Comprehensive calculation of the characteristics of landscape images

Algorithms have been developed for computing the fractal dimensions of textural images by overlaying them with standards and by overlaying them with two-dimensional surfaces, which showed the best results. The main fractal characteristics (the fractal dimension and lacunarity) of landscape images, which are natural fractal formations, have been calculated. These were supplemented with statistical indices such as the mean brightness, variance, third moment, homogeneity, entropy, and certain other characteristics. A software complex has been developed for estimating the fractal dimension of textures when recognizing landscape images. Two databases of natural textures were used for the studies: Textures Library--Forest (400 images) and Brodatz (110 images).

[1]  M. E. Jernigan,et al.  Texture Analysis and Discrimination in Additive Noise , 1990, Comput. Vis. Graph. Image Process..

[2]  James M. Keller,et al.  Texture description and segmentation through fractal geometry , 1989, Comput. Vis. Graph. Image Process..