Deep Belief Networks for Multimodal, Images-based Non Contact Material Characterization

Authors

  • Magdelena Stoyanova CISBI Ca’ Foscari University, Venice, Italy
  • Lilia Pavlova Laboratory of Telematics, Bulgarian Academy of Sciences, Sofia, Bulgaria

DOI:

https://doi.org/10.55630/dipp.2017.7.17

Keywords:

Deep Belief Networks, Knowledge Representation and Reasoning, Knowledge Extraction

Abstract

Our growing cognisance about the chemo-physical properties of electromagnetic waves and of their interaction with various materials provides expanding range of possibilities for quantification and non contact material characterization even in difficult to computation, « non-standard » domains as this of the cultural heritage. This review - thematically part of the IFIDA project dedicated to digitization of archaeometric and conservation-restoration praxis - deals with the possibilities latest neuroinformatic methods offer for a more efficient and fast interrogation, understanding and classification of multi-modal spectral records of paintings on various supports. The workflow followed during their routine non destructive (ND) analysis by human experts is described in terms of specific technical characteristics, research objectives and concrete application for the artworks' characterization as a prototype to simulate artificially. References to techniques for high semantic level information extraction from low-level features are suggested. The strengths and limitations in application of Deep Belief Networks (DBNs) for compression, visualization and data recognition are also considered. Examples of most appropriate architectures and topological maps of Deep Learning Networks for interpretation of UV, IR, XR, β, γ and CT records performing simple characterization and classification tasks - from single layer perceptron to multi-layer deep belief neural networks for unlabelled data - are presented and explained.

References

Amin, J., & Zafar, A. (2014). A survey: content based image retrieval. Int. J. Advanced Networking and Applications, 5 (6), 2076-2083.

ANN. (n.d.). Artificial Neural Networks . Retrieved 06 13, 2017, from (Искусственная нейронная сеть): https://ru.wikipedia.org/wiki /Искусственная_нейронная_сеть

Burdajewicz, J. (2009). MetigoMAP - an innovative graphic software for conservation documentation. Conservation documentation: on-going projects and perspectives. Contributions to the ICOM-CC Working Group Paintings Meeting.

Erastov, D. (1997). The use of photographic technologies for investigation of historical documents, Conservation of cultural heritage: science and praxis. 2nd workshop Exposition and conservation of cultural and historic monuments . Sanct Peterburg, Russia.

FANN. (n.d.). Fast Artificial Neural Network Library. Retrieved 06 13, 2017, from http://leenissen.dk/fann/wp/

Golovko, V. (2015). From single layer perceptrons to Deep Belief Networks: paradigms of training and application (От многослойных персептронов к нейронным сетям глубокого доверия: парадигмы обучения и применение). In Lectures in neuroinformatics (Лекции по нейроинформатике) (pp. 47-84). Moscow, Russia: MIFI.

Gonsales, R. (2005). Digital assessment of images (Цифровая обработка изображений). In R. Gonsales, & R. W. (Eds.), Digital assessment of images (p. 1072). Moscow, Russia: Technosfera.

Grenberg, J. (2000). Technology and investigation of easel and wall painting (Технология и исследование произведений станковой и настенной живописи). Мoscow, Russia.

Heaton, J. (2015). Encog: Library of Interchangeable Machine Learning Models for Java and C#. Journal of Machine Learning Research , 1243−1247.

Hendriks, E. (2009). Automated thread counting from x-rays of canvas picture supports Van Gogh Museum Amsterdam. Conservation documentation: on-going projects and perspectives. Contributions to the ICOM-CC Working Group Paintings Meeting.

Hinton, G. E., Osindero, S., & Teh, Y. W. (2006). Fast learning algorithm for deep belief nets. Neural computation, 18 (7), 1527-1554.

Imran, M., Hashim, R., Irtaz, A., Azhar, M., & Abdullah, U. (2017). Class wise image retrieval through scalable color descriptor and edge histogram descriptor. International Journal of Advanced and Applied Sciences, 3 (12), 32-36. Retrieved from http://www.science-gate.com/IJAAS.html

Lebedeva, V. (1986). Technique of optical spectroscopy (Техника отпической спектроскопии) (2nd ed.). Moscow, Russia: Moscow University.

LeCun, Y., & Bottou, L. e. (1998). Efficient Back Prop. G. Orr and K. Muller (Eds.): Neural Neworks: Tricks of the Trade. Springer.

LeCun, Y., Bottou, L., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE , 86(11) , pp. 2278–2324.

