Disclosed　is　a　visual　perception　feature-fused　scaled　video　coding　method,　　comprising　:　　the　setting　of　a　priority　of　a　visual　region-of-interest　and　setting　of　a　video　coding　resource　allocation　solution.　The　former　comprises　:　　in　consideration　of　the　richness　of　contents　in　a　video　image　and　a　vision　selective　attention　mechanism　for　human　eyes,　　respectively　performing　a　time　domain　and　a　space　domain　visual　feature　prominence　analysis　on　the　contents　of　the　video,　　and　finishing　labelling　of　the　priority　of　the　visual　region-of-interest.　The　latter　demonstrates　that　:　　in　order　to　improve　the　video　coding　real-time　performance　while　ensuring　a　video　coding　quality　and　compression　efficiency,　　according　to　the　priority　of　the　visual　region-of-interest,　　firstly,　　optimization　of　coding　resources　of　a　macro　block　in　the　region-of-interest　is　satisfied,　　scaled　coding　is　realized,　　and　the　contradiction　between　the　coding　complexity　and　the　coding　efficiency　is　effectively　relieved;　　and　compared　with　an　H.264/AVC　international　video　coding　standard,　　an　average　coding　speed　is　improved　by　about　80%　under　the　premise　of　ensuring　a　higher　video　image　quality　and　a　low　compression　code　rate,　　and　the　video　coding　performance　are　obviously　improved.