FAIR-CSAR: A Benchmark Dataset for Fine-Grained Object Detection and Recognition Based on Single-Look Complex SAR Images

Object detection and recognition (OD&R) based on deep learning is a hot topic in the application of synthetic aperture radar (SAR). These methodologies based on deep learning are inherently data-driven, which means that their performance is subjected to the corresponding datasets. Although existing datasets have included some common targets collected from real-valued intensity SAR images, there still exist some limitations in terms of quantity, categories, diversities, and data domain. Hence, it is urgent to establish a large-quantity benchmark for fine-grained OD&R on complex-valued SAR images, which contains rich signal-domain features well coupled with classical physical modeling. In addition, considering the unique imaging characteristics and diverse imaging conditions, some important attribute information, such as incidence and attitude angles, is necessary to be attached. In this article, we propose a novel benchmark dataset with more than 340k instances for fine-grained OD&R based on single-look complex (SLC) SAR images, which is named FAIR-CSAR. We collected complex-valued SAR images with a resolution of 1–5 m from 175 entire images of Gaofen-3 covering 32 cities and multiple sea areas worldwide. All instances in the FAIR-CSAR are annotated by oriented bounding boxes (OBBs), covering five major categories and 22 subcategories. Compared with existing datasets dedicated to OD&R, the FAIR-CSAR dataset has four particular advantages: 1) it contains complex-valued SAR images from various acquisition modes and polarization modes, including full-scale signal-domain features for object recognition; 2) it is much larger than other existing OD&R datasets in terms of quantity of instances; 3) it provides more fine-grained category annotation and more detailed attribute information; and 4) it provides more challenging images with some common imaging phenomena, such as speckle noise and azimuth ambiguities. To establish a baseline adapted for SLC SAR images, a multidomain feature extraction and fusion network (MDNet) is proposed as a novel framework to mine detailed information underlying various domains. A series of state-of-the-art (SOTA) algorithms are applied on the FAIR-CSAR to build the fine-grained OD&R benchmark. Experimental results indicate that FAIR-CSAR is closer to practical application and more challenging than existing datasets for SAR images.

I. Introduction

Synthetic aperture radar (SAR) is a common remote sensing payload that actively emits electromagnetic waves and receives its echo to acquire Earth surface scattering information. Contributed to its waveband being less affected by atmospheric attenuation, SAR can achieve global monitoring and information collection all day and all weather [1], [2]. The energy of fully collected information scattered by ground targets disperses within the complex-valued echo, lacking the remarkable features to distinguish various targets [3]. In order to achieve clear features with sufficient resolution and signal-to-noise ratio (SNR), the echo energy needs to undergo specific processes, such as pulse compression [4] and range cell migration correction (RCMC) [5] in the complex-valued signal domain. The processed results are usually presented as real-valued intensity images, which abandon some physical signal-domain features but retain the most intuitive, basic visual features for image understanding and interpretation [6].