ABSTRACT Coffee farmers do not have efficient tools to have sufficient and reliable information on the maturation stage of coffee fruits before harvest. In this study, we propose a computer vision system to detect and classify the Coffea arabica (L.) on tree branches in three classes: unripe (green), ripe (cherry), and overripe (dry). Based on deep learning algorithms, the computer vision model YOLO (You Only Look Once), was trained on 387 images taken from coffee branches using a smartphone. The YOLOv3 and YOLOv4, and their smaller versions (tiny), were assessed for fruit detection. The YOLOv4 and YOLOv4-tiny showed better performance when compared to YOLOv3, especially when smaller network sizes are considered. The mean average precision (mAP) for a network size of 800 × 800 pixels was equal to 81 %, 79 %, 78 %, and 77 % for YOLOv4, YOLOv4-tiny, YOLOv3, and YOLOv3-tiny, respectively. Despite the similar performance, the YOLOv4 feature extractor was more robust when images had greater object densities and for the detection of unripe fruits, which are generally more difficult to detect due to the color similarity to leaves in the background, partial occlusion by leaves and fruits, and lighting effects. This study shows the potential of computer vision systems based on deep learning to guide the decision-making of coffee farmers in more objective ways.

ABSTRACT A common agricultural problem in many regions of Brazil is maize lodging, as a consequence of strong winds and rain which impacts on crop growth and yield. However, collecting data using ground-based, manual field measurement methods is inefficient. An emerging tool is the Remotely Piloted Aircraft System (RPAS), capable of delivering spatial data with high resolution and flexible periodicity. In this study, the potential to detect the maize lodging using crop surface models derived from RPAS was assessed. Our RPA-based approach uses a quantitative threshold to determine lodging percentage. The threshold values of plant height, used to detect the occurrence of lodging, were based on fixed and variable values. The validation of percentage lodging was performed using the RGB orthomosaic. The derived lodging estimates showed a very high correlation to the reference data. High correlations were observed for the fixed threshold at 60% (R2 = 0.93, RSME = 8.72%) and the variable thresholds, Jenks natural breaks and iso-clusters (R2 = 0.92, RSME = 8.89% and R2= 0.92, RSME = 9%, respectively). This study demonstrated the potential of the use of this technique, reducing the subjectivity of ground-based evaluation and the laborious traditional technique of lodging inference.