Brain is the most advanced part of the nervous system. As an emerging neighborhood exploring the brain, the brain-computer interface (BCI) may completely influence people's current communication and lifestyle in the future. We propose an architecture for feature extraction and classification model in brain-computer interface (BCI). Three features of EEG signal are extracted and combined in an attempt to describe the EEG signal more comprehensively including Wavelet packet decomposition, Information entropy and Co-space pattern (CSP). We construct the C-LSTM model combining the Convolutional Neural Networks (CNN) and the Long Short-Term Memory (LSTM) to perform the classification. Comparison was made with traditional approaches, and results show the method has a better performance for classification of motor imagery.