The objective in the Scenic Beauty Estimation (SBE) problem is to develop an automatic classification algorithm that matches human subjective ratings. In order to get a better result, we use a new statistical modeling technique, dividing each images into 3 regions, then extracting features from each of them separately and use the Principal Components Analysis (PCA) to do training and testing. Compared with the previous experiment in which processing are carried on the whole image, the new classification method get better results on some aspects.