Automated Identification and Classification of Diatoms from Water Resources

  • Jose LibrerosEmail author
  • Gloria Bueno
  • Maria TrujilloEmail author
  • Maria Ospina
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11401)


The quantity of certain types of diatoms is used for determining water quality. Currently, a precise identification of species present in a water sample is conducted by diatomists. However, different points of view of diatomists along with different sizes and shapes that diatoms may have in samples makes diatoms identification difficult, which is required to classify them into genera to which they belong to. Additionally, chemical processes, that are applied to eliminate unwanted elements in water samples (debris, flocs, etc.) are insufficient. Thus, diatoms have to be differentiated from those structures before classifying them into a genus. In fact, researchers have a special interest on looking for different ways to perform an automated identification and classification of diatoms. In spite of applications, an automatic identification of diatom has a high level of difficulty, due to the present of unwanted elements in water samples. After diatoms have been identified, diatoms classification into genera is an additional problem.

In this paper, an automatic method for identification and classification diatoms from images is presented. The method is based on the combination of Scale and Curvature Invariant Ridge Detector (SCIRD-TS), following by a post processing method, and the use of a nested Convolutional Neural Networks (CNN). Whilst the identification approach is able to identify well-defined ridge structures, the nested CNN is able to classify a diatom into the genus to which it belongs to.


Diatoms Handcraft filters Nested CNN Paleo-environmental studies Water quality monitoring 



The first author thanks to Santander Bank for the financial support for his mobility to Universidad de Castilla-La Mancha, Ciudad Real, Spain.

Gloria Bueno acknowledges financial support of the Spanish Government under the Aqualitas-retos project (Ref. CTM2014-51907-C2-2-R-MINECO).

The authors acknowledge the contribution to this work of Dr. E. Peña from Universidad del Valle. The authors are also grateful to the anonymous reviewers for their valuable comments, suggestions and remarks, which contributed to improve the paper.


  1. 1.
    Smol, J.P., Stoermer, E.F. (eds.): The Diatoms: Applications for the Environmental and Earth Sciences, vol. 17, pp. 283–284. Cambridge University Press, Cambridge (2010)Google Scholar
  2. 2.
    Annunziata, R., Trucco, E.: Accelerating convolutional sparse coding for curvilinear structures segmentation by refining SCIRD-TS filter banks. IEEE Trans. Med. Imag. 35, 2381–2392 (2016)CrossRefGoogle Scholar
  3. 3.
    Bueno, G., et al.: Automated diatom classification (Part A): handcrafted feature approaches. Appl. Sci. 7, 753 (2017)CrossRefGoogle Scholar
  4. 4.
    Pedraza, A., Bueno, G., Deniz, O., Cristóbal, G., Blanco, S., Borrego-Ramos, M.: Automated diatom classification (Part B): a deep learning approach. Appl. Sci. 7, 460 (2017)CrossRefGoogle Scholar
  5. 5.
    Rojas Camacho, O., Forero, M., Menéndez, J.: A tuning method for diatom segmentation techniques. Appl. Sci. 7, 762 (2017)CrossRefGoogle Scholar
  6. 6.
    Pech-Pacheco, J.L., Alvarez-Borrego, J.: Optical-digital system applied to the identification of five phytoplankton species. Mar. Biol. 132, 357–365 (1998)CrossRefGoogle Scholar
  7. 7.
    Pappas, J.L., Stoermer, E.F.: Legendre shape descriptors and shape group determination of specimens in the Cymbella cistula species complex. Phycologia 42, 90–97 (2003)CrossRefGoogle Scholar
  8. 8.
    Du Buf, H., et al.: Diatom identification: a double challenge called ADIAC. In: 10th International Conference on Image Analysis and Processing, pp. 734–739 (1999)Google Scholar
  9. 9.
    Lai, Q.T.K., Lee, K.C.M., Tang, A.H.L., Wong, K.K.Y., So, H.K.H., Tsia, K.K.: High-throughput time-stretch imaging flow cytometry for multi-class classification of phytoplankton. Opt. Soc. Am. 24, 28170–28184 (2016)Google Scholar
  10. 10.
    Davidson, M.W., Abramowitz, M.: Molecular expressions microscopy primer: digital image processing-difference of gaussians edge enhancement algorithm. Olympus America Inc., and Florida State University (2006)Google Scholar
  11. 11.
    Alvarez, T., et al.: Classification of microorganisms using image processing techniques. In: 2001 International Conference on Image Processing (Cat. No. 01CH37205), Thessaloniki, vol. 1, pp. 329–332. IEEE Conferences (2001)Google Scholar
  12. 12.
    Hawickhorst, B.A., Zahorian, S.A., Rajagopal, R.: A comparison of three neural network architectures for automatic speech recognition. In: Intelligent Engineering Systems Through Artificial Neural Networks, vol. 5, pp. 221 (1995). In: Advances in Neural Information Processing Systems. Neural Information Processing Systems Foundation Inc., La Jolla, CA, USA, pp. 1097–1105 (2012)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Multimedia and Computer Vision GroupUniversidad del ValleCaliColombia
  2. 2.Grupo de Visión y Sistemas InteligentesUniversidad de Castilla La ManchaCiudad RealSpain
  3. 3.Grupo de Investigación en Biología de Plantas y MicroorganismosUniversidad del ValleCaliColombia

Personalised recommendations