空白区域匹配法处理图像碎片复原问题  

Using Blank Space Matching to Deal with Image Fragment Restoration

作　　者：潘娟娟[1] 凌雪岷[1] PAN Juan-juan;LING Xue-min(Department of Common Education,Anhui Xinhua University,Hefei,Anhui 230088,China)

机构地区：[1]安徽新华学院通识教育部,安徽合肥230088

出　　处：《河北北方学院学报（自然科学版）》2020年第7期7-11,共5页Journal of Hebei North University:Natural Science Edition

基　　金：安徽省高校自然科学重点研究项目(KJ2017A622);安徽省质量工程一般教研项目(2017jyxm0530)。

摘　　要：目的利用空白区域匹配法对图像碎片进行拼接,最终完成破损图像的匹配。方法利用MATLAB把所给的碎纸片转换为相应的像素矩阵,任取1个像素矩阵作为配源矩阵,求出各行的和,得出1个源配列向量。结果比较源配列向量和待配列向量中的元素为255所在的行数是否一致,完全吻合的归为一组,对于剩下的像素矩阵重复上述过程,最后得出若干组。对没有归类的任何一组的列向量,根据边界匹配度来判断其归属。结论2次边界匹配法得出最后完整的图片。Objective Blank space matching was used to splice image fragments,and the damaged image was matched.Methods Scrap paper was converted to the corresponding pixel matrix with MATLAB.Any pixel matrix was taken as the source matrix to find the sum of the rows.The source collocation column vector was obtained.Results The consistency on row number of element 255 in the source matching column vector and the column vector was determined by comparison,and those that fitted perfectly were grouped together.This process was repeated for the remaining pixel matrices.Thus,several groups were obtained.For any group of column vectors that were not classified,their affiliation was determined according to the boundary matching degree.Conclusion Performing the boundary matching method twice can result in the final full picture.

关 键 词：边界匹配法 源配列向量 像素矩阵 配源矩阵 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]