Image Stitching Method of Bone Stick Fragment Based on Similarity Freeman Code Matching

Abstract

A large number of bone stick fragments were unearthed at the Weiyang Palace, and the manual stitching method is inefficient and time-consuming, therefore Image Stitching Method of Bone Stick Fragment Based on Similarity Freeman code Matching is proposed by this paper to improve bone stick stitching efficiency. The Freeman code is used to represent the contour of the true color RGB image of bone stick fragments, and the direction change dense feature of the bone stick fragment fracture is fused with the Freeman code difference to represent the change direction feature to locate the fracture boundary. At the same time, the integer basic operation is used to replace the complex calculus calculation to improve the efficiency of the algorithm; The longest common subsequence algorithm is used to perform local similarity rough matching on Freeman code of the contour at the fragment fracture, and then the angle and distance error threshold is set according to the prior knowledge of manual stitching to reduce the matching error; Finally, according to the matching results, the complete bone stick contour binary image is drawn. Compared with other algorithms, the contour Cosine similarity accuracy of the bone stick image stitched by this algorithm is improved by 21.68%, which can be accurate to 97.80%; the Editing distance similarity accuracy is improved by 14.29%, which can be accurate to 91.87%;the Peak signal-to-noise ratio is improved by 83%, which can be accurate to 31.13%;the Structural Similarity is improved by 29%, which can be accurate to 0.87; the Figure of merit is improved by 39%, which can be accurate to 0.85. The experiment results show that the algorithm proposed in this paper can effectively assist the stitching work of bone stick cultural relics.