The　Motor　Imagery　electroencephalogram　(MI-EGG)　is　time　varying　and　subject-specific,　its　recognition　needs　the　perfect　adaptability　and　combination　of　feature　extraction　method　and　classifier.　In　this　paper,　Deep　Belief　Networks　(DBN)　is　integrated　with　Wavelet　Packet　Transform　(WPT)　to　yield　a　novel　recognition　method,　denoted　as　WPT-DBN.　Firstly,　the　MI-EEG　is　transformed　into　power　signal　and　analyze　the　effective　time　domain.　Then,　WPT　is　applied　to　each　channel　of　MI-EEG　to　obtain　the　effective　time-frequency　information.　Finally,　DBN　is　used　for　the　identification　and　classification　simultaneously.　Experiments　are　conducted　on　a　publicly　available　dataset,　and　the　5-fold　cross　validation　experimental　results　show　that　WPT-DBN　yields　relatively　higher　classification　accuracies　compared　to　the　existing　approaches.