Advanced search
1 file | 1.23 MB

Superpixel quality in microscopy images: the impact of noise & denoising

Joris Roels (UGent) , Jonas De Vylder (UGent) , Jan Aelterman (UGent) , Saskia Lippens (UGent) , Yvan Saeys (UGent) and Wilfried Philips (UGent)
(2016) IFMBE Proceedings. 57. p.258-263
Author
Organization
Abstract
Microscopy is a valuable imaging tool in various biomedical research areas. Recent developments have made high resolution acquisition possible within a relatively short time. State-of-the-art imaging equipment such as serial block-face electron microscopes acquire gigabytes of data in a matter of hours. In order to make these amounts of data manageable, a more data-efficient representation is required. A popular approach for such data efficiency are superpixels which are designed to cluster homogeneous regions without crossing object boundaries. The use of superpixels as a pre-processing step has shown significant improvements in making computationally intensive computer vision analysis algorithms more tractable on large amounts of data. However, microscopy datasets in particular can be degraded by noise and most superpixel algorithms do not take this artifact into account. In this paper, we give a quantitative and qualitative comparison of superpixels generated on original and denoised images. We show that several advanced superpixel techniques are hampered by noise artifacts and require denoising and parameter tuning as a pre-processing step. The evaluation is performed on the Berkeley segmentation dataset as well as on fluorescence and scanning electron microscopy data.
Keywords
segmentation, superpixels, Fluorescence microscopy, denoising, electron microscopy

Downloads

  • MEDICON 2016.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 1.23 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Roels, Joris, Jonas De Vylder, Jan Aelterman, Saskia Lippens, Yvan Saeys, and Wilfried Philips. 2016. “Superpixel Quality in Microscopy Images: The Impact of Noise & Denoising.” In IFMBE Proceedings, ed. Efthyvoulos Kyriacou, Stelios Christofides, and Constantinos S Pattichis, 57:258–263. Cham, Switzerland: Springer.
APA
Roels, Joris, De Vylder, J., Aelterman, J., Lippens, S., Saeys, Y., & Philips, W. (2016). Superpixel quality in microscopy images: the impact of noise & denoising. In E. Kyriacou, S. Christofides, & C. S. Pattichis (Eds.), IFMBE Proceedings (Vol. 57, pp. 258–263). Presented at the 14th Mediterranean conference on Medical and Biological Engineering and Computing (MEDICON 2016), Cham, Switzerland: Springer.
Vancouver
1.
Roels J, De Vylder J, Aelterman J, Lippens S, Saeys Y, Philips W. Superpixel quality in microscopy images: the impact of noise & denoising. In: Kyriacou E, Christofides S, Pattichis CS, editors. IFMBE Proceedings. Cham, Switzerland: Springer; 2016. p. 258–63.
MLA
Roels, Joris, Jonas De Vylder, Jan Aelterman, et al. “Superpixel Quality in Microscopy Images: The Impact of Noise & Denoising.” IFMBE Proceedings. Ed. Efthyvoulos Kyriacou, Stelios Christofides, & Constantinos S Pattichis. Vol. 57. Cham, Switzerland: Springer, 2016. 258–263. Print.
@inproceedings{7023634,
  abstract     = {Microscopy is a valuable imaging tool in various biomedical research areas. Recent developments have made high resolution acquisition possible within a relatively short time. State-of-the-art imaging equipment such as serial block-face electron microscopes acquire gigabytes of data in a matter of hours. In order to make these amounts of data manageable, a more data-efficient representation is required. A popular approach for such data efficiency are superpixels which are designed to cluster homogeneous regions without crossing object boundaries. The use of superpixels as a pre-processing step has shown significant improvements in making computationally intensive computer vision analysis algorithms more tractable on large amounts of data. However, microscopy datasets in particular can be degraded by noise and most superpixel algorithms do not take this artifact into account. In this paper, we give a quantitative and qualitative comparison of superpixels generated on original and denoised images. We show that several advanced superpixel techniques are hampered by noise artifacts and require denoising and parameter tuning as a pre-processing step. The evaluation is performed on the Berkeley segmentation dataset as well as on fluorescence and scanning electron microscopy data.},
  author       = {Roels, Joris and De Vylder, Jonas and Aelterman, Jan and Lippens, Saskia and Saeys, Yvan and Philips, Wilfried},
  booktitle    = {IFMBE Proceedings},
  editor       = {Kyriacou, Efthyvoulos and Christofides, Stelios and Pattichis, Constantinos S},
  isbn         = {9783319327037},
  issn         = {1680-0737},
  keyword      = {segmentation,superpixels,Fluorescence microscopy,denoising,electron microscopy},
  language     = {eng},
  location     = {Paphos, Cyprus},
  pages        = {258--263},
  publisher    = {Springer},
  title        = {Superpixel quality in microscopy images: the impact of noise \& denoising},
  url          = {http://dx.doi.org/10.1007/978-3-319-32703-7\_52},
  volume       = {57},
  year         = {2016},
}

Altmetric
View in Altmetric
Web of Science
Times cited: