Principal component analysis for feature extraction of image sequence

This paper presents a method to extract features by PCA (Principal component analysis) from a series of woods surfaces´ images. The method introduces a principal component subspace and can reserve original information while extraction mainly information. The emulated results show that we can fuse the image series and extract features from the four images of a same surface by using this method. After processing the sequences of images, we get a feature which is good for the next classification and recognition process.