Image　inpainting　based　on　Partial　Differential　Equation　(PDE)　requires　a　lot　of　computation　which　makes　it　hard　to　be　applied　to　industry.　In　this　paper,　to　realize　high　image　processing　speed　and　flexibility,　the　development　of　parallel　algorithms　for　image　inpainting　based　on　BSCB　model　under　Message　Passing　Interface　(MPI)　environment　is　presented.　Three　methods　to　partition　an　image　inpainting　task　are　proposed　which　successfully　avoid　much　communication　among　subtasks.　Two　images　are　tested　and　the　experimental　results　on　a　cluster　of　processors　show　good　speedup.