350 rub
Journal Biomedical Radioelectronics №11 for 2010 г.
Article in number:
Research of the Usage of Dynamical Segmentation Methods, Based on the Optical Flow, in Application to the Angiography Diagnostics Problems
Authors:
A. P. Nemirko, S. A. Ivanovskiy, E. L. Maryaskin
Abstract:
Number of images processing problems add up to the detection of moving objects, their boundaries, etc. Methods of dealing with these problems are based on the dynamic segmentation of images. A lot of problems in medical diagnostics researches are also connected with the motion analyses in video sequences. One of the areas, in which dynamical segmentation of images may be actively used, is angiography diagnostics. Motion analysis has long been a specialized subject area. Motion and brightness alteration are not equivalent. Here two terms appear: motion field and optical flow. On basis of these equations several algorithms for optical flow calculation have been developed and implemented. Even the elementary optical flow calculation makes it possible to carry out a segmentation. Every video sequence procession consists of two stages: data pre-procession and data post-procession. Modifica-tions, introduced to the both of the procession stages, are described here. Traditional method has a significant disadvantage. Special software tool for fast flow calculation and its structure display was developed. Filtration algorithms are used to minimize aberration influence. Flow vectors field analysis makes it possible to effectively resolve some of the angiography diagnostics problems. The problems of sanguimotion process reconstruction, of the complete scheme of sanguimotion in the observed area reconstruction, of the calculation of blood flow speed in the observed area, of vessel sectional plane assess-ment are considered. Described processing method can be proposed after improvement for usage in the different areas of the biomedi-cal researches and can help the specialists in their work.
Pages: 10-15
References
  1. Яне Б. Цифровая обработка изображений. М.: Техносфера. 2007. 583c.
  2. Шапиро Л., Стокман Ж. Компьютерное зрение. М.: БИНОМ. ЛабораторияЗнаний. 2006. 752с.
  3. Ballard D.H. Computer vision. Rochester. New York: Brown Department of computer science. University of Rochester. 2006. 573p.
  4. Baker S., Matthews I. Lucas-Kanade 20 Years On: A Unifying Framework. The Robotics Institute Carnegie Mellon University Pittsburgh. 2002. 30 p.
  5. Bruhn A. Variationelle Optische Flussberechnung. Prazise Modellierung und Effiziente Numerik. Saarbrucken: Universitat des Saarlandes. 2006. 194 p.