This　paper　presents　a　hybrid　cameras　visual　servo　robot　system,　which　combines　both　Eye-in-hand　and　Eye-to-hand　cameras,　to　perform　the　task　of　picking　and　precise　placing　for　moving　workpieces　on　conveyor　belt.　The　Eye-to-hand　camera　is　applied　to　capture　images　that　evaluate　the　velocity　of　moving　workpieces　and　detect　the　position　error　of　robot　gripper　and　workpieces.　The　precise　velocity　of　workpieces　is　measured　by　using　visual　geometry　algorithm　and　the　position　error　is　calculated　by　the　pinhole　imaging　mode.　Since　lots　of　errors,　including　image　processing,　structure　of　robot　gripper,　calculation　of　velocity,　etc,　exist　in　system,　it　is　necessary　to　correct　the　position　error　between　workpieces　and　robot　gripper,　so　that　workpieces　are　able　to　be　precise　placed　to　expected　position.　The　results　show　that　the　proposed　hybrid　cameras　visual　servo　robot　system　can　grasp　and　place　workpieces　effectively.