Identification of artifacts in scenery images using color and line information by RBF network and K-means

In this paper, we propose an identification method of artifacts in scenery images using color and line information by Radial Basis Function (RBF) network and K-means. In the proposed method, the original image is divided into some areas based on color information by the K-means algorithm, and whether or not artifacts are included for each divided area are judged using color and line information by the RBF network. We carried out a series of computer experiments and confirmed that the proposed method can identify the area including artifacts with considerable accuracy.