UDK 004.932.2
IMAGE TEXTURE RECONSTRUCTION BASED ON MULTIAGENT APPROACH
A. N. Bolgov
Siberian State Aerospace University named after academician M. F. Reshetnev 31, Krasnoyarsky Rabochiy Av., Krasnoyarsk, 660014, Russian Federation E-mail: anbolgov@gmail.com
Development and implementation of a multiagent system that designed for texture restoration of image background when deleting objects are examined in the thesis. General problems of photo and video reconstruction and its application are reviewed. The analysis of existing methods for restore gaps in big data arrays is held. Solution by using one of popular method of clustering analysis: special self-organizing neural network is offered. Algorithm of learning Kohonen map is covered step by step: preparation initial map data, best matching unit choice, shifting adjacent nodes by neighborhood function, determination of training measures, ordering and fine tuning. Adaptation and modification of the Kohonen map algorithm for image restoring is proposed. Algorithms of initialization of map, choice of content of training vectors, best matching unit algorithm, neighborhood functions are highlighted in this paper. Kohonen multimap principle and cooperative solution by using multiagent paradigm is considered. Mutliagent algorithm is expanded by presegmentation step for defining of adjacent areas. Architecture of a multiagent system is developed, behavior of individual agents and communication ways are determined. Multiagent system that allows restoring background pixels of is implemented. Results of series of experiments are presented. Experiments were carried out on the database of images to search for objects of interest (CBIR database). Assessments of efficiency parameters of multimapping are presented. Evaluation of the optimal number of maps for by multimapping is based on a percentage of the original image palette coverage. The accuracy and performance of agent-based mechanism for the subsystems of training and reconstruction are represented. Assessments of accuracy metrics using SSIM, MSE, PSNR and method by pixel-per-pixel comparison are conducted. Performance estimation for various configurations of the multi-system is conducted in order to determine the most effective ones.
image restoration, texture restoration, Kohonen maps, multiagent systems.
References
- Forsayt D. A. Komp'yuternoe zrenie. Sovremennyy podkhod [Computer Vision. Modern view]. Moscow, Williams Publ., 2004, 928 p.
- Favorskaya M. N., Petukhov N. Yu. [Recognition of nature objects on airphotos with using neural networks]. Avtometriya. 2011, vol. 47, no. 3, p. 34–40 (In Russ.).
- Kohonen T. Self-Organizing Maps. Springer Series in Information Sciences. 2001, 501 p.
- Rassel S. Iskusstvennyy intellekt: sovremennyy podkhod [Artificial Intelligence: Modern view]. Moscow, Williams, 2006, 1408 p.
- Andreas L. Agent intelligence through data mining. Munchen, Technische Universität München, 2008, 215 p.
- Bolgov A. N. [Using of Kohonen maps to restore background on images]. V materialakh nauchno-tekhnicheskoy konferentsii “Tekhnicheskoe zrenie v sistemakh upravleniya – 2014” [Proceeding of “Technical Vision in Control Systems – 2014”]. Moscow, 2014, p. 437 (In Russ.).
- Littl R. Dzh. A., Rubin D. B. Statisticheskiy analiz dannykh s propuskami [Statistical analysis of data with gaps]. Moscow, Finansy i statistika Publ., 1991, 430 p.
- Smigiel E. Self-Organizing Maps and Ancient Documents. Proceeding 6th International Workshop on Document Analysis Systems VI. Berlin, 2004, p. 125–134.
- Abonyi J., Feil B. Cluster Analysis for Data Mining and System Identification. Oxford, Blackwell, 2007, 320 p.
- Serra J. Image analysis and mathematical morphology. London, Academic Press Inc., 1982, 610 p.
- Nisbet R., Elder J., Miner G. Handbook of Statistical Analysis and Data Mining Applications. London, Academic Press Inc., 2009, 864 p.
- Avcibas I., Sankur B., Akarun L., Anarim E. Image Quality Statistics and Their Use in Steganalysis and Compression. Istanbul, Boğaziçi University, 2001, p. 211–218.
- Szeliski R. Computer Vision Algorithms and Applications. Berlin, Springer, 2011, 812 p.
- Datta R., Joshi D., Wang J. Z. Image retrieval: Ideas, influences, and trends of the new age. ACM Computing Surveys. 2008, vol. 40, p. 1–60.
- Horton N. J, Lipsitz S. R. Multiple Imputation in Practice: Comparison of Software Packages for Regression Models with Missing Variables. The American Statistician. 2001, vol. 55, p. 244–254.
- Shamir A., Avidan S. Seam carving for content-aware image resizing. Proceeding SIGGRAPH '07 ACM SIGGRAPH 2007 papers, 2007, P. 341–350.
Bolgov Andrey Nikolaevich – postgraduate student, Siberian State Aerospace University named after academician M. F. Reshetnev. E-mail: anbolgov@gmail.com
