The invention discloses a feedback system based on a motor imagery brain-computer interface. The feedback system comprises a feedback module, an electroencephalogram collector, a wireless transmission module and a terminal device; the feedback module is used for compiling a feedback interface; the electroencephalogram collector is used for collecting an electroencephalogram signal associated with motion; the electroencephalogram signal is amplified and filtered by the electroencephalogram collector after scalp electrode detection and is transmitted to the terminal device by the wireless transmission module; and the terminal device is used for carrying out data processing on the electroencephalogram signal to extract a motor imagery characteristic signal, and the motor imagery characteristic signal is used for forming visual feedback after mode identification and controlling the feedback interface to form a closed loop control system. Compared with the traditional MI-BCI system, the feedback system disclosed by the invention better conforms to a normal thinking action control process and approaches to actual interactive application, thereby being expected to provide critical technological guarantee for novel MI-BCI. The feedback system disclosed by the invention can be applied to the fields of disabled rehabilitation, electronic entertainment, industrial control, aerospace engineering, etc.