Research And Implementation Of Rule - Based Simulation Of Shaped Paper Mosaic Algorithm

As the shredder apply in work and life, scraps of paper recovery in criminal investigation and national security, and other fields are playing a more and more important role. This paper use many ways to solve the problem that to recovery scraps of paper, like the number recognition algorithm, the greedy genetic algorithm and k-means clustering algorithm and parallel genetic algorithm. In this paper, the problem of scraps of paper recovery includes: Chinese and English recognition,shredding grain of clustering, transverse splices, longitudinal splicing. Because of the need for splicing of shredding grains from the vertical and horizontal, so the first grain of clustering for shredding, then carried out within the class transverse splicing,longitudinal joining together outside of class.1. Practice Chinese and English recognition: because of the number of Chinese text is generally larger and English letters is small, so the vertical through The Times as the distinction between Chinese and English to the main feature of classification of language and completed the scraps of characters’ recognition in both Chinese and English.2. Grain of clustering, shredding: according to the grain of the text in the corresponding algorithm to extract shredding in both English and Chinese characteristics of high line and line spacing, and on the basis of the same text line features of the shredding grain has the similarity to clustering, on a scrap of paper between the Euclidean distance as the standard of text line features, using the K-means algorithm completed the scraps of clustering in both English and Chinese.3. Lateral stitching longitudinal chopped note transverse stitching to the edge of similarity as a standard of judging, the greedy genetic algorithm completed longitudinal chopped note transverse splicing; ripped up grain of transverse stitching to the edge of similarity and width matching degree as evaluation standard, by using double fitness GA completed the shredding transverse splices.4. In view of the longitudinal stitching: to extract the edge height matchingdegrees between the scraps of paper as a standard of judging, the greedy genetic algorithm solving the problem of joining together.In this paper, based on the VC++ language, to write about more than 9000 lines of code to implement these algorithms, the experimental results show the feasibility of the above algorithm. Through the experimental results and comprehensive experimental data, the paper has analyzed the algorithm and the text, the text line characteristics features use of inadequate.