We have shown in this paper that the combination of Soft
Computing techniques allows the improvement of intelligent
systems with different hybrid approaches. In this paper we
considered two parts of a modular neural network for image
recognition, where a Type-2 Fuzzy Inference System(FIS-2) helps us
improves the performance results in image recognition. The first FIS-
2was used for feature extraction in training data, and the second one
to find the ideal parameters for the integration method of the
modularneuralnetwork.Once again Fuzzy Logic is showntobe a tool
that canhelpimprovethe resultsof aneural system,whenfacilitating
the representation of the human perception. In this case, the hybrid
fuzzy neural approach can be considered a good alternative for
improving the performance of the modular neural model.