With the continuous development of the modern intelligent household, people found that brain wave can be used for controlling household appliances. EEG signal, as a typical brain wave signal carrying the brain state information, has walked into the researcher's view. Brain wave carries detailed information about the state of the brain. As a result, using computer to collect and analyze EEG signal plays a great role in smart home. This paper uses positive and negative emotional EEG signal as the research object, begins with introducing the research status of brain waves, and then uses the Chinese Affective Picture System (CAPS [12]) of Chinese Academy of Sciences, designs the watch pictures to experiment, utilizes machine learning algorithm of support vector machine (SVM) for data analysis, and obtains an accuracy of 58.3% eventually. This paper provides a feasible scheme for the study of EEG in the field of emotion analysis. Introduction In 1884, the father of American psychology James put forward that emotion was a kind of feeling of people for their body changes, the change of the body appeared before emotion perception and any emotion produced must be accompanied by some changes of the body[1]. This is the earliest definition of emotion. In 1927, Connon thought the mood was determined by the hypothalamus[2]. In 1937, Papez linked emotions to physical activity[3]. At present, a large number of results of neurological and cognitive science research show that the generation of emotion is related to physiological activities, particularly the activity of the brain which provides a theoretical basis for analyzing the activities of the cerebral cortex for emotion recognition. Emotional classification, like emotion, has not yet formed a unified theory. Modern scholars through research and experiment put forward their own emotional set, such as James's emotional set, Ekman's emotional set, etc. Through further research, the researchers found that there is a certain correlation between emotions, such as anger and disgust sometimes both appear, so there is a two-dimensional Lange emotional classification model[4]. The model is also the most common classification model using the longitudinal expression of mood pleasure, dislike to gradually over like; use abscissa to indicate the state of excitement, from depression to excitement. Common emotion recognition methods have two categories.The first one is based on the recognition of non-physiological signals, including the recognition of facial expressions and the recognition of speech tones.The advantage is that data acquisition is relatively simplewithoutthe need of special equipment. The disadvantage is that the reliability of emotion recognition is not ensured, because the data can be subject to subjective factors. The second one is based on physiological signal recognition.At present, the main physiological signals have EOG, GSP,BVP,EMG,EEG, ECG,etc. This paper uses the brain waves in physiological signals for emotion recognition, through the acquisition, identification, analysis of EEG signals.The main steps of emotion recognition based on EEG include: emotional evoked, EEG signal collection, EEG preprocessing, feature extraction, feature dimension reduction, emotion pattern learning and classification.First, we design the relevant experiments for emotional evoked. Second, we normalize the data extracted from EEG signals. Finally we use support vector machine model for data training and testing, to get the classification results. 2nd International Conference on Automation, Mechanical Control and Computational Engineering (AMCCE 2017) Copyright © 2017, the Authors. Published by Atlantis Press. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). Advances in Engineering Research, volume 118
[1]
P. Cochat,et al.
Et al
,
2008,
Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.
[2]
P. Lang.
International Affective Picture System (IAPS) : Technical Manual and Affective Ratings
,
1995
.
[3]
Branden Kahn.
Electroencephalogram (EEG) signal processing, wave identification, and emotion recognition
,
2015
.
[4]
Tanja Schultz,et al.
Towards emotion recognition from electroencephalographic signals
,
2009,
2009 3rd International Conference on Affective Computing and Intelligent Interaction and Workshops.
[5]
K. M. Dallenbach,et al.
The American Journal of Psychology: 1887-1937
,
1937
.
[6]
ScienceDirect.
Electroencephalography and clinical neurophysiology. Electromyography and motor control
,
1998
.