FORMAL REPRESENTATION OF THE PIXEL-BY-PIXEL CLASSIFICATION PROCESS USING A MODIFIED WANG-MENDEL NEURAL NETWORK

Authors

DOI:

https://doi.org/10.30837/ITSSI.2020.13.122

Keywords:

segmentation, classification, fuzzy set of the second type, fuzzy neural network, production model

Abstract

The subject of research in the article are the processes of formalization of the pixel-by-pixel classification problem using the modified fuzzy neural production network of Wang-Mendel for segmentation of urban structures in the automated analysis of space and aerial photographs of the city. The purpose of the work is to develop the architecture of the modified fuzzy neural production network of Wang-Mendel as a classifier for image segmentation to increase the values of efficiency and reliability of urban monitoring. The following tasks are solved in the article: analysis of possibilities of Wang-Mendel network modification based on representation of membership functions in terms of interval fuzzy sets of the second type (IFST2) and realization of phasing, aggregation and activation operations using IFST 2 operations, development of the architecture of the modified fuzzy neural production network of Wang-Mendel as a classifier for image segmentation. The following methods and models are used: methods and models of fuzzy set theory (fuzzy Wang-Mendel neural network, interval fuzzy sets of the second type), methods and models of deep learning methodology (convolutional neural network for image segmentation (auto coder) U-net). The following results were obtained: the use of a fuzzy Wang-Mendel neural network as a classifier of a modified U-Net decoder based on the representation of membership functions in IFST2 and the implementation of phasing, aggregation and activation operations using operations on IFST2; introduction of an additional operation of type reduction in the phase of dephasification of the original variable based on the classical method of the center of gravity (centroid); introduction of several outputs of the network to recognize the appropriate number of classes (subclasses) of the subject area. To do this, the third layer is represented as a set of several pairs of adder neurons, and the fourth implements several normalizing neurons, the number of which corresponds to the number of pairs of the third layer. Conclusions: the use in the architecture of a convolutional neural network for segmentation of U-net images as a classifier of the modified fuzzy neural production network of Wang-Mendel will provide an additional increase in the accuracy of pixel-by-pixel classification of certain objects. Instead of fuzzy sets of the first type (FST1) in this network IFST2 are used. The proposed IFST2, on the one hand, provide a formalization of more additional degrees of uncertainty compared to FST1, on the other hand, are "implemented" in the development of fuzzy systems (models) and have less computational complexity, compared to fuzzy sets of the second type (FST2).

Author Biographies

Oleksii Kolomiitsev, National Technical University is the "Kharkiv Polytechnic Institute"

Honored Inventor of Ukraine, Doctor of Sciences (Engineering), Senior Research, Professor of the Department of Computing Engineering and Programming

Volodymyr Pustovarov, Kharkov Representative Office of General Customer – The State Space Agency of Ukraine

Bread-Winner of Scientific Degree of PhD of Engineering Sciences, Chief of Group

References

Mishchenko, V. A. Mishchenko, V. A., Astakhova, I. F., Krasnoyarov, A. A. (2012), "Learning algorithm for the Wang-Mendel fuzzy neural network for recognizing handwritten characters in the work of the postal service" ["Algoritm obucheniya nechetkoy neyronnoy seti Vanga-Mendelya dlya raspoznavaniya rukopechatnykh simvolov v rabote pochtovoy sluzhby"], Bulletin of Voronezh State University. Series: System Analysis and Information Technologies, No. 1, P. 141–146.

"Semantic segmentation: a quick guide" ["Semanticheskaya segmentatsiya: kratkoye rukovodstvo"], available at : https://neurohive.io/ru/osnovy-data-science/semantic-segmention/.

Ronneberger, O., Fischer, P., Brox, T. (2015), "U-net: Convolutional networks for biomedical image segmentation", In International Conference on Medical image computing and computer-assisted intervention, Springer, P. 234–241.

