ROBUST CLASSIFICATION OF TEXTURE LAND FOREST INVENTORY BASED ON MODEL OF MINIMALLY SUFFICIENT FEATURES
Abstract
Method for automated classification of ground forest inventory images based on the proposed mathematical model developed. The general model is represented by the statistical characteristics of images and fractal dimension of texture. Experimental means were determined minimally sufficient characteristics to solve the problem of robust classification. Neural network based on unsupervised self-organizing maps used as a classifier. Figures obtained discounts of the proposed approach on real digital images.
References
1. John A. Richards, Xiuping Jia Remote Sensing Digital Image Analysis. – Berlin: Springer, 2006. – 454 p.
2. Schowengerdt RA. Remote Sensing – Models and Methods for Image Processing. 2nd ed. London, San Diego:Academic Press; 1997. ISBN: 0-12-628981-6.
3. IpatovY.A., KrevetskyA.V. Algoritm lokalizatsii granits teksturnykh uchastkov drevesiny na ikh tsifrovykh izobrazheniyakh // Izv. vuzov. Priborostroyeniye. 2009. Volumе. 52, № 7. – pp. 12-17.
Ipatov Y.A., Krevetsky A.V. Segmentation of color aerial photographs on the basis of selective training algorithm // Vestnik of Volga State University of Technology. Series Radio Engineering and Infocommunication Systems. № 2 2008. pp.22-26.
Krevetsky A.V.,Ipatov Y.A. Integration Optical-Location Detection and Estimation Parameters of Ground Based Forest Taxation // Achievements of Modern Radioelectronics 2011. №5. – p. 56-60.
4. Ipatov Y.A. Process automation of an estimation for relative density of wood plantings the instrumentality of methods of digital image processing // 8-th International Conference «PRIA: New Information Technologies»: Conference Proceedings. Vol. 1. – Yoshkar-Ola, 2007. – P.307-309.
5. Krevetsky A.V., Ipatov Y.A. High Technologies in Measuring Problems of Forestry Complex on The Basis of Scene Analysis and Image Recognition Method// 8-th International Conference «PRIA: New Information Technologies»: Conference Proceedings. Vol. 2. – Yoshkar-Ola, 2007. – P.287-289.
6. Novyye teksturnyye kharakteristiki i ikh ispol'zovaniye v klassifikatsii tekstur, invariantnykh k izmeneniyu ugla povorota, DzhiangoZhang, TiyenuTan, 2001 [Elektronnyy resurs]. – Rezhim dostupa: http://lear.inrialpes.fr/people/zhang/ECCV_paper.pdf
Chi-Man Pun ; Moon-Chuen Lee 2003 Log-polar wavelet energy signatures for rotation and scale invariant texture classification , // IEEE Transactionson Pattern Analysisand Machine Intelligence, vol. 25, no. 5, May 2003. 590 – 603 DOI: 10.1109/TPAMI.2003.1195993
/9/ Haralick R.M., Shanmugam K., Dinstein I., Textural features for image classification // IEEE Trans.Syst.ManCybern. v.3, pp.610-621, 1973.
/10/ Ilyasova N.Y., Kupriyanov A.V., Khramov A.G., Klassifikatsiya kristallogramm s ispol'zovaniyem metodov statisticheskogo analiza teksturnykh izobrazheniy // Computer Optics, vyp. 20, pp. 122-127, 2000
http://www.problem-info.sscc.ru/2011-5/13.pdf
/12/ Kuznetsov A.V. Klassifikatsiya izobrazheniy s ispol'zovaniyem yader na strukturirovannykh dannykh //Sb. nauchnykh trudov NGTU.– 2011. –№3(65) pp.55-60.
/13/ Potapov A. A., Pakhomov A. A., Nikitin S. A., Gulyayev YU. V., Noveyshiye metody obrabotki izobrazheniy. – M.: Fizmatlit, 2008.– 496 p.
/14/ Khaykin S. neural networks. – M.: «Vilyams», 2006. – p. 1104
/15/ Teuvo Kohonen, "Self-Organizing Maps", Springer-Verlag, Heidelberg, 1995.
/16/ Korn G., Korn T. Handbook of Mathematics for Scientists and Engineers. M.: Nauka, 1984. - 832p.
