Moving　objectives　detection　is　the　basis　for　the　application　of　computer　vision　technology,　in　intelligent　video　surveillance　systems　in　many　areas　have　a　wide　range　of　applications　for　dynamic　in　the　context　of　the　moving　object　detection,　presented　in　this　paper　that　will　improve　the　pyramid　optical　flow　law　and　the　combination　of　edge　detection　moving　object　detection　method.　First　of　video　pre-processing　applications,　Wavelet　Threshold　to　remove　noise;　second,　through　the　pyramid　optical　flow　to　achieve　maximum　displacement　appears　moving　objectives;　then　uses　a　canny　edge　detection　for　the　edge　of　an　image　information;　and　finally　will　get　the　optical　flow　of　information　and　the　edge　of　the　characteristics　of　the　information　so　the　second　value　after　the　integration　of　the　morphology　through　the　follow-up　to　the　campaign　to　get　accurate　target　information.　Lab　results　indicate　that　the　algorithm　real-time　high,　the　versatile　good,　you　can　more　accurately　detect　moving　objectives.