Aerial　image　sequence　mosaicking　is　one　of　the　chal-lenging　research　fields　in　computer　vision.　To　obtain　large-scale　orthophoto　maps　with　object　detection　information,　we　propose　a　vision-based　image　mosaicking　algorithm　without　any　extra　location　data.　According　to　object　detection　results,　we　define　a　complexity　factor　to　describe　the　importance　of　each　input　ima-ge　and　dynamically　optimize　the　feature　extraction　process.　The　feature　points　extraction　and　matching　processes　are　mainly　guided　by　the　speeded-up　robust　features　(SURF)　and　the　grid　motion　statistic　(GMS)　algorithm　respectively.　A　robust　refer-ence　frame　selection　method　is　proposed　to　eliminate　the　trans-formation　distortion　by　searching　for　the　center　area　based　on　overlaps.　Besides,　the　sparse　Levenberg-Marquardt　(LM)　al-gorithm　and　the　heavy　occluded　frames　removal　method　are　ap-plied　to　reduce　accumulated　errors　and　further　improve　the　mo-saicking　performance.　The　proposed　algorithm　is　performed　by　using　multithreading　and　graphics　processing　unit　(GPU)　accel-eration　on　several　aerial　image　datasets.　Extensive　experiment　results　demonstrate　that　our　algorithm　outperforms　most　of　the　existing　aerial　image　mosaicking　methods　in　visual　quality　while　guaranteeing　a　high　calculation　speed.