A Novel Automatic Image Stitching Algorithm for Ceramic Microscopic Images

The polyphase ceramic images taken by electron microscope contain rich material science information. However, in practical scientific engineering, the researchers can only obtain a small part material image by microscope. Therefore, it is necessary to apply image stitching technology to obtain full-size, information-rich images. To solve this problem, this paper proposes a novel automatic image stitching algorithm for ceramic microscopic images. Firstly, an algorithm for sorting out-ordered ceramic images into ordered sequences was proposed, which is based on local matching relationship and specific threshold. Next, an image stitching algorithm based on Principal Component Analysis (PCA) and Speeded Up Robust Feature (SURF) was proposed to get the complete image. The experimental result shows that the proposed method is more feasible and effective than traditional Scale Invariant Feature Transform (SIFT).