EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Mix MSTAR: A Synthetic Benchmark Dataset for Multi-Class Rotation Vehicle Detection in Large-Scale SAR Images

Zhigang Liu, Shengjie Luo, Yiting Wang

Remote Sensing · 16.09.2023 · 15 (18) · 4558

DOI: 10.3390/rs15184558 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 16.09.2023. Verfasst von Zhigang Liu und 2 weiteren Autor:innen. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Young Talent Fund of the University Association for Science and Technology in Shaanxi, China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Because of the counterintuitive imaging and confusing interpretation dilemma in Synthetic Aperture Radar (SAR) images, the application of deep learning in the detection of SAR targets has been primarily limited to large objects in simple backgrounds, such as ships and airplanes, with much less popularity in detecting SAR vehicles. The complexities of SAR imaging make it difficult to distinguish small vehicles from the background clutter, creating a barrier to data interpretation and the development of Automatic Target Recognition (ATR) in SAR vehicles. The scarcity of datasets has inhibited progress in SAR vehicle detection in the data-driven era. To address this, we introduce a new synthetic dataset called Mix MSTAR, which mixes target chips and clutter backgrounds with original radar data at the pixel level. Mix MSTAR contains 5392 objects of 20 fine-grained categories in 100 high-resolution images, predominantly 1478 × 1784 pixels. The dataset includes various landscapes such as woods, grasslands, urban buildings, lakes, and tightly arranged vehicles, each labeled with an Oriented Bounding Box (OBB). Notably, Mix MSTAR presents fine-grained object detection challenges by using the Extended Operating Condition (EOC) as a basis for dividing the dataset. Furthermore, we evaluate nine benchmark rotated detectors on Mix MSTAR and demonstrate the fidelity and effectiveness of the synthetic dataset. To the best of our knowledge, Mix MSTAR represents the first public multi-class SAR vehicle dataset designed for rotated object detection in large-scale scenes with complex backgrounds.

Förderung

  • National Natural Science Foundation of China
  • Young Talent Fund of the University Association for Science and Technology in Shaanxi, China

Zitieren

APA

Liu, Z., Luo, S., & Wang, Y. (2023). Mix MSTAR: A Synthetic Benchmark Dataset for Multi-Class Rotation Vehicle Detection in Large-Scale SAR Images. Remote Sensing, 15(18), 4558. https://doi.org/10.3390/rs15184558

Vancouver

Liu Z, Luo S, Wang Y. Mix MSTAR: A Synthetic Benchmark Dataset for Multi-Class Rotation Vehicle Detection in Large-Scale SAR Images. Remote Sensing. 2023;15(18):4558. doi:10.3390/rs15184558

BibTeX
@article{Liu2023,
  author = {Liu, Zhigang and Luo, Shengjie and Wang, Yiting},
  title = {Mix MSTAR: A Synthetic Benchmark Dataset for Multi-Class Rotation Vehicle Detection in Large-Scale SAR Images},
  journal = {Remote Sensing},
  year = {2023},
  volume = {15},
  number = {18},
  pages = {4558},
  doi = {10.3390/rs15184558}
}
Kostenlos starten →