Objective To propose a method combining electroencephalogram (EEG) and electromyogram (EMG) for the control of home system to enhance the control safety.Methods A hybrid braincomputer interface control method based on steady-state visual evoked potential (SSVEP) and EMG of temporal muscle was proposed and a virtual home control system integrating five different devices was designed and implemented.SSVEP was used for target selection, and EMG of temporal muscle was used for system switch, graphical user interface switchover and target confirmation.EEG and EMG signals were processed by using canonical correlation analysis and threshold method respectively, and then the generation of system control commands was determined combining EEG and EMG recognition results and system state.Six subjects were involved in the system verification experiments.Indices including command ratio and false operation rate were defined to evaluate the performance of the system.Results All of the six subjects successfully completed the control for the five kinds of devices.No operation error occurred during the control.The average command ratio for device selection/control command was 106.3%.Conclusions This control method of combining EEG and EMG effectively reduces operation errors and enhances the control safety for the home system.