In　the　paper,　an　efficient　spatial　clustering　algorithm,　the　improved　DBSCAN　(Density　Based　Spatial　Clustering　of　Applications　with　Noise),　is　proposed,　for　cluster　analysis　of　trajectory　points　gathered　from　Digital　Movies　Mobile　Playing　Systems(DMSs).　By　searching　the　density　of　points　which　are　greater　than　a　given　threshold　in　rural　areas,　the　densityreachable　maximum　movie　playing　clusters　are　established.　Moreover,　the　improved　DBSCAN　algorithm　is　based　on　the　adjacent　list　structure　to　save　space　and　time　resources.　The　paper　also　adopts　the　real　mobile　projection　GPS　data　in　a　province　to　analyze　the　clusters　of　movie　trajectory　points.　Experiment　results　show　that　the　presented　algorithm　can　not　only　reflect　quickly　the　overall　distribution　of　the　projection　location　information,　but　dig　out　the　projection　regions　with　high　density.　And　the　conclusions　drawn　from　the　paper　are　valuable　for　the　management　department　to　evaluate　the　tasks　of　the　projectionists.