-In this paper, we propose a method for classifying Fractal Models, not only describe texture but also synthesize it textures using Genetic Programming (GP). Texture features are using a specific model. extracted from the energy of subimages of the wavelet Genetic Programming (GP) [4], which is an evolutionary decomposition. The GP is then used to evolve rules, which are method that lends itself naturally to the development of arithmetic combinations of energy features, to identify whether a program with the ability of automatically constructing texture image belongs to certain class. Instead of using only one rule to discriminate the samples, a set of rules are used to vappropriate structures for the solution as well as selectig the perform the prediction by applying the majority voting variables, has been applied to extract features from textures in technique. In our experiment results based on Brodatz dataset, the past. A GP method iS proposed based on raw pixel data the proposed method has achieved 99.6% test accuracy on an using dynamic range selection [5][6]. Gray level histogram, average. In addition, the experiment results also show that which is reported to outperform raw pixel features in the classification rules generated by this approach are robust to some cluster accuracy and convergence speed, is also used as noises on textures.features to evolve the classification rules [7].