Panic　emotion　appears　frequently　in　an　emergency　evacuation,　affecting　evacuation　efficiency　and　pedestrian　safety.　Individual　diversity　in　emotion　propagation　causes　pedestrians　to　behave　differently.　To　better　describe　the　emotion　propagation　in　the　movement　of　pedestrians,　we　analyze　the　characteristics　of　the　emotions　of　different　pedestrians　in　emergencies.　Considering　pedestrian　heterogeneity,　we　divide　pedestrians　into　three　types　(conservative,　moderate,　and　radical)　according　to　the　strength　of　the　expression　of　individual　emotions.　Based　on　an　analysis　of　the　characteristics　of　the　three　types　of　pedestrian,　an　improved　susceptible-infected-recovered　(SIR)　model　is　built　to　describe　the　pedestrian　emotion　propagation　process.　In　addition,　a　conflict　avoidance　method　that　considers　pedestrian　speed　is　proposed　to　demonstrate　pedestrian　motion　more　realistically.　Based　on　this　foundation,　a　cellular　automata　(CA)　model　that　integrates　emotion　propagation　and　conflict　avoidance　techniques　is　employed　to　replicate　the　process　of　pedestrian　emergency　evacuation.　The　simulation　results　suggest　that　a　higher　proportion　of　pedestrians　with　initial　panic　emotions　and　a　radical　behavior　type　leads　to　a　prolonged　evacuation　time.　Many　pedestrians　in　emergency　evacuations　gather　at　the　exit　in　a　short　time　(approximately　37　time　steps),　the　density　then　reaches　a　high-level　(approximately　5.5　pedestrians/m2)　for　a　period　of　time,　and　the　duration　increases　with　the　increase　in　the　number　of　pedestrians.　Methods　for　guiding　pedestrians　to　evacuate　in　a　rational　manner　are　worthy　of　promotion　and　research.