The classification of precise data has been studied and analyzed in various ways and in a wide range, but the data used for classification do not always have a specific and precise value. Since the type of data scale is different, the data value may fall within an interval, in which case the problem of classifying imprecise data arises. In recent years, assuming that the distribution governing imprecise data is normal, various estimates have been presented for the mean and variance of this distribution. In this article, assuming that the distribution governing imprecise data is a bivariate normal distribution, we have estimated the mean and variance of this distribution using the maximum likelihood method on the values ​​at both ends of the imprecise data interval. Then, using simple Bayesian classification, we have presented a Bayesian mixture model for classifying precise and imprecise data. The accuracy and efficiency of the presented model have also been investigated.