Issue
Natl Sci Open
Volume 3, Number 5, 2024
Special Topic: Microwave Vision and SAR 3D Imaging
Article Number 20240038
Number of page(s) 2
Section Information Sciences
DOI https://doi.org/10.1360/nso/20240038
Published online 02 August 2024

Synthetic Aperture Radar (SAR) has significant applications in terrain mapping, environmental monitoring, disaster investigation and many other fields. Traditional SAR acquires two dimensional (2D) images of the observed scenarios, while interferometric SAR (InSAR) can acquire digital surface model (DSM, 2.5D images). However, in areas with steep terrain changes or complex infrastructures, there will be severe layover phenomenon, resulting in many targets being difficult to detect and interpret. SAR 3D imaging can solve this problem and significantly enhance the target recognition and 3D modeling capabilities. It has become an important trend in the current development of SAR technology. Currently, SAR 3D imaging techniques mainly utilize multi-incident-angle observations to construct a synthetic aperture in the third dimension, so as to obtain the third dimensional resolution ability. However, dozens of tomographic flights or multi-channel observations are required, leading to long data acquisition cycles or extremely sophisticated radar systems, which restrict its popularization.

In recent years, advancements in device technology have well satisfied the computational power demands by the massive data processing, injecting new vitality into the fields of computer vision, computational electromagnetics and so on. Catching sight of this development trend, Professors Chibiao Ding and Xiaolan Qiu proposed the concept of “SAR microwave vision 3D imaging”. The core of this concept is the combination of 3D imaging and image interpretation in an intelligent way. It proposes to extract 3D clues from 2D SAR images as prior information, thereby leveraging the reliance of SAR 3D imaging on multi-angle observations, achieving efficient and low-cost 3D imaging. Furthermore, it enhances our understanding of the scenario under consideration. SAR microwave vision 3D imaging pioneers a new direction in SAR 3D imaging, and it has also brought new inspiration to SAR imaging, SAR image interpretation and other aspects.

Here we are delighted to organize this special topic on Microwave vision and SAR 3D imaging in National Science Open (NSO). The selected papers highlight significant progress and novel methodologies in microwave vision and SAR 3D imaging, paving the way for enhanced imaging radar sensing capabilities. This special topic features four papers that exemplify the diversity and depth of current research in this field.

Phase unwrapping plays an important role in array 3D SAR imaging. Zhang et al. addresses the issue of elevation ambiguity in low-altitude UAV (Unmanned Aerial Vehicle) platforms [1]. The authors propose a novel algorithm that integrates Markov Random Field (MRF) for image segmentation with optimal spatial phase unwrapping. Their results, based on real airborne array-InSAR data, show substantial improvements in resolving elevation ambiguities, thereby enhancing the coherence and resolution of 3D SAR imaging.

Non-line-of-sight (NLOS) imaging is a novel radar sensing technology that enables the reconstruction of hidden targets. In the second contribution, Liu et al. introduce a complex total variation (CTV) regularization-based sparse reconstruction method for NLOS 3D imaging radar [2]. The proposed RM-CSTV method significantly enhances imaging quality and speed by incorporating range migration and complex sparse joint total variation constraints. Through experiments with various NLOS targets, the authors demonstrate the effectiveness of their approach in providing high-resolution 3D imaging with detailed edge information. This research effectively demonstrates the significant role of scattering mechanisms in SAR 3D imaging.

As mentioned before, advancements in the field of computer vision have significantly enhanced SAR 3D imaging. In the third paper, Wang et al. explore a new paradigm in SAR 3D imaging by leveraging SAR visual semantics [3]. They investigate the extraction and utilization of scene conceptual structural information at different levels—low, middle and high. Through a case study on open-source TomoSAR datasets, they validate the effectiveness of SAR visual semantics in improving TomoSAR for 3D building modeling, offering insightful perspectives on the future of semantic-driven SAR imaging.

In the final paper, Qiu et al. review the evolution and current state of SAR microwave vision 3D imaging, emphasizing the role of artificial intelligence in recent advancements [4]. The authors propose a comprehensive theoretical framework for SAR microwave vision, encompassing representation models, computational models, processing paradigms, and evaluation systems. They also discuss their research progress and prospects, highlighting the effectiveness of microwave vision to powerfully promote SAR 3D imaging, which is leading the new development of SAR imaging technology.

Together, these papers provide a comprehensive overview of the state-of-the-art in SAR microwave vision 3D imaging technologies. They illustrate the innovative approaches being developed to overcome existing limitations. We hope this special topic will inspire further research and development in this exciting field, contributing to its continued growth and application across various domains.

We would like to thank all the authors for their valuable contributions and the reviewers for their critical insights. We are confident that the advancements presented in this special topic will significantly impact the field of SAR microwave vision 3D imaging and beyond.

References


© The Author(s) 2024. Published by Science Press and EDP Sciences.

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.