The automatic recognition of emotional states are quite researched problem. In this work, emotion recognition problem was investigated and the system has been developed that can recognize expressions of happy, sad, surprised, angry, afraid, disgusted and neutral from an image. Previously, a feature vector was extracted for the detected face and the classification was made using the distance information between the vector and vectors of emotional expression. SVM(Support Vector Machine) approach was used. Middle-aged faces were studied without glasses and clean-shaven. KDEF dataset was used.