A　registration　method　based　on　morphological　feature　fusion　is　proposed.　First,　ISS　and　3D　Harris　feature　points　were　projected　onto　the　XY　plane　respectively　to　generate　binary　images,　and　then　morphological　operations　were　performed　to　remove　the　interference　of　outliers　and　maintain　the　continuity　of　features.　Then,　conditional　filtering　was　performed　on　the　feature　points　according　to　the　binary　images　to　obtain　fusion　feature　points,　and　coarse　registration　was　completed　using　PPF.　In　the　fine　registration　stage,　IQR　statistical　analysis　is　carried　out　on　the　curvature　of　the　point　cloud,　the　threshold　is　set　to　retain　the　area　of　high　curvature,　and　finally　GICP　registration　is　carried　out.　The　experimental　results　show　that　the　proposed　method　is　significantly　superior　to　the　traditional　method　in　terms　of　registration　accuracy,　and　has　certain　practical　application　value.　©　2025　IEEE.