Traffic　pollution　reduction　policy　has　shown　the　trend　of　diversification,　refinement　and　differentiation,　the　limitations　of　the　traditional　traffic　emission　model　in　evaluation　scope　and　scale　have　become　increasingly　prominent.　This　paper　focus　on　the　research　of　traffic　flow　simulation　model　based　on　real-time　monitoring　data,　analysis　of　fleet　structure　based　on　visual　detection　method,　the　correction　relationship　of　velocity　emission　factor　based　on　local　operating　model,　from　which　a　high　resolution　analysis　method　of　urban　traffic　emission　is　proposed.　This　method　can　reach　the　hour　and　the　road　section　level.　The　emission　sources　can　be　distinguished　to　the　different　traffic　modes　and　the　local　field.　In　addition,　based　on　a　case　study　of　Beijing,　an　empirical　analysis　of　the　temporal　and　spatial　distribution　characteristics　of　NOx　emission　from　the　road　traffic　is　conducted.　The　result　shows　that　during　the　peak　period,　the　NOx　emission　on　the　road　network　can　take　up　to　31.2%　of　the　whole　day.　Besides,　highway　and　expressway　account　for　37.9%　and　38.8%　of　the　total　emission　respectively.　The　emission　intensity　between　the　5th　and　6th　ring　road　is　the　highest,　which　accounts　for　38.32%　within　the　6th　ring.　The　emission　of　freight　vehicles　can　reach　up　to　47%　of　all.　Finally,　the　research　points　out　that　the　high　resolution　analysis　method　of　urban　traffic　emission　is　of　significance　to　the　accurate　temporal,　spatial　and　vehicle　positioning　of　urban　traffic　pollution　control,　as　well　as　contributes　to　the　fine　decision-making.　Copyright　©　2018　by　Science　Press.