate extraction of litchi branches even when the semantic segmentation model’s prediction accuracy is not high. In the second stage, the segmented and extracted litchi images are passed as input into the Wonder3D network to generate multi-view of the litchis. After comparing different semantic segmentation and multi-view synthesis networks, the Mask2Former and Wonder3D networks demonstrated the best performance. The Mask2Former network achieved a mean Intersection over Union (mIoU) of 79.79% and a mean pixel accuracy (mPA) of 85.82%. The Wonder3D network achieved a peak signal-to-noise ratio (PSNR) of 18.89 dB, a structural similarity index (SSIM) of 0.8199, and a learned perceptual image patch similarity (LPIPS) of 0.114. Combining the Mask2Former model with the Wonder3D network resulted in an increase in PSNR and SSIM scores by 0.21 dB and 0.0121, respectively, and a decrease in LPIPS by 0.064 compared to using the Wonder3D model alone. Therefore, the proposed two-stage model effectively achieves automatic extraction and multi-view generation of litchis with high accuracy.

Utilizing agricultural UAVs for pesticide and insecticide spraying is an effective measure for plant protection. However, achieving effective coverage on the back side of target is often challenging. To address this issue, this study combined a contact-charging spraying system with a UAV to develop an electrostatic plant protection UAV system. Upon activating the electrostatic component, strong electrostatic effects were observed at the nozzle, altering the distribution of the liquid flow; the distribution within the liquid flow became more homogeneous, while the edge regions experienced electrostatic repulsion, leading to changes in droplet size and an increase in droplet density. In the central area, droplet size reduced from 159 μm to 135 μm, while in the edge area, it changed from no value to 120 μm. During field tests using the UAV, the results showed an increase of 1.0 m in effective spray width (at a flight height of 4.0 m), indicating that the charges and propellor wind field contributed to the diffusion of droplets towards the edges. Additionally, the droplet density increased by an average of 19.7 droplets/cm2, and the overall deposition increased by 0.12 μL/cm2, resulting in an approximate three-fold increase compared to conventional spray, which aids in insect control and reduces pesticide usage.

In order to achieve accurate detection of mature Zanthoxylum in their natural environment, a Zanthoxylum detection network based on the YOLOv5 object detection model was proposed. It addresses the issues of irregular shape and occlusion caused by the growth of Zanthoxylum on trees and the overlapping of Zanthoxylum branches and leaves with the fruits, which affect the accuracy of Zanthoxylum detection. To improve the model’s generalization ability, data augmentation was performed using different methods. To enhance the directionality of feature extraction and enable the convolution kernel to be adjusted according to the actual shape of each Zanthoxylum cluster, the coordinate attention module and the deformable convolution module were integrated into the YOLOv5 network. Through ablation experiments, the impacts of the attention mechanism and deformable convolution on the performance of YOLOv5 were compared. Comparisons were made using the Faster R-CNN, SSD, and CenterNet algorithms. A Zanthoxylum harvesting robot vision detection platform was built, and the visual detection system was tested. The experimental results showed that using the improved YOLOv5 model, as compared to the original YOLOv5 network, the average detection accuracy for Zanthoxylum in its natural environment was increased by 4.6% and 6.9% in terms of mAP@0.5 and mAP@0.5:0.95, respectively, showing a significant advantage over other network models. At the same time, on the test set of Zanthoxylum with occlusions, the improved model showed increased mAP@0.5 and mAP@0.5:0.95 by 5.4% and 4.7%, respectively, compared to the original model. The improved model was tested on a mobile picking platform, and the results showed that the model was able to accurately identify mature Zanthoxylum in its natural environment at a detection speed of about 89.3 frames per second. This research provides technical support for the visual detection system of intelligent Zanthoxylum-harvesting robots.

The use of drones in agriculture is expanding at a brisk pace in crop production due to the superiority in precision, efficiency, and safety of their applicators. However, their potential drift risk also raises concern for users and regulatory authorities. The method of wind tunnel research can effectively evaluate the weighted influence of each drift factor, especially the drift characteristics of the nozzle and spray solution. Based on the wind tunnel test results, centrifugal nozzles have a higher drift risk than hydraulic nozzles, even with a similar DV50. The cumulative drift rate of the centrifugal nozzle at 2 m downwind was 90.1% compared to the LU12001 nozzle’s 40.6% under the wind speed of 3.5 m/s. Compared with the same coding as the flat fan hydraulic nozzle, the IDK nozzle can effectively reduce the drift rate. For the tested nozzles, DV50 and wind speed had a linear relationship with drift rate, and the sampling location had an exponential or logarithmic relationship with drift rate. Spray adjuvants, especially modified vegetable oils, had a significant effect on reducing the amount of drift. The results of this experiment provide a reference for the selection of nozzles and the addition of spray adjuvants. Further clarifying the spray drift characteristics of drones until a drift prediction model is available is still the focus of research.

