Abstract
A fast, reliable, and objective computerized system for classification of color textured perthite images is proposed. In order to locate a crystal's borders, color and texture features are combined in a pixel classification segmentation operation, followed by probabilistic relaxation. Once a binary image is extracted, noise is suppressed using mathematical morphology. An operator based on vertical-horizontal run length ratios is used in order to identify and separate vein elements. Joint patch elements are disconnected using an iterative patch core locating procedure. Based on geometrical properties, the texture element's distribution is computed. Experimental results are compared to manual classification.
Original language | English |
---|---|
Pages (from-to) | 1533-1545 |
Number of pages | 13 |
Journal | Pattern Recognition |
Volume | 30 |
Issue number | 9 |
DOIs | |
State | Published - 1 Jan 1997 |
Keywords
- Color texture analysis
- Computer vision
- Perthite crystals
- Relaxation
- Shape analysis
ASJC Scopus subject areas
- Software
- Signal Processing
- Computer Vision and Pattern Recognition
- Artificial Intelligence