N. B. Ampilova, V. D. Sergeev, I. P. Soloviev
Price: 50 руб.
The paper describes the method of digital image analysis, which is based on the representation of an image by an oriented graph. Vertices correspond to image pixels, edges connect the nearest neighbors. We assign a measure to all edges so that to obtain a Markov chain on the graph. In accordance with the initial measure distribution the stationary fl ow is constructed and weighted entropy is calculated. The algorithm is implemented both for the base case (vertex corresponds to one pixel) and the optimized one — vertex corresponds to a cell of the image partition. The choice of the maximum allowed cell size depends on the image structure and may be obtained experimentally — comparing the weighted entropy values and run times for the base and optimized variants. The results of calculations for some classes of biomedical preparations images are given. The described optimization reduces the run time by 3–4 times.
Keywords: Digital images, oriented graph, Markov chain, stationary distribution,
weighted entropy.
REFERENCES
1. Marcus, B., & Lind, D. An introduction to symbolic dynamics and coding (1st
ed.). 1995: Cambridge University Press.
2. Ampilova, N., & Soloviev, I. On application of entropy characteristics to texture
analysis. WSEAS Transactions on Biology and Biomedicine, 2014, 11(1), 194–202.
3. Ampilova, N.B. Stationary processes on graphs and image analysis [Стационарные
процессы на графах и анализ изображений]. Computer Tools in Education, 2013,
2, 24–29.
4. Sheleikhovsky, G.V. The composition of city plan as the transport problem
[Композиция городского плана как проблема транспорта]. 1946, M.: Giprogor.
5. Bregman, L.M. The proof of the convergence of G.V.Sheleihovsky method for the
task with transport restrictions [Доказательство сходимости метода Шелейховского
для задачи с транспортными ограничениями]. Journal of Computational Mathematics
and Mathematical Physics, 1967, 7(1), 147–156.
6. Bregman, L.M. Relaxation method for detecting a common point of convex set
and its application to the solving convex programming tasks [Релаксационный метод
нахождения общей точки выпуклых множеств и его применение для решения
задач выпуклого программирования]. Journal of Computational Mathematics and
Mathematical Physics, 1967, 7(3), 620–631.
7. Romanovskii, I.V. The optimization of the stationary control by a discrete
deterministic process of dynamical programming [Оптимизация стационарного
управления дискретным детерминированным процессом]. Cybernetic, 1967, 2,
66–78.
8. Ampilova, N.B., Romanovskii, I.V., & Petrenko, E.I. On the maximiza-
tion of entropy for linear restrictions [О максимизации энтропии при линейных
ограничениях]. Proceedings of the International Science Conference “Cosmos,
astronomy and programming”, 2008, pp. 181–185.