基于聚类的规则文档碎纸片拼接模型

针对碎纸机破碎文档后的规则碎纸片拼接问题，通过对碎纸片上边缘的灰度向量将文档分为上边缘为空白和非空白区域两大类，再分别以上边缘非空白区高度和空白区高度作为聚类参数，将纸片分为若干簇，在每一个簇中利用相邻两张碎纸片左右边缘向量相似度来进行拼接，得到若干横条的纸片，然后以行距和横条间上下边缘相似度为参数来将若干横条拼接为完整文档。

The reconstruction of paper fragments of rule document based on clustering

In this paper, a method that the paper fragments of rule document is reconstructed is provided .The paper fragments is divided into the upper edge of the blank and non -blank area into two categories by the gray vector on the edges .Respec-tively, the height of the upper edge of the blank and non -blank area as the clustering parameters is calculated .The paper fragments will be divided into several clusters .The reconstruction of paper fragments depends on computing the similarity of the left and right edges of adjacent pieces of paper in each cluster .After getting the number of bars of the paper , the paper fragments of rule document is reconstructed by computing the similarity between the top and bottom edges of the bar .