Affiliation:
1. School of Electronic Information Engineering, Beihang University, Beijing 100191, China
2. Institute of Unmanned System, Beihang University, Beijing 100191, China
Abstract
Significant progress has been achieved in the field of oriented object detection (OOD) in recent years. Compared to natural images, objects in remote sensing images exhibit characteristics of dense arrangement and arbitrary orientation while also containing a large amount of background information. Feature extraction in OOD becomes more challenging due to the diversity of object orientations. In this paper, we propose a semantic-driven rotational feature enhancement method, termed SREDet, to fully leverage the joint semantic and spatial information of oriented objects in the remote sensing images. We first construct a multi-rotation feature pyramid network (MRFPN), which leverages a fusion of multi-angle and multiscale feature maps to enhance the capability to extract features from different orientations. Then, considering feature confusion and contamination caused by the dense arrangement of objects and background interference, we present a semantic-driven feature enhancement module (SFEM), which decouples features in the spatial domain to separately enhance the features of objects and weaken those of backgrounds. Furthermore, we introduce an error source evaluation metric for rotated object detection to further analyze detection errors and indicate the effectiveness of our method. Extensive experiments demonstrate that our SREDet method achieves superior performance on two commonly used remote sensing object detection datasets (i.e., DOTA and HRSC2016).
Funder
National Natural Science Foundation of China
Reference65 articles.
1. A comprehensive survey of oriented object detection in remote sensing images;Wen;Expert Syst. Appl.,2023
2. Object detection in optical remote sensing images: A survey and a new benchmark;Li;ISPRS J. Photogramm. Remote Sens.,2020
3. Methods for small, weak object detection in optical high-resolution remote sensing images: A survey of advances and challenges;Han;IEEE Geosci. Remote Sens. Mag.,2021
4. Yang, L., Jiang, H., Cai, R., Wang, Y., Song, S., Huang, G., and Tian, Q. (2021, January 20–25). Condensenet v2: Sparse feature reactivation for deep networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA.
5. Remote sensing image super-resolution and object detection: Benchmark and state of the art;Wang;Expert Syst. Appl.,2022