BBoxCut: A Targeted Data Augmentation Technique for Enhancing Wheat Head Detection Under Occlusions
This work addresses the challenge of automating wheat head detection for breeders, but it is incremental as it builds on existing object detectors with a new augmentation method.
The paper tackles the problem of detecting wheat heads under occlusions in field conditions by proposing BBoxCut, a targeted data augmentation technique that simulates occlusions, resulting in mAP gains of 2.76, 3.26, and 1.9 for three state-of-the-art object detectors.
Wheat plays a critical role in global food security, making it one of the most extensively studied crops. Accurate identification and measurement of key characteristics of wheat heads are essential for breeders to select varieties for cross-breeding, with the goal of developing nutrient-dense, resilient, and sustainable cultivars. Traditionally, these measurements are performed manually, which is both time-consuming and inefficient. Advances in digital technologies have paved the way for automating this process. However, field conditions pose significant challenges, such as occlusions of leaves, overlapping wheat heads, varying lighting conditions, and motion blur. In this paper, we propose a novel data augmentation technique, BBoxCut, which uses random localized masking to simulate occlusions caused by leaves and neighboring wheat heads. We evaluated our approach using three state-of-the-art object detectors and observed mean average precision (mAP) gains of 2.76, 3.26, and 1.9 for Faster R-CNN, FCOS, and DETR, respectively. Our augmentation technique led to significant improvements both qualitatively and quantitatively. In particular, the improvements were particularly evident in scenarios involving occluded wheat heads, demonstrating the robustness of our method in challenging field conditions.