METHOD FOR DETECTING CHAOTIC STRUCTURES IN A SPECIFIC ENVIRONMENT

摘要

The invention concerns a method for detecting chaotic structures in a specific environment which consists in calculating all the components of the luminous intensity gradient vector E in all points of a window F, centred on a point of a block representing the milieu. Said method is characterised in that it further consists in: summing all the elementary matrices M for all the points of the window F; diagonalizing said matrix sum A to determine its characteristics values lambda 1, lambda 2, lambda 3; quantifying in the centre of the window, the minimal aggregate error of the optical flux vector U on the basis of said characteristic values lambda 1, lambda 2, lambda 3 and with the constraint U<T> x D = 1; eliminating the contribution of the greatest characteristic value; defining a multidirectional error by integrating it in the plane defined by the characteristic vectors corresponding to the remaining characteristic values; assigning the multidirectional error to the image point whereon is centred the window F, and computing the multidirectional errors assigned to all the image points of the image block.