1. Science
  2. Publications
  3. Information Processing Systems
  4. 4(111)'2013
  5. Development of statistical methods for excluding the outliers which provide adequate application of image noise spatial filters

Development of statistical methods for excluding the outliers which provide adequate application of image noise spatial filters

K.S. Smelyakov, O.V. Vodolazhko, S.V. Smelyakov, S.V. Osievskey
Annotations languages:


Description: A series of statistical methods are proposed for excluding singular and multiple outliers which present brightness being atypical for a pixel vicinity processed by a filter. Making use of these methods allows us not only to solve the problem of adequate smoothing of object image interior, but also to solve an important problem of adequate smoothing the bounds of objects presenting both lines and nonsingular domains.


Keywords: image, noise, filter, mask, vicinity, outliers

Reference:
Smeliakov, K.S., Vodolazhko, O.V., Smeliakov, S.V. and Osievskii, S.V. (2013), “Postroenie statisticheskikh metodov iskliucheniia ekstremalnykh nabliudenii dlia obespecheniia adekvatnogo primeneniia prostranstvennykh filtrov shuma izobrazhenii”, Information Processing Systems, Vol. 4(111), pp. 35-41.