Reliable Detection of Minute Targets in High-Resolution Aerial Imagery across Temporal Shifts
It addresses a domain-specific challenge in agriculture by improving crop detection reliability, but it is incremental as it applies an existing method to new data with minor adaptations.
This paper tackled the problem of detecting small rice seedlings in aerial imagery for precision agriculture, achieving consistent performance across different temporal test sets through transfer learning with a Faster R-CNN architecture.
Efficient crop detection via Unmanned Aerial Vehicles is critical for scaling precision agriculture, yet it remains challenging due to the small scale of targets and environmental variability. This paper addresses the detection of rice seedlings in paddy fields by leveraging a Faster R-CNN architecture initialized via transfer learning. To overcome the specific difficulties of detecting minute objects in high-resolution aerial imagery, we curate a significant UAV dataset for training and rigorously evaluate the model's generalization capabilities. Specifically, we validate performance across three distinct test sets acquired at different temporal intervals, thereby assessing robustness against varying imaging conditions. Our empirical results demonstrate that transfer learning not only facilitates the rapid convergence of object detection models in agricultural contexts but also yields consistent performance despite domain shifts in image acquisition.