The Research Of Music Image Recognition Based On Hausdorff Distance

From 1980s to the 90's, with the development of computer technology and each applied science, the image processing and analysis have gradually formed its own independent discipline system. As a result, the digital image processing, a new interdisciplinary research, was born.This paper summarizes the history, the present situation, the topic research background and the significance of digital image development. It introduces the concepts of it, elaborates the digital image processing technology, the main research and lists the essential features. Then it thoroughly contrasted the continual method and the separate method of digital image processing, and summarized the merits.The paper summarizes various parts of realization in deference to music recognition structure drawing. The specification was on the image purifies, the deletion spectral line, the packing slit, the boundary track algorithm and its realization specifically, explained by the effect chart. All of the efforts were used to complete the pretreatments before the music image match.After introduces the concepts of the template match, the article dilates the Hausdorff distance and the Euclidean distance. First carries on Distant Transform of the music image to obtain the distance image. Then acts according to the recognition algorithm flow, builds the template database, matches one by one, shows table matrix to gives the match results. Finally, analyzes the summary of the entire match experiment.