A new metric algorithm of image textural coarseness

Aim at the address of fractal dimension for the measure of texture coarseness is low on calculation accuracy and the lack of robustness on rotating and noise metric. A kind of combining multi-scale and multi-directional metric model of Gabor filter with fractal dimension is proposed. First, Gabor wavelet model based on multi-scale and multi-directional transform is constructed and image filtering is executed on three-scale and four-directional metric model of Gabor filter. Then to establish fractal overlay model, 12 fractal dimensions are extracted based on filtered 12 images, and finally the average of fractal dimension is calculated as presented the model of measuring global texture coarseness. In the database Mosaics of various natural textures from the Brodatz album on the simulation experiments show that the new algorithm to measure the local texture metric comparison of the original global coarseness of the fractal model has high measurement accuracy and a strong rotation and noise robustness.