Threatening Patch Attacks on Object Detection in Optical Remote Sensing Images

Advanced patch attacks (PAs) on object detection in natural images have pointed out the great safety vulnerability in methods based on deep neural networks (DNNs). However, little attention has been paid to this topic in optical remote sensing images (O-RSIs). To this end, we focus on this research, i.e., PAs on object detection in O-RSIs, and propose a more threatening patch attack (TPA) without the scarification of the visual quality. Specifically, to address the problem of inconsistency between the local and global landscapes in existing patch selection schemes, we propose leveraging the first-order difference (FOD) of the objective function before and after masking to select the subpatches to be attacked. Furthermore, considering the problem of gradient inundation when applying existing coordinate-based loss (CBL) to PAs directly, we design an IoU-based objective function specific for PAs, dubbed bounding box (Bbox) drifting loss (BDL), which pushes the detected Bboxes far from the initial ones until there are no intersections between them. Finally, on two widely used benchmarks, i.e., DIOR and DOTA, comprehensive evaluations of our TPA with four typical detectors (faster R-CNN, FCOS, RetinaNet, and YOLO-v4) witness its remarkable effectiveness. To the best of our knowledge, this is the first attempt to study the PAs on object detection in O-RSIs, and we hope this work can get our readers interested in studying this topic.