For　the　shortage　of　artificial　methods　in　classification　of　apple　level,　a　method　of　classification　of　apple　level　based　on　BP　artificial　neural　networks　is　presented.　Artificial　neural　network　provides　technical　means　for　research　on　apple　classification,because　it　can　provide　nonlinear　mapping　of　the　input　vector　and　output　vector　with　arbitrary　dimension,　and　also　can　reach　any　nonlinear　continuous　system.　The　BP　neural　network　model　is　established　for　the　classification　by　taking　feature　parameters　as　input　vectors　and　apple　level　as　output　vectors.　The　results　has　shown　that　the　classification　precision　of　model　is　very　high,　and　which　has　good　application　in　realizing　automatic　identification　of　apple　level.　©　2011　IEEE.