EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ISTD-PDS7: A Benchmark Dataset for Multi-Type Pavement Distress Segmentation from CCD Images in Complex Scenarios

Weidong Song, Zaiyan Zhang, Bing Zhang, Guohui Jia, Hongbo Zhu, Jinhe Zhang

Remote Sensing · 24.03.2023 · 15 (7) · 1750

DOI: 10.3390/rs15071750 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 24.03.2023. Verfasst von Weidong Song und 5 weiteren Autor:innen. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 83 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Basic Scientific Research Expenses of Heilongjiang Provincial Universities, China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The lack of large-scale, multi-scene, and multi-type pavement distress training data reduces the generalization ability of deep learning models in complex scenes, and limits the development of pavement distress extraction algorithms. Thus, we built the first large-scale dichotomous image segmentation (DIS) dataset for multi-type pavement distress segmentation, called ISTD-PDS7, aimed to segment highly accurate pavement distress types from natural charge-coupled device (CCD) images. The new dataset covers seven types of pavement distress in nine types of scenarios, along with negative samples with texture similarity noise. The final dataset contains 18,527 images, which is many more than the previously released benchmarks. All the images are annotated with fine-grained labels. In addition, we conducted a large benchmark test, evaluating seven state-of-the-art segmentation models, providing a detailed discussion of the factors that influence segmentation performance, and making cross-dataset evaluations for the best-performing model. Finally, we investigated the effectiveness of negative samples in reducing false positive prediction in complex scenes and developed two potential data augmentation methods for improving the segmentation accuracy. We hope that these efforts will create promising developments for both academics and the industry.

Förderung

  • National Natural Science Foundation of China
  • Basic Scientific Research Expenses of Heilongjiang Provincial Universities, China

Zitieren

APA

Song, W., Zhang, Z., Zhang, B., Jia, G., Zhu, H., & Zhang, J. (2023). ISTD-PDS7: A Benchmark Dataset for Multi-Type Pavement Distress Segmentation from CCD Images in Complex Scenarios. Remote Sensing, 15(7), 1750. https://doi.org/10.3390/rs15071750

Vancouver

Song W, Zhang Z, Zhang B, Jia G, Zhu H, Zhang J. ISTD-PDS7: A Benchmark Dataset for Multi-Type Pavement Distress Segmentation from CCD Images in Complex Scenarios. Remote Sensing. 2023;15(7):1750. doi:10.3390/rs15071750

BibTeX
@article{Song2023,
  author = {Song, Weidong and Zhang, Zaiyan and Zhang, Bing and Jia, Guohui and Zhu, Hongbo and Zhang, Jinhe},
  title = {ISTD-PDS7: A Benchmark Dataset for Multi-Type Pavement Distress Segmentation from CCD Images in Complex Scenarios},
  journal = {Remote Sensing},
  year = {2023},
  volume = {15},
  number = {7},
  pages = {1750},
  doi = {10.3390/rs15071750}
}
Kostenlos starten →