Image classification is most emerging area in today’s world. Variety of images classified using different methods. In this paper image classification based on two different approaches Artificial neural network and Neurofuzzy system and it is seen that Neurofuzzy system is better classification technique than ANN. The design used the discrete cosine transform (DCT) for feature extraction and artificial neural networks and neurofuzzy system for Classification. As DCT works on gray level image, the color image is transformed into gray levels. A neuro-fuzzy approach was used to take advantage of neural network’s ability to learn, and membership degrees and functions of fuzzy logic. This paper proves that neuro-fuzzy model performed better than the neural network in classification of texture image of 2 different types.