Kuksa, P. P. "Analysis of neural network-fuzzy models" ["Analiz neyrosetevykh-nechetkikh modeley"], Moscow State Technical University, available at : http://pkuksa.org/~pkuksa/publications/nfm-analysis-c-0-sept-03.pdf

Kruglov, V. V., Dli, M. I., Golunov, R. Yu. (2001), Fuzzy logic and artificial neural networks [Nechetkaya logika i iskusstvennyye neyronnyye seti], Fizmatlit, 224 p.

Olizarenko, S. A., Kapranov, V. A., Safronov, R. V. (2016), "Development of a fuzzy convolutional neural network architecture for the recognition of compact (point) objects in a digital aerial photograph" ["Rozrobka arkhitektury nechitkoyi z·hortochnoyi neyronnoyi merezhi dlya rozpiznavannya kompaktnykh (tochkovykh) obʺyektiv na tsyfrovomu aerofotoznimku"], Systems of Arms and Military Equipment, Kharkiv, HUPS, Issue 4 (48), P. 38–41.

Ivanov, E. S., Tishchenko, I. P., Vinogradov, A. N. (2019), "Segmentation of multispectral images using convolutional neural networks" ["Segmentatsiya mul'tispektral'nykh snimkov s primeneniyem svortochnykh neyronnykh setey"], Modern problems of remote sensing of the Earth from space, Vol. 16, No. 1, P. 25–34.

Soloviev, R. A., Telpukhov, D. V., Kustov, A. G. (2017), "Automatic segmentation of satellite images based on a modified convolutional neural network UNET" ["Avtomaticheskaya segmentatsiya sputnikovykh snimkov na baze modifitsirovannoy svortochnoy neyronnoy seti UNET"], Engineering Bulletin of the Don, No. 4, available at : ivdon.ru/ru/magazine/archive/n4y2017/4433

Long, J., Shelhamer, E., Darrell, T. (2015), "Fully Convolutional Networks for Semantic Segmentation", Computer Vision and Pattern Recognition (cs.CV), available at : https://arxiv.org/abs/1411.4038

Gordienko, A. S. (2015), "The use of index images for detecting changes in space images of different times" ["Primeneniye indeksnykh izobrazheniy pri vyyavlenii izmeneniy po raznovremennym kosmicheskim snimkam"], Interexpo GEO-Siberia-2015. XI Int. scientific. Congr. : Int. scientific. conf. "Remote sensing methods of the Earth and photogrammetry, environmental monitoring, geoecology": collection of articles. materials in 2 volumes (Novosibirsk, April 13-25, 2015), Novosibirsk : SGUGiT, Vol. 1, P. 67–70.

Gavrilova, T. A., Khoroshevsky, V. F. (2000), Knowledge bases of intelligent systems [Bazy znaniy intellektual'nykh sistem], SPb. : Peter, 384 p.

Iskander, Yu. M. (2003), Creation of knowledge bases of intelligent systems [Sozdaniye baz znaniy intellektual'nykh sistem], Ministry of Defense of the Russian Federation, 233 p.

Milletari, F., Navab, N., Ahmadi, S. (2016), "V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation", 2016 Fourth International Conference on 3D Vision (3DV), Stanford, CA, P. 565–57.

Takagi, T., Sugeno, M. (1985), "Fuzzy Identification of Systems and Its Applications to Modeling and Control", IEEE Trans. on Systems, Man, and Cybernetics, Vol. 15, No. 1, P. 116–132.

Mendel, J. M., Hagras, H., John, R. I. "Standard Background Material About Interval Type-2 Fuzzy Logic Systems", IEEE CIS Standards Committee, available at : http://ieee-cis.org/technical/standards

Downloads

How to Cite

Kolomiitsev, O., & Pustovarov, V. (2020). FORMAL REPRESENTATION OF THE PIXEL-BY-PIXEL CLASSIFICATION PROCESS USING A MODIFIED WANG-MENDEL NEURAL NETWORK. INNOVATIVE TECHNOLOGIES AND SCIENTIFIC SOLUTIONS FOR INDUSTRIES, (3 (13), 122–128. https://doi.org/10.30837/ITSSI.2020.13.122

Issue

Section

ENGINEERING & INDUSTRIAL TECHNOLOGY