In　the　paper,　massive　3D　laser　point　cloud　data　is　obtained　through　vehicular　laser　scanning　system.　Massive　3D　laser　point　cloud　data　is　characterized　by　large　data　size,　uneven　distribution,　complex　data　attributes,　etc.,　and　these　characteristics　are　analyzed.　Compared　with　traditional　quadtree　algorithm　for　processing,　it　is　proposed　that　massive　3D　laser　point　cloud　data　can　be　organized　through　utilizing　improved　quadtree　algorithm.　Characteristics　of　Hilbert　curve　are　combined　for　further　improving　organization　efficiency　of　massive　3D　laser　point　cloud　data　on　the　basis　of　improved　algorithm.　In　the　paper,　related　experiment　is　implemented　on　corresponding　platform　with　java　language.　Experiment　proves　that　point　cloud　data　is　organized　through　improved　quadtree　algorithm.　Efficiency　is　correspondingly　improved　compared　with　traditional　quadtree　algorithm.　The　improved　algorithm　is　combined　with　Hibert　curve　synchronously　for　further　improving　organization　efficiency.