Advanced blocs classification for fast encoding in fractal-based gray-scale images compression

Fractal-based images compression entails a computationally costly search for matching range and domain blocs. One way to remedy at this problem is to classify image blocs into categories and only search among domain blocs which are in the same category as the target range bloc. Since image blocs with a simple edge (blocs with a distinct edge running through them) are a very important portions of the perceptual information content in image, we propose in this paper a method to both identify and classify this kind of blocs according to their edge presentation. We refer to this method as forced classification (FC). This method is combined with other suitable methods of blocs classification available in the literature to allow a fast and efficient encoding of grey-scale images. The result is surprisingly good, the encoding time for 512/spl times/512 lena image is reduced by a factor of 37.52% than using only Yuval Fisher (1995) classification, while the loss of image quality is low.