Method for detecting an anomaly on the surface of a tire comprising the following steps in the course of which: A-the image of a given anomaly present on the surface of at least one tire is produced, B-with the aid of a collection of filters, a multivariate image of the said surface is constructed in a space of the filters, in which each pixel is represented in the form of a pixel vector, the components of each pixel vector having a value corresponding to the value of this pixel in the image transformed with the aid of each of the filters of the said collection, C-with the aid of a linear function, this multivariate image is transformed from the space of the filters into a spectral space of given dimension whose variables are the filters or combinations of filters of the said collection, so as to form a spectral image, D-a classifier is constructed by determining, for this anomaly,; those zones representative of the spectral space which contain, in a statistically representative manner, the points of the spectral image of the said anomaly transformed into the said spectral space.