One of the major concerns in the field of computer vision and pattern recognition is emotion detection. One difficulty in face recognition is how to handle the variations in the expression, pose and illumination when only a limited number of training samples are available. In this paper KNN Regression algorithm with SURF feature is proposed for facial expression detection. Initially the eigenspace was created with eigenvalues and eigenvectors. . From this space, the eigenfaces are constructed, and the most relevant eigenfaces have been selected using Principal Component Analysis (PCA). The proposed method was carried out by taking the picture database. The database was obtained with 50 photographs of a person at different expressions. Another database was also prepared for testing phase by taking 10 photographs of that person in different expressions but in similar conditions ( such as lighting, background, distance from camera etc.) and these database images were stored in test folder.