Regarding　the　strong　noise　in　the　vibration　signal　induced　by　the　roller　bearing　fault　and　the　difficulty　to　obtain　fault　characteristic　frequencies　in　practice,　a　new　fault　diagnosis　method　is　proposed　based　on　dual-tree　complex　wavelet　transform　(DT-CWT)　and　singular　value　difference　(SVD)　spectrum.　Firstly,　original　fault　signals　are　decomposed　into　several　different　frequency　band　components　through　DT-CWT;　Secondly,　the　Hankel　matrix　is　constructed　by　the　component　which　contains　the　fault　information,　and　the　singular　value　difference　spectrum　can　be　obtained　after　SVD.　Then　the　maximum　catastrophe　point　is　used　to　identify　the　number　of　singular-value　reconstruction　component.　Finally,　the　fault　frequency　can　be　identified　accurately　by　the　Hilbert　envelope　spectrum.　The　results　of　the　experiments　and　engineering　application　show　that　the　fault　characteristics　of　the　roller　bearing　can　be　separated　and　extracted　effectively,　getting　the　feasibility　and　effectiveness　of　the　method　verified.