Image　fusion　is　a　very　important　step　for　image　mosaic.　Fusion　algorithms　affect　the　visual　effect　and　quality　of　the　mosaic　images　directly.　This　paper　presents　an　adaptive　weighted　coefficients　algorithm　for　image　fusion.　The　algorithm　can　adjust　weighted　coefficients　adaptively　along　with　changes　of　the　range　and　shape　of　the　overlapping　regions.　First,　Harris　operator　is　used　to　detect　feature　points.　Then,　we　use　normalized　cross　correlation　(NCC)　and　random　sample　consensus　(RANSAC)　algorithm　to　carry　out　coarse　matching　and　precise　matching　respectively,　and　to　obtain　the　transformation　matrix.　Finally,　the　proposed　algorithm　is　used　to　fuse　images.　Compared　with　other　fusion　algorithms,　the　experiment　results　indicate　that　the　proposed　fusion　algorithm　eliminates　stitching　seams　more　effectively,　and　has　better　fusion　effect.