Scenery video type classification based on SVM

In this paper SVM algorithm is applied to classify the scenery video types in compressed domain. Firstly we extract video sequences randomly from scenery video and detect representative frames from the video sequences; secondly we extract features such as color layout, dominant color, edge histogram and face feature; then according to SVM, representative frames are classified as natural scenery, personality, animal and plant. Experimental results have shown that the result of our algorithm is very high accuracy.