Super-resolution　image　reconstruction　refers　to　as　restoring　a　high-resolution　and　high-quality　image　from　multiple　low-resolution　observations　degraded　by　warping,　blurring,　noise　and　aliasing.　The　projections　onto　convex　sets　(POCS)　algorithm　is　widely　used　for　super-resolution　image　reconstruction.　In　this　paper,　we　propose　an　improved　POCS　algorithm　that　reduces　the　amount　of　edge　artifacts　present　in　the　high-resolution　reconstructed　image.　The　blur　point　spread　function　(PSF)　centered　at　an　edge　pixel　is　weighted　by　an　exponential　function,　so　that　the　coefficients　of　the　modified　PSF　could　decrease　in　the　direction　orthogonal　to　the　edge.　Experimental　results　show　that　the　modification　effectively　reduces　the　visibility　of　the　artifacts　on　the　edges　and　obviously　improves　the　quality　of　the　reconstructed　image.