Image Recognition of Male Oilseed Rape (Brassica napus) Plants Based on Convolutional Neural Network for UAAS Navigation Applications on Supplementary Pollination and Aerial Spraying

To ensure the hybrid oilseed rape (OSR, Brassica napus) seed production, two important things are necessary, the stamen sterility on the female OSR plants and the effective pollen spread onto the pistil from the OSR male plants to the OSR female plants. The unmanned agricultural aerial system (UAAS) has developed rapidly in China. It has been used on supplementary pollination and aerial spraying during the hybrid OSR seed production. This study developed a new method to rapidly recognize the male OSR plants and extract the row center line for supporting the UAAS navigation. A male OSR plant recognition model was constructed based on the convolutional neural network (CNN). The sequence images of male OSR plants were extracted, the feature regions and points were obtained from the images through morphological and boundary process methods and horizontal segmentation, respectively. The male OSR plant image recognition accuracies of different CNN structures and segmentation sizes were discussed. The male OSR plant row center lines were fitted using the least-squares method (LSM) and Hough transform. The results showed that the segmentation algorithm could segment the male OSR plants from the complex background. The highest average recognition accuracy was 93.54%, and the minimum loss function value was 0.2059 with three convolutional layers, one fully connected layer, and a segmentation size of 40 pix × 40 pix. The LSM is better for center line fitting. The average recognition model accuracies of original input images were 98% and 94%, and the average root mean square errors (RMSE) of angle were 3.22° and 1.36° under cloudy day and sunny day lighting conditions, respectively. The results demonstrate the potential of using digital imaging technology to recognize the male OSR plant row for UAAS visual navigation on the applications of hybrid OSR supplementary pollination and aerial spraying, which would be a meaningful supplement in precision agriculture.

Influence of the Downwash Wind Field of Plant Protection UAV on Droplet Deposition Distribution Characteristics at Different Flight Heights

The aerial spraying of pesticides by plant protection unmanned aerial vehicles (UAV) is a process in which the spray droplet deposition on target sites occurs under the influence of the downwash wind field. The downwash wind field is the most important factor affecting droplet deposition distribution characteristics in an aerial spray. To understand the mechanism of the downwash wind field, spray tests were conducted at different flight heights by using a DJI UAV, and the downwash wind field in the three-dimensional direction (X-directional wind, Y-crosswind, and Z-vertical wind) was measured by using a wind speed measurement system for UAV. Combined with the droplet deposition of aerial spray, the distribution characteristics of the downwash wind field and the influence of the downwash wind field on droplet deposition were studied. The results showed that it had obvious differences in the distribution of the downwash wind field for UAV at different flight heights. As the flight height increases, the downwash wind field in X-direction and Z-direction showed a strong to weak trend, while the downwash wind field in Y-direction showed an opposite trend. In addition, it was found that the downwash wind field in Y-direction and Z-direction both have a significant influence on droplet deposition. With the increase of flight height, the change of the downwash wind field led to a gradual decrease in droplet deposition in the effective spray area, and droplets deposited more uniformly. For the DJI T16 plant protection UAV in this test, the optimal flight height was 2.0 m, and the downwash wind field had a better improvement effect on droplet deposition. Therefore, in order to make full use of the downwash wind field of UAV, the appropriate flight height should be selected to improve droplet deposition of liquid pesticide and achieve a better control effect for crop disease and pests when UAV is used for aerial spray operations in the field. This study revealed the influence mechanism of the downwash wind field on droplet deposition of aerial spray, and proposed appropriate operation parameters from the perspective of practical operation. It was expected to provide data support for improving the operation quality of aerial spraying and the formulation of field operation specifications.

The Role for Drones in Future Aerial Pesticide Applications

Aircraft have been used to apply pesticides for decades, but their use has been largely confined to spraying crops grown on very large fields, mostly in north, central and south America. Aerial spraying has also been vital in controlling locust swarms spread over vast areas. The development of unmanned aircraft, known as drones, has benefitted smallscale farmers as the drone can be controlled effectively over specific areas. Drones have also benefitted vector control by detecting the breeding sites of mosquitoes, which can be sprayed subsequently with larvicides, or controlled by applying space sprays. Smaller drones are an important way of replacing ground methods in difficult to reach areas such as wetlands, steep terrain for vineyards, and small areas of trees.

Efficacy of Insecticide Aerial Spraying Programs to Reduce Tree Mortality During a Spruce Budworm Outbreak (1967‒1992) in the Quebec Province.

Spruce budworm (Choristoneura fumiferana) is the most important pest of conifer forests in Eastern North America. The main approach to protect Quebec forests against this defoliator is aerial spraying of insecticides. Despite the crucial role of aerial spraying of insecticides in the global forest protection strategy, little is known about its real impact on tree mortality reduction or the stand characteristics that may affect its long-term efficacy. We evaluate the efficacy of the protection program implemented between 1967 and 1992 in reducing tree mortality during the previous outbreak in Quebec and how its efficacy is affected by stand characteristics such as drainage quality and stand age. We established 422 plots in Eastern Quebec based on the following parameters: insecticide application (0-13 years), stand composition, drainage quality, and stand age at the beginning of the previous outbreak. Our results show that unprotected plots exhibited in average between 18 and 32.6% higher mortality than protected plots. Insecticide efficacy was affected by drainage quality. For example, protected plots established on sites with bad drainage quality exhibited 49% less balsam fir mortality than unprotected plots. Furthermore, immature and mature protected stands exhibited a similar reduction in balsam fir mortality (32.2 and 32.1% less mortality respectively) compare to unprotected stands. Our results may be useful for decision-makers not only to decide where and when apply insecticides during a spruce budworm outbreak but also, they may help to determine the pertinence of continuous protection during longer than usual outbreaks depending on protection goal.