The invention relates to a coding method applied to digital video data available in the form of a video stream consisting of consecutive frames divided into macroblocks. These frames are coded in the form of at least I-frames, coded independently, P-frames, predicted from at least a previous I- or P-frame, and B-frames, bidirectionally predicted from at least two frames between which they are disposed. According to the invention, the coding method comprises the following steps: a structuring step, provided for capturing for all the macroblocks of the current frame related coding parameters characterizing the fact that they have been coded, or not, according to a predetermined intra prediction mode; a computing step, for delivering statistics related to said parameters; an analyzing step, provided for analyzing said statistics for determining the number of blocks which exhibit, or not, said intra prediction mode; a detecting step, provided for detecting, each time said number is greater than a given threshold, the occurrence of an image, or of a sub-region of an image, which is either monochrome or with a repetitive pattern; a description step, provided for generating description data of said occurrences of images or sub-images either monochrome or with a repetitive pattern; a coding step, for coding both description data and original data.