The droplet size distribution following pesticide application practices can significantly impact droplet drift and non-target organisms (animals and plants). However, the relationship among liquid sheet breakup, drop formation, and droplet drift is an area that has been studied over the past 65 years but is still not fully understood. The objectives of this study were severalfold: to examine the liquid sheet breakup following the use of different adjuvants (sodium dodecyl sulfate, aerosol OT, and silicone at 1%) and their effects on the drift via three commonly used commercial spray nozzles (XR, AIXR, and TXVK). The spray sheet and initial droplet size spectrum for each spray was detected by a particle image velocimetry (PIV) system, and the drift for each treatment was measured in a wind tunnel. The nozzle type and the spray solution were found to significantly affect the liquid sheet breakup characteristics. AIXR produced large droplets, and TXVK with short liquid sheet length produced more small droplets. All adjuvants used in these experiments increased the formation of large droplets upon atomization while simultaneously reducing driftable fines (e.g., drops %V0.5 was positive for XR and TXVK nozzles, and there was a significant positive correlation between %

Chemical application using unmanned aerial vehicles (UAVs) has received significant attention from researchers and the market in recent years. The concept of using drones for collaborative spraying was proposed by manufacturers for improving intelligence and work efficiency. However, chemical spraying is a professional technology in which spraying quality is the main concern. Using drones to achieve multi-unmanned aerial vehicle formation spraying and evaluating the spraying effect has not yet been reported. In this study, an indoor test platform and two UAVs for field experiments were built. Indoor and outdoor trials of close formation spraying were carried out in Guangzhou and Changji, China from the end of 2018 to 2019, respectively. The droplet density and distribution uniformity of droplets were evaluated from multiple spray overlap areas. It can be seen that simultaneous spraying was better than sequential spraying with the indoor spraying results in the outer fuselage overlap area (S1), and spraying in a short-interval mode can improve the droplet deposition distribution in the overlapping spraying area. Additionally, the droplet distribution result of sequential spraying was better than that of simultaneous spraying in the route center overlap area (S2). Also, the droplet distribution result of the long-interval mode was better than that of the short-interval mode. The uniformity of the droplets’ distribution in two spray width areas (S3) did not change significantly among the treatments.

Identifying and extracting male and female parent of hybrid rice and then accurately judging the spikelet flowering of male parents is the basis of hybrid rice pollination. Currently, male parent flowering information extraction for hybrid rice is basically obtained by manual recognition. In this study, remote sensing images of parental rice fields were obtained with a multispectral camera carried by a UAV (Umanned Aerial Vehicle). Six kinds of visible light vegetation indices and four kinds of multispectral vegetation indices, together with two classification methods, pixel-based supervised classification and sample-based object-oriented classification, were applied to identify the male and female parents of hybrid rice, after which the accuracies of the methods were compared. The results showed that the visible vegetation index had a better effect in pixel-based supervised classification. The kappa coefficient of ExGR (Excess Green minus Excess Red index) classification was 0.9256 and the total accuracy was 0.9552. The extraction accuracy was higher than that of the other vegetation indices and object-oriented classification. In pixel-based supervised classification, the maximum likelihood method achieved the highest identification accuracy and shortest calculation time. Taking the remote sensing images obtained with a UAV as a data source, maximum likelihood supervised classification based on ExGR index can more effectively and quickly identify the field information of male and female parents of hybrid rice so as to provide a reference for determining optimal pollination timing for hybrid rice in large-scale seed production farms.

Effective detection of rice spikelet flowering is crucial to the determination of optimal pollination timing for hybrid rice seed production. Currently, the detection of rice spikelet flowering status relies on manual observation of farmers, which has low efficiency and large errors. This study attempts to acquire rice spikelet flowering information using a hyperspectral technique and machine learning in order to meet the needs of hybrid rice seed pollination rapidly and automatically. Hyperspectral data of rice male parents with flowering and non-flowering in two experimental sites were collected with an ASD FieldSpec® HandHeld™2 spectrometer. Three traditional classifiers, Random Forest (RF), Support Vector Machine (SVM) and Back Propagation (BP) neural network, and Convolutional Neural Network (CNN), were used to build classification models for rice spikelets flowering detection. Three data processing methods, PCA feature extraction, GA feature selection, and the PCA and GA combination algorithm, were used for data dimensionality reduction. By comparing the precision and recall rate of different algorithms and data processing methods, the algorithms applicable to identify rice spikelet flowering were investigated. Results show that by evaluating different feature reduction methods and classifiers, the optimal model for rice spikelets flowering detection is the BP model with PCA feature extraction. The accuracy of the model reaches up to 96–100%. Hyperspectral technology and machine learning algorithm are capable of effective detection of rice spikelet flowering. This study provides technical reference for accurate judgment of rice flowering and helps to determine the optimal operation time for supplementary pollination of hybrid rice.

