The　influence　of　the　built　environment　on　dockless　bike　-sharing　(DBS)　trips　connecting　to　urban　metro　stations　has　always　been　a　significant　problem　for　planners.　However,　the　evidence　for　correlations　between　microscale　built-environment　factors　and　DBS-metro　transfer　trips　remains　inconclusive.　To　address　this,　a　framework,　augmented　by　big　data,　is　formulated　to　analyze　the　correlation　of　built　environment　with　DBS-metro　transfer　trips　from　the　macroscopic　and　microscopic　views,　considering　Beijing　as　a　case　study.　The　trip　density　and　cycling　speed　are　calculated　based　on　11,120,676　pieces　of　DBS　data　and　then　used　to　represent　the　characteristic　of　DBS-metro　transfer　trips　in　a　multiple　linear　regression　model.　Furthermore,　a　novel　method　is　proposed　to　determine　the　built-environment　sampling　area　around　a　station　by　its　corresponding　DBS　travel　distances.　Accordingly,　6　microscale　built-environment　factors　are　extracted　from　street-view　images　using　deep　learning　and　integrated　into　the　analysis　model,　together　with　14　macroscale　built-environment　factors　and　8　potential　influencing　factors　of　socioeconomic　attributes　and　metro　station　attributes.　The　results　reveal　the　significant　positive　influence　of　greenery　and　presence　of　barriers　on　trip　density　and　cycling　speed.　Additionally,　presence　of　streetlights　is　found　to　be　negatively　correlated　with　both　trip　density　and　cycling　speed.　Presence　of　signals　is　also　found　to　have　an　influence　on　DBS-metro　transfer　trips,　but　it　only　negatively　impacts　trip　density.