Attention-Free Global Multiscale Fusion Network for Remote Sensing Object Detection

Remote sensing object detection (RSOD) encounters challenges in complex backgrounds and small object detection, which are interconnected and unable to address separately. To this end, we propose an attention-free global multiscale fusion network (AGMF-Net). Initially, we present a spatial bias module (SBM) to obtain long-range dependencies as a part of our proposal global information extraction module (GIEM). GIEM efficiently captures the global information, overcoming challenges posed by complex backgrounds. Moreover, we propose multitask enhanced structure (MES) and multitask feature pretreatment (MFP) to enhance the feature representation of multiscale targets, while eliminating the interference from complex backgrounds. In addition, an efficient context decoupled detector (ECDD) is presented to provide distinct features for regression and classification tasks, aiming to improve the efficiency of RSOD. Extensive experiments demonstrate that our proposed method achieves superior performance compared with the state-of-the-art detectors. Specifically, AGMF-Net obtains the mean average precision (mAP) of 73.2%, 92.03%, 95.21%, and 94.30% on detection in optical remote sensing images (DIOR), high resolution remote sensing detection (HRRSD), Northwestern Polytechnical University Very High Resolution-10 (NWPU VHR-10), and RSOD datasets, respectively.

I. Introduction

Object detection is a popular research task in the analysis of remote sensing images (RSIs), which is widely used in many fields, such as intelligent transportation, urban planning, military reconnaissance, and environmental monitoring [1]. Given the vast amount of remote sensing satellite data, detecting remote sensing objects in the presence of intricate backgrounds, densely clustered small objects, and extensive scale variations has been consistently a focal point of research [3]. Differing from natural scene images, RSIs are captured from a top–down perspective, leading to remote sensing objects exhibiting diverse scales and uncertain orientations, where small targets are often submerged within complex background noise.