Robust calculation of fractal dimension of images and its applications to classification of ultrasonic liver images and texture images

The estimation of the fractal dimension is crucial in fractal geometry. The popular estimation approach is based on box-counting. However, that scheme, which is easily disturbed by noise, produces many non-negligible plateaus that cause underestimation. This paper proposes a more robust and efficient computation of the fractal dimension. Finally, the approach is applied to classify ultrasonic liver images and natural texture images. Experimental results reveal that the proposed computation method is trustworthy.