With　the　development　of　artificial　intelligence　technology,　urban　traffic　management　has　become　increasingly　convenient,　and　the　task　of　illegal　parking　detection　has　become　a　major　research　focus.　Currently,　most　illegal　parking　detection　schemes　use　fixed-point　cameras,　which　not　only　waste　resources　but　also　have　the　limitation　of　a　small　detection　range.　To　overcome　this　issue,　we　have　designed　a　highly　mobile　system　for　detecting　illegally　parked　vehicles.　Considering　the　different　relative　positions　of　the　tires　and　parking　lines　in　roadside　parking　areas,　we　use　an　optimized　You　Only　Look　Once(YOLOv5)　algorithm　to　classify　the　tires　and　determine　whether　the　vehicle　is　illegally　parked　based　on　different　category　of　tires　combinations.　Subsequently,　we　applied　the　above-mentioned　illegal　parking　detection　strategy　to　a　embedded　device　which　mounted　on　the　mobile　vehicle　to　realize　the　real-time　mobile　detection　and　information　collection　of　illegal　vehicles　in　the　roadside　parking　area.　©　2024　Copyright　held　by　the　owner/author(s).