Abstract: A method for the non-destructive inspection of an aeronautical component having a step of obtaining a plurality of digital images of a unit area of the aeronautical component, a step of estimating a characteristic image representative of the unit area, each pixel of the characteristic image having a characteristic vector, a step of dividing the characteristic image into a plurality of micro prediction zones, a step of comparing the characteristic vector of each pixel in each micro prediction zone with a previously estimated local statistical model of the micro prediction zone, the local model of a micro prediction zone being obtained by means of a learning algorithm from characteristic vectors of pixels in a micro learning zone of the annotated characteristic image which includes the micro prediction zone.