Screen　content　denotes　images　generated　by　computers,　which　consists　of　texts,　graphics　and　synthetic　images.　Conventional　encoding　strategies　developed　for　natural　content,　which　are　premised　on　the　strong　temporal　relevance,　may　not　work　well　for　screen　content.　This　is　because　the　irregular　motion　of　objects　results　in　the　weak　temporal　relevance　in　screen　content,　which　also　makes　the　selection　of　reference　pictures　not　optimal.　This　paper　proposes　an　optimal　picture　reordering　scheme　for　screen　content　coding　to　enhance　the　temporal　relevance　of　screen　content.　In　our　scheme,　a　concept　of　inter-picture　correlation　is　introduced　to　help　formulate　the　temporal　relevance　problem,　and　a　model　of　permutation　is　proposed　to　produce　the　new　order　of　the　pictures.　To　further　improve　the　coding　efficiency,　a　reference　picture　selecting　algorithm　is　also　proposed.　Better　compression　performance　can　then　be　achieved.　Compared　with　SCC　extension　of　HEVC,　on　average,　2.82%　bit　rate　saving　can　be　achieved　by　reordering　the　pictures　and　4.40%　bit　saving　is　achieved　by　further　utilizing　the　reference　pictures　selection　algorithm.