Remote sensing is important to precision agriculture and the spatial resolution provided by Unmanned Aerial Vehicles (UAVs) is revolutionizing precision agriculture workflows for measurement crop condition and yields over the growing season, for identifying and monitoring weeds and other applications. Monitoring of individual trees for growth, fruit production and pest and disease occurrence remains a high research priority and the delineation of each tree using automated means as an alternative to manual delineation would be useful for long-term farm management. In this paper, we detected citrus and other crop trees from UAV images using a simple convolutional neural network (CNN) algorithm, followed by a classification refinement using superpixels derived from a Simple Linear Iterative Clustering (SLIC) algorithm. The workflow performed well in a relatively complex agricultural environment (multiple targets, multiple size trees and ages, etc.) achieving high accuracy (overall accuracy = 96.24%, Precision (positive predictive value) = 94.59%, Recall (sensitivity) = 97.94%). To our knowledge, this is the first time a CNN has been used with UAV multi-spectral imagery to focus on citrus trees. More of these individual cases are needed to develop standard automated workflows to help agricultural managers better incorporate large volumes of high resolution UAV imagery into agricultural management operations.

Extracting the shapes of individual tree crowns from high-resolution imagery can play a crucial role in many applications, including precision agriculture. We evaluated three CNN models—MASK R-CNN, YOLOv3, and SAM—and compared their tree crown results with OBIA-based reference datasets from UAV imagery for seven dates across one growing season. We found that YOLOv3 performed poorly across all dates; both MASK R-CNN and SAM performed well in May, June, September, and November (precision, recall, and F1 scores over 0.79). All models struggled in the early season imagery (e.g., March). MASK R-CNN outperformed other models in August (when there was smoke haze) and December (showing end-of-season variation in leaf color). SAM was the fastest model, and, as it required no training, it could cover more area in less time; MASK R-CNN was very accurate and customizable. In this paper, we aimed to contribute insight into which CNN model offers the best balance of accuracy and ease of implementation for orchard management tasks. We also evaluated its applicability within one software ecosystem, ESRI ArcGIS Pro, and showed how such an approach offers users a streamlined efficient way to detect objects in high-resolution UAV imagery.

Extracting the irregular and complex shapes of individual tree crowns from high-resolution imagery can play a crucial role in many applications, including precision agriculture. We evaluated three CNN models - MASK R-CNN, YOLOv3, and SAM - and compared their tree crown results with OBIA-based reference datasets from UAV imagery for seven dates across one growing season. We found that YOLOv3 performed poorly across all dates; both MASK R-CNN and SAM performed well in May, June, September, and November (Precision, Recall and F1 scores over 0.79). All models struggled in the early season imagery (e.g., March). MASK R-CNN outperformed other models in August (when there was smoke haze) and December (showing end of season red leaf senescence). SAM was the fastest model, and as it required no training, it could cover more area in less time; MASK R-CNN was very accurate and customizable. In this paper, we aimed to contribute insight into which CNN model offers the best balance of accuracy and ease of implementation for orchard management tasks. We also evaluated their applicability within one software ecosystem, ESRI ArcGIS Pro, and showed how such an approach offers users a streamlined, efficient way to detect objects in high resolution UAV imagery.