A method for a signal evaluation of an electronic image sensor in the case of the pattern recognition of image contents of a test specimen

摘要

A method for a signal evaluation of an electronic image sensor in the case of the pattern recognition of image contents of a test specimen, wherein said image sensor a light input signal and receives an electrical output signal which is correlated to the light input signal, characterised in that the light input signal is processed as follows,– Dividing the image of the test object in n × n arranged in a grid window (01) of the size n × n pixels (02),– Analysis of the image contents (03) of a window (01) of the size n × n pixels (02) by,– Conversion of the directly or indirectly by the image sensor output signal output into at least one trans modulation invariant feature value (08) by means of at least one computing rule (04, 07),– the computing algorithm (04) for the conversion of the signal of the image sensor into a modulation invariant feature value (08) is a two-dimensional mathematical spectrally transformation method (04), in particular a two-dimensional fourier -, or walsh -, or hadamard -, or circularly transformation,– Weighting of the feature value (08) with at least a fuzzy membership function (13), wherein the membership function (13) into functional relationship..