An image classification algorithm integrating principal component analysis and spatial pyramid matching features

To speed the image classification by effectively eliminating of redundancy in image features is a hot research topic in computer vision. Integrating the Spatial Pyramid Matching (SPM) and Principal Component Analysis (PCA), we propose a new SPM-PCA algorithm for effective image classification. Our method can improve time efficiency and meanwhile improve classification performance through dimensionality reduction in multiple levels of spatial pyramid. Experiments demonstrate our method's performance.