There are several techniques for analyzing multispectral images. In general, those are performed by linear transformation methods. In this paper, we present a new interactive method for classifying multispectral images using a Hilbert curve which is a one-to-one mapping from N- dimensional space to one dimensional space and preserves the neighborhood as much as possible. The merit of our system is that the user can extract clusters without computing any distance in N- dimensional space, and analyze multidimensional data from gross data distribution to fine data distribution hierarchically. In the experiments using LANDSAT TM data, it is confirmed that the user can get the real time response from the system after once making the data tables, and understand distribution of data that correspond to categories in feature space.