When　people　observe　pictures,　different　pictures　will　generate　different　emotions,　and　the　painters　often　convey　emotional　energy　to　the　audience　through　the　media.　Through　the　effect　of　this　emotional　transfer,　people　get　emotional　satisfaction　psychologically,　and　the　author　can　better　express　the　emotion　he　needs　to　express　by　using　the　principle.　For　those　who　do　not　know　art　appreciation,　it　is　difficult　to　accurately　describe　the　feelings　brought　by　pictures　and　describe　them　accurately.　It　is　also　difficult　to　quickly　and　accurately　find　the　picture　of　the　corresponding　feeling　sought　by　oneself.　This　paper　presents　a　quantitative　picture　emotion　analysis　method.　Based　on　the　analysis　of　the　characteristics　and　colors　of　the　pictures　that　arouse　people＇s　emotions,　the　classification　planning　is　carried　out　in　RGB　space　and　HSV　color　space　respectively.　And　the　combination　analysis　of　the　color　features　with　strong　influence　is　carried　out　to　select　a　more　appropriate　color　space　for　the　expression　of　human　feelings.　At　the　same　time,　the　paper　also　discusses　the　effect　of　the　black-and-white　filling　and　the　brightness　saturation　of　the　image.　Finally,　it　makes　use　of　the　corresponding　design　principles　to　carry　out　a　quantitative　analysis　of　the　emotion　generated　by　it.　The　system　performs　a　combined　analysis　of　a　picture　from　these　directions　and　presents　the　results　to　the　user.