Artifacts and selection of relevant patterns in dynamic speckle images by means of principal components

J. M. LÓPEZ ALONSO, J. ALDA; E. E. GRUMEL, N. CAP, H. RABAL, M.TRIVI

Santiago de Compostela

Congreso; XXIII General Congress International Commission for Optics (ICO-23); 2014

International Commission for Optics ICO

Dynamic speckle patterns convey information on the dynamics of the underlying physical phenomenaoccurring in the sample. Several phenomena usually concur to the final observed intensity variations,including artifacts coming from the measurement equipment (camera, non-uniformity illuminations).Several measurements have been proposed in the literature to characterize these dynamic patterns [1-3].In this work we propose the use of a multivariate statistical method (Principal Component Analysis-PCA), which decomposes the original image speckle frames into a new set of frames (PrincipalComponents). The main relevant property of these new frames is that they exhibit a null correlationbetween them. On the contrary, the original image speckle exhibits a correlation among them due to theunderlying dynamics.