The　new　MPEG-H/H.265　High　Efficiency　Video　Coding(HEVC)　standard　greatly　promotes　the　video　compression　efficiency　and　also　creates　new　opportunities　and　challenges　in　perceptual　video　coding.　To　measure　the　perceptual　quality,　Mean　Opinion　Score　(MOS)　is　obtained　by　averaging　human　scores　in　subjective　test.　But　its　value　cannot　necessarily　represent　exact　difference　between　the　human　opinions.　To　solve　this　problem,　interval　MOS　(iMOS)　method,　which　employs　the　distribution　information　of　human　scores　to　measure　the　difference　between　MOS　values　with　deviations,　is　introduced.　Furthermore,　the　feasibility　of　using　iMOS　to　evaluate　compressed　video　quality　is　confirmed　by　experiments.