Music features recognition and its application in national music protection

Music features recognition is an important research hotspot in computer music, which can be widely applied in many fields such as multimedia retrieval, human-machine interaction, digital heritage, digital entertainment and so on. Aimed at digital heritage protecting engineering of Chinese national music, this thesis discusses the topic of music features recognition technology by simulating human´s cognitive process based on MIDI files, mainly including such high-level features recognition technologies as melody extraction, music section recognition and partition, and emotion recognition. At last, the prototype system of virtual editor and exhibition for Chinese national music by means of aforementioned technologies are developed to serve for digital popularising of Chinese national music