Mid- to late-season weeds that escape from the routine early-season weed management threaten agricultural production by creating a large number of seeds for several future growing seasons. Rapid and accurate detection of weed patches in field is the first step of site-specific weed management. In this study, object detection-based convolutional neural network models were trained and evaluated over low-altitude unmanned aerial vehicle (UAV) imagery for mid- to late-season weed detection in soybean fields. The performance of two object detection models, Faster RCNN and the Single Shot Detector (SSD), were evaluated and compared in terms of weed detection performance using mean Intersection over Union (IoU) and inference speed. It was found that the Faster RCNN model with 200 box proposals had similar good weed detection performance to the SSD model in terms of precision, recall, f1 score, and IoU, as well as a similar inference time. The precision, recall, f1 score and IoU were 0.65, 0.68, 0.66 and 0.85 for Faster RCNN with 200 proposals, and 0.66, 0.68, 0.67 and 0.84 for SSD, respectively. However, the optimal confidence threshold of the SSD model was found to be much lower than that of the Faster RCNN model, which indicated that SSD might have lower generalization performance than Faster RCNN for mid- to late-season weed detection in soybean fields using UAV imagery. The performance of the object detection model was also compared with patch-based CNN model. The Faster RCNN model yielded a better weed detection performance than the patch-based CNN with and without overlap. The inference time of Faster RCNN was similar to patch-based CNN without overlap, but significantly less than patch-based CNN with overlap. Hence, Faster RCNN was found to be the best model in terms of weed detection performance and inference time among the different models compared in this study. This work is important in understanding the potential and identifying the algorithms for an on-farm, near real-time weed detection and management.

As one of the key crop traits, plant height is traditionally evaluated manually, which can be slow, laborious and prone to error. Rapid development of remote and proximal sensing technologies in recent years allows plant height to be estimated in more objective and efficient fashions, while research regarding direct comparisons between different height measurement methods seems to be lagging. In this study, a ground-based multi-sensor phenotyping system equipped with ultrasonic sensors and light detection and ranging (LiDAR) was developed. Canopy heights of 100 wheat plots were estimated five times during a season by the ground phenotyping system and an unmanned aircraft system (UAS), and the results were compared to manual measurements. Overall, LiDAR provided the best results, with a root-mean-square error (RMSE) of 0.05 m and an R² of 0.97. UAS obtained reasonable results with an RMSE of 0.09 m and an R² of 0.91. Ultrasonic sensors did not perform well due to our static measurement style. In conclusion, we suggest LiDAR and UAS are reliable alternative methods for wheat height evaluation.

Latest advances in unmanned aerial vehicle (UAV) technology and convolutional neural networks (CNNs) allow us to detect crop lodging in a more precise and accurate way. However, the performance and generalization of a model capable of detecting lodging when the plants may show different spectral and morphological signatures have not been investigated much. This study investigated and compared the performance of models trained using aerial imagery collected at two growth stages of winter wheat with different canopy phenotypes. Specifically, three CNN-based models were trained with aerial imagery collected at early grain filling stage only, at physiological maturity only, and at both stages. Results show that the multi-stage model trained by images from both growth stages outperformed the models trained by images from individual growth stages on all testing data. The mean accuracy of the multi-stage model was 89.23% for both growth stages, while the mean of the other two models were 52.32% and 84.9%, respectively. This study demonstrates the importance of diversity of training data in big data analytics, and the feasibility of developing a universal decision support system for wheat lodging detection and mapping multi-growth stages with high-resolution remote sensing imagery.

Nitrogen Sufficiency Index (NSI) is an important nitrogen (N) stress indicator for precision N management. It is usually calculated using variables such as leaf chlorophyll meter readings (SPAD) and vegetation indices (VIs). However, no consensus has been reached on the most preferred variable. Additionally, conventional NSI (NSI_uni) calculation assumes N being the sole yield-limiting factor, neglecting other factors such as soil water variability. To tackle these issues, this study compared various variables for NSI calculation and evaluated two new N stress indicators in minimizing the impact of confounding water treatment. The following ground- and aerial-derived variables were compared for NSI_uni calculation: SPAD, sampled leaf and canopy N content (LNC, CNC), LNC and CNC estimated using hyperspectral images acquired by an Unmanned Aerial Vehicle, and three VIs (Normalized Difference Vegetation Index (NDVI), Normalized Red Edge Index (NDRE), and Chlorophyll Index) from the hyperspectral images. Results demonstrated that ground-measured variables outperformed aerial-based variables in deriving N-responsive NSI. Especially, LNC derived NSI_uni responded to N treatment significantly in ten out of thirteen site-date datasets. For the second objective, a modified NSI (NSI_w) and the NDRE/NDVI ratio were compared to NSI_uni. NSI_w reduced water treatment effects in over 80% of the datasets where NSI_uni showed evident impacts. NDRE/NDVI performed similarly to NSI_w, with the notable advantage of not requiring prior knowledge of soil water spatial distribution. This research pioneers the optimization of N stress indicators by identifying the best variables for NSI and mitigating the effects of soil water variability. These advancements significantly contribute to precision N management in complex field conditions.