With the acceleration of the Belt and Road Initiative, Poland–China agricultural trade has increasingly strengthened, but there is little exchange or cooperation in agricultural machinery. China’s agricultural UAV industry has flourished over the past 7 years. In China, by using typical food crops and economic crops to control diseases and pests, agricultural drones can reduce the use of fertilizer, pesticides, and water, improve operational efficiency, open up new markets through the ‘sale + services’ mode, and reduce production costs and labor shortages. The spraying of agricultural UAVs and related pest-disease-defense services applied in China are also suitable for Poland’s decentralized, small-scale production modes. By learning from China’s development progress of precision-agriculture aviation, Poland can develop 5th-generation (5G) unmanned intelligent organic farms from traditional organic agriculture, use agricultural UAVs in the spraying of Plant Protection Products (PPPs), and carry out special protection or loss management on typical fruits. Furthermore, by building its own spraying system, aviation industry, and service team, Poland can realize resource optimization, technological empowerment, application expansion, and industrial innovation. Therefore, this paper focuses on the development experience of Chinese agricultural UAVs and discusses its enlightenment to the precision-agriculture aviation application of Poland.

The efficient use of water and fertilizer is vital for optimizing plant growth and yield in rice production. To achieve sustainable rice production and resource management, the ways in which applied water and nitrogen affect the root and shoot morpho-physiology, as well as yield, must be understood. In this study, a pot experiment was conducted to investigate the effects of slow-release nitrogen fertilizer (sulfur-coated urea) application at three levels (light nitrogen (NL), medium nitrogen (NM), and heavy nitrogen (NH)) on the growth, yield, and nitrogen use efficiency (NUE) of rice grown under three water regimes (wetting and soil saturation (WSS), wetting and moderate drying (WMD) and wetting and severe drying (WSD)). The results revealed that differences in water regimes and fertilizer rates led to significant differences in the roots, shoots, yield, and NUE of rice. Increasing the N dosage by 5% enhanced the root and biomass production by 16% in comparison with that of the other groups. The NH×WSS treatment produced the greatest root length, weight, density, active absorption, and oxidation. However, the integration of WSS × NL generated the maximum value of nitrogen apparent recovery efficiency (63.1% to 67.6%) and the greatest value of nitrogen partial factor productivity (39.9 g g−1 to 41.13 g g−1). Transmission electron microscopy (TEM) images showed that plants grown under high and medium nitrogen fertilizer rates with WSS had improved leaf mesophyll structure with normal starch grains, clear cell walls, and well-developed chloroplasts with tidy and well-arranged thylakoids. These results show that TEM images are useful for characterizing the nitrogen and water status of leaves in the sub-micrometer range and providing specific information regarding the leaf microstructure. The findings of this study suggest that the application of NH×WSS can produce improvements in growth traits and increase rice yield; however, the NL×WSS treatment led to greater NUE, and the authors recommend its usage in rice agriculture.

Soil moisture is an important factor determining yield. With the increasing demand for agricultural irrigation water resources, evaluating soil moisture in advance to create a reasonable irrigation schedule would help improve water resource utilization. This paper established a continuous system for collecting meteorological information and soil moisture data from a litchi orchard. With the acquired data, a time series model called Deep Long Short-Term Memory (Deep-LSTM) is proposed in this paper. The Deep-LSTM model has five layers with the fused time series data to predict the soil moisture of a litchi orchard in four different growth seasons. To optimize the data quality of the soil moisture sensor, the Symlet wavelet denoising algorithm was applied in the data preprocessing section. The threshold of the wavelets was determined based on the unbiased risk estimation method to obtain better sensor data that would help with the model learning. The results showed that the root mean square error (RMSE) values of the Deep-LSTM model were 0.36, 0.52, 0.32, and 0.48%, and the mean absolute percentage error (MAPE) values were 2.12, 2.35, 1.35, and 3.13%, respectively, in flowering, fruiting, autumn shoots, and flower bud differentiation stages. The determination coefficients (R2) were 0.94, 0.95, 0.93, and 0.94, respectively, in the four different stages. The results indicate that the proposed model was effective at predicting time series soil moisture data from a litchi orchard. This research was meaningful with regards to acquiring the soil moisture characteristics in advance and thereby providing a valuable reference for the litchi orchard’s irrigation schedule.