CIOP   05384
CENTRO DE INVESTIGACIONES OPTICAS
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
The fractal dimension curve as a descriptor of biospeckle images
Autor/es:
. E. E. GRUMEL, N. CAP, A. LONGARZO, H. RABAL, M.TRIVI; A LONGARZO
Lugar:
Santiago de Compostela
Reunión:
Congreso; XXIII General Congress International Commission for Optics (ICO-23); 2014
Institución organizadora:
International Commission for Optics ICO
Resumen:
Dynamic speckle or ?biospeckle? [1] is a phenomenon occurring when an active rough sample isilluminated with a coherent light. Dynamic speckle patterns analysis from active materials gives usefulinformation of the illuminated samples and it has been used in some biologic and industrial applications[2]. Many algorithms for quantitative and qualitative characterization of dynamic speckle have beenreported [3], with some advantages and drawbacks. Therefore, we continue to investigate new algorithmsfor different dynamic speckle phenomena.Fractal dimension has been described for time evolution of dynamic speckles by Passoni et al [4]. Thespace version of fractal dimension [5] applied to images through its measurement called ?fractaldimension box? is used in this work to characterize the degree to which an image, conveniently binarized,covers the plane. The result, in this case a number between 0 and 2, depends on the chosen thresholdvalue for binarizing the image. By varying the threshold continuously, the obtained box fractal dimensiondescribes a cumulative curve which is characteristic of the distribution of gray in the image. For a speckleimage it is a sigmoid curve analogous, in some cases, to the statistical distribution of Fermi Dirac (FD).This function is used to adjust the curve and define an equivalent to a ?temperature? parameter descriptor.This temperature characterizes the slope of the curve and is zero if the original image is binary.In this paper we use the formalism of fractal dimension to characterize images of speckle diagrams invarious situations of interest