Fingerprint　matching　is　the　most　important　step　in　the　Automatic　Fingerprint　Identification　System　(AFIS).　In　this　paper,　a　new　method　of　fingerprint　matching　using　minutia　disturbance　is　proposed,　in　which　the　matching　process　is　divided　into　two　parts.　First,　at　the　alignment　stage,　the　local　structure　relationships　between　different　minutiae　have　been　used　to　adjust　the　coordinate　of　minutia,　which　include　the　triangle　structure　and　the　local　structure　of　the　point　and　line.　It　shows　a　good　result　especially　when　there　is　not　too　much　minutia　in　one　finger.　Then,　at　the　fingerprint　verdict　stage,　an　innovative　method　has　been　used　to　match　two　fingerprint　images.　The　main　idea　of　which　is　adding　random　minutia　into　a　fingerprint　to　disturb　the　set　of　minutia　after　alignment　and　get　the　statistic　matching　result　from　a　number　of　experiments.　The　method　avoids　the　cumbersome　job　in　the　past　like　setting　the　threshold　and　measuring　similarity　according　to　the　score.　Experimental　results　show　that　the　proposed　algorithm　improves　the　performance　of　the　fingerprint　matching　process.