A Study of Texture Analysis Algorithms.
Abstract
This research has focused upon evaluating and developing texture analysis algorithms. During the current year theory has been developed for evaluating texture analysis algorithms. It has been shown that the Spatial Gray Level Dependence algorithm is a superior algorithm under fairly weak assumptions. The assumption required is that the textures be translation stationary random fields of order two. Under these assumptions, the SGLDM algorithm is superior to the Power Spectral Method, the Gray Level Run Length Method and the Gray Level Difference Method. The SGLDM algorithm has therefore been the focus of our subsequent work. Efforts have progressed on developing a combined structural and statistical texture analysis algorithm. Methods have been developed for characterizing the unit cell which gives the basic repeated texture pattern. Combined with a study of the unit cell has been the development of a mosaic model for textures that utilizes tiling theory. Tiling theory provides the basic mathematical structure required for analysis. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 23, 1980
- Accession Number
- ADA086656
Entities
People
- Charles A. Harlow
- Richard W. Conners
Organizations
- Louisiana State University