Extraction of small-size objects by segmentation algorithms using the model of the system with abruptly changing random structure
A.N. Malov, B.M. Mironov, V.A. Kuznetsov

Irkutsk Higher Air Force Engineering School (Military Institute)

Full text of article: Russian language.

Abstract:
We now present substrate-surface segmentation algorithms based on a model of the system with abruptly changing random structure and smoothing sufficient statistic. We use a simulation method in performing research of the efficiency of segmentation algorithms in extraction of small-size objects. These characteristics allow us to compare the efficiency of different algorithms and to determine size limits for segmented sections of the objects.

Key words:
substrate-surface segmentation, simulation method, small-size objects.

Citation: Malov AN, Mironov BM, Kuznetsov VA.  Extraction of small-size objects by segmentation algorithms using a model of the system with abruptly changing random structure [In Russian]. Computer Optics 2008; 32(1): 89-92.

References:

  1. Verdenskaya NV. Image segmentation: statistic models and methods [In Russian]. Telecommunications and Radio Engineering 2002; 12: 33-47.
  2. Skripnik ON, Lezhankin BV, Malov AN, Mironov BM, Galiev SF. Formation of spreading surface classification map on coherent radar images [In Russian]. Computer Optics 2006; 29: 151-159.
  3. Klekis EA. Optimal detection of jump structures of discrete dynamic systems by noise-free observations [In Russian]. Vilnius: Mathematics and Cybernetics Institute of the RAS of the Lithuanian Soviet Socialist Republic. Statistical Control Problems Publisher 1986; 73: 89-99.

© 2009, ИСОИ РАН
Россия, 443001, Самара, ул. Молодогвардейская, 151; электронная почта: ko@smr.ru ; тел: +7 (846) 332-56-22, факс: +7 (846 2) 332-56-20