Optical　navigation　technology　is　widely　used　in　precision　guidance　for　autonomous　landing　and　hazard　avoidance.　According　to　the　requirements　of　the　computer　vision　landing　location　system　on　a　planetary　lander,　this　paper　proposed　an　algorithm　to　automatically　obtain　the　precise　landing　position,　which　is　based　on　optical　image　features.　The　main　contents　of　this　paper　are　as　follows:　Firstly,　an　integrated　corner　detection　algorithm　is　firstly　applied　to　detect　the　interest　corners　on　asteroid　images,　which　performs　quite　well　compared　with　other　feature　extraction　methods　in　many　aspects;　Secondly,　the　paper　introduces　a　mutual　matching　method　based　on　moment　invariant　and　regional　feature　mutual　information;　Finally,　the　landing　position　can　be　established　by　calculating　the　vectors　between　the　goal　landing　point　and　the　feature　points　obtained　on　the　base　image　and　correctly　matching　the　feature　point　pairs　on　the　base　image　with　images　captured　by　CCD　camera　in　real　time　when　landing　on　the　asteroid.　The　simulated　experiments　have　been　conducted　using　Eros　Asteroid　images.　The　results　show　that　the　algorithm　can　achieve　high　matching　accuracy　and　is　qualified　for　engineering　application.