Mestezkij, L. (2004). Mathematical methods for image recognition ( Математические методы распознавания образов). Moscow, Russia: Moscow State University.

Montagnuolo, M. (2005). Introduzione alle tecniche di Image Analysis per la classificazione automatica degli archivi audiovisivi. Elettronica e telecomunicazioni, 2 , 23-34.

Pyykkö, J., & Głowacka, D. (2017). Interactive Content -Based Image Retrieval with Deep Neural Networks. In L. e. (Eds.), Symbiotic 2016 (pp. 77-88). LNCS 9961.

Rajchenok, T., Skornijakov, I., Tolstorozhev, G., Gorelenko, A., Zaharich, M., & Rak, A. (2010). Fotostability of printing colours using luminescence and IR spectroscopic methods (Фотостойкость печатных красок при использовании методов люминесценции и ик - спектроскопии). 8th International Scientific Conference "Laser Physics and Optical Technologies”. Minsk: Belarus Institute of Physics B.N.Stepanov.

Schüller, H. (1997). Radiologie im Dienst der Kunst — Ikonen. Ivan Bentchev und Eva Haustein- Bartsch (Eds.) Ikonen. Restaurierung und naturwissenschaftliche Erforschung. Beiträge des Internationalen Kolloquiums in Recklinghausen (pp. 11-16). München, Germany: Editio Maris.

Sheshkus, A., Limonova, E., Nikolaev, D., & Krivtsov, V. (2016). Combining convolutional neural networks and Hough Transform for classification of images containing lines. Proceedings of the Ninth International Conference on Machine Vision (ICMV). Nice, France.

Sil'chenko, T. (1955). Investigation of easel paintings by XR and UV (Исследование картин рентгеновскими и ультрафиолетовыми лучами). In Restoration and investigation of artworks. Moscow, Russia.

Sirotenko, M. (2009). Application of neural networks in image recognition (Применение нейросетей в распознавании изображений). Retrieved from Habrahabr page: http://habrahabr.ru/post/74326/

Stoyanova, M. (2014). Towards a more efficient use of spectral techniques in the attribution of easel painting. COSCH Workshop & Working Group Meeting. Belgrade, Serbia. Retrieved 06 13, 2017, from http://www.info.cosch/

Stoyanova, M. (2015). Spectral investigation of Serbian Baroque icons for their scientific documentation. Technical Report (Reference COST-STSM- TD1201-48807, Affiliation: COST TD1201). Retrieved from http://artcon.ru/node/6105 last accessed 2017/06/13

Stoyanova, M. (2017). An integrated technical-technological investigation of the Archangel icon in the National museum in Belgrade (Комплексное технико - технологическое исследование иконы архангела из национального музея в Белграде). München, Germany: GRIN Verlag.

Stoyanova, M., Luchev, D., & Paneva-Marinova, D. (2016). Optimization of Natural Dyes' Non-contact Characterization and Interdisciplinary Application Using Ensemble Classifiers and Genetic Algorithms. Proceedings of the Sixth International Conference Digital Presentation and Preservation of Cultural and Scientific Heritage – DiPP2016. , pp. 147-160. Veliko Tarnovo, Bulgaria: Institute of Mathematics and Informatics, Bulgarian Academy of Sciences.

Stoyanova, M., Paneva-Marinova, D., Pavlova, L., & Pavlov, R. (2014). The IFIDA Project: Intelligent Fast Interconnected Devices and Tools for Applications in Archaeometry and Conservation Practice. Proceedings of the Fourth International Conference Digital Presentation and Preservation of Cultural and Scientific Heritage – DiPP2014. , pp. 256-262. Velito Tarnovo, Bulgaria: Institute of Mathematics and Informatics, Bulgarian Academy of Sciences.

Stoyanova, M., Stojanović, M., & Provorova, I. (2015). Technical visualization of easel painting functional to the mapping of its conservation status and attribution. COSCH 5th WG Meeting 2015. St. Étienne, France.

Van den Bulcke, J., Boone, J., Van Acker, J., & Van Hoorebeke, L. (2010). Highresolution X-ray imaging and analysis of coatings on and in wood. Journal of Coatings Technology and Research, 7 (2), 271–277.

Downloads

Published

2017-09-10

How to Cite

Stoyanova, M., & Pavlova, L. (2017). Deep Belief Networks for Multimodal, Images-based Non Contact Material Characterization. Digital Presentation and Preservation of Cultural and Scientific Heritage, 7, 191–204. https://doi.org/10.55630/dipp.2017.7.17

Most read articles by the same author(s)

<< < 1 2 3 > >>