Abstarct: Brain signals or electroencephalography show the electrical activity of the brain; Brain signals are received, amplified (amplified and removed from noise) and recorded by electrodes attached to the scalp. Recorded signals are analyzed for clinical diagnoses and human-computer interaction identification systems. Emotion recognition has become a very controversial issue in the brain-computer interface. There are also important definitions and theories about human emotions. In this essay, we try to cover important issues related to the field of emotion recognition. Electroencephalogram (EEG) signal analysis is a biomarker in emotional changes. Due to its low cost, good timing, and spatial resolution, EEG has become very common and is widely used in BCI applications and studies. In this dissertation, we seek to improve the existing methods for recognizing human emotion from EEG signals. In this dissertation, he seeks to achieve the best interpretation for human emotional states, so that we can adapt to their behaviors and receive a proper response, and thus communicate with the computer so that the best state of learning and knowing human for different applications in the computer. To be used. This dissertation uses data from the laboratory databaxse of the University of Shanghai, which showed 4 types of emotions through video to 15 people during 3 periods. In fact, 24 films with 4 emotions showed these people that among These 24 films evoke 6 feelings of fear, 6 feelings of happiness, 6 feelings of neutrality and the other 6 feelings of sadness. The first step is to extract the appropriate features, which first begins by specifying the type of extraction, which is a time-frequency method, then in the next step, the success rate of each of these methods is evaluated, the third step should be appropriate features Categorize to be the best and most appropriate method of classification according to the previous results.