The method of image pre-filtering to improve the accuracy of recognition
Published: 17.11.2014
Authors: Koval Yu.A., Filippov M.V.
Published in issue: #12(36)/2014
DOI: 10.18698/2308-6033-2014-12-1307
Category: Information technology
Solution of the recognition problem of various objects in images is usually difficult because of the presence of various kinds of distortions. This paper describes a method of pretreatment of images using bilateral and shock filters for subsequent pattern recognition method of contour analysis. Assessment of the recognition quality, when using these filters, and without filters for different parameters is made.
References
[1] Gonzalez R.C., Woods R.E. Digital Image Processing. 2nd ed. Pearson Education, Inc., 2002.
[2] Krasilnikov N.N. Tsifrovaya obrabotka 2D i 3D izobrazheniy [ Digital processing of 2D- and 3D-images]. BHV-Petersburg, 2011, 608 p.
[3] Furman Ya.A. Vvedenie v konturnyi analiz [Introduction to circuit analysis].Moscow, FIZMATLIT, 2003, 598 p.
[4] Tomasi C., Manduchi R. Bilateral Filtering for Grey and Color Images. Proceedings of the 1998 IEEE International Conference on Computer Vision, Bombay, India, 1998, pp. 839-848.
[5] Weickert J. Coherence-Encahncing Shock Filters. Lecture Notes in Computer Science, vol. 2781, Springer, 2003, pp. 1-8.