An　artificial　neural　network　(ANN)　is　a　parallel,　distributed　processing　system　consisting　of　a　large　number　of　interconnected　neurons.　At　present,　artificial　neural　networks　have　been　widely　used　in　signal　processing,　automation,　control　systems,　image　recognition　and　many　other　fields.　Usually,　the　realization　of　neural　networks　is　based　on　software.　However,　because　the　software　implementation　method　cannot　achieve　real-time　calculation　in　many　cases,　the　hardware　implementation　method　can　reflect　the　inherent　parallel　processing　characteristics　of　neural　networks.　This　paper　proposes　an　ANN　neural　network　field　programmable　neuron　array　based　on　the　design　of　reusable　ANN　artificial　neural　network　IP　core.　It　not　only　can　save　a　lot　of　hardware　resources　and　improve　processing　speed,　but　also　has　a　rapid　development　cycle　and　good　reconfigurability　and　other　advantages.