Method of texture segmentation of aerospace images
Автор: Karlushin K.A., Kurbatova E.E., Medvedeva E.V.
Журнал: Инфокоммуникационные технологии @ikt-psuti
Рубрика: Технологии компьютерных систем и сетей
Статья в выпуске: 4 т.10, 2012 года.
Бесплатный доступ
Method of texture segmentation of aerospace images based on two-dimensional Markov processes is offered. As a textural feature the probability of transition between image elements is used. The method of segmentation has been researched experimentally and compared with known contour methods (Canny, Sobel, Laplacian gausian). The method effectively allocates texture regions with different statistical characteristics and allows to reduce computing resources.
Texture segmentation, digital grayscale images, markov process, the probability of transition, an entropy
Короткий адрес: https://sciup.org/140191584
IDR: 140191584
