Human motor imagery tasks evoke electroencephalogram (EEG) signal changes. A brain computer interface is a system that can translate the electrical activity of brain for using in communication and control. To distinguish signals of interest from the background activity various feature extraction methods have been applied, we describe a new technique for the classification of motor imagery EEG recordings. The technique is based on a time-frequency analysis of EEG signals, regarding the relations between the EEG data obtained from the C3/C4 electrodes; the features were reduced according the Fisher distance. This reduced feature set is finally fed to a linear discriminant for classification. The algorithm was applied to 3 subjects, and analyzed the different frequency band, Kappa number and time period of EEG. The classification performance of the proposed algorithm varied between 65% and 93.1% across subjects.