METHOD FOR THE SEGMENTATION ENCODING OF AN IMAGE

Jean-Pierre Morard (1) , Olivier Pietquin (2) , Stéphane Vialle (2, 3)

1 SAGEM Défense Sécurité [Massy]
2 SUPELEC-Campus Metz
3 ALGORILLE - Algorithms for the Grid

The invention relates to a method for encoding an image, said encoding being a mixed encoding with the possibility of using a first lossless (100) compression type and a second lossy (101) compression type, wherein said method comprises the operation that consists of dividing the image into a plurality of elementary blocks (201), characterised in that said method comprises the different additional steps of: determining which of the elementary blocks (302; 303) have a high level of detail; allocating the first compression type to each of the elementary blocks (302; 303) having a high level of detail; and allocating the second compression type to each of the elementary blocks (302; 303) not having a high level of detail.

Jean-Pierre Morard, Olivier Pietquin, Stéphane Vialle. METHOD FOR THE SEGMENTATION ENCODING OF AN IMAGE. N° de brevet: WO/2010/072983. 2010. ⟨hal-00554405⟩