According　to　the　advantages　of　invariant　moments　extraction　and　support　vector　machine　(SVM)　classification,　an　efficient　method　of　SAR　image　target　recognition　is　proposed.　Firstly,　training　sets　of　SAR　images　are　preprocessed.　Then　in　the　target　fields　of　SAR　image,　seven　Hu＇s　invariant　moments　are　extracted,　and　average　gray　level　values　are　also　calculated.　Following　that,　by　using　the　vectors　composed　of　normalized　invariant　moments　and　average　gray　level　values,　SVM　is　trained.　Finally,　the　testing　sets　of　SAR　images　are　recognized　by　trained　SVM.　In　the　experiments　of　recognizing　bridges　and　tanks,　better　results　have　been　obtained　with　the　new　method.