A　diagnosis　of　corrosion　failure　in　steel　wire　ropes　based　on　wavelet　and　support　vector　machine　(SVM)　is　investigated.　Wavelet　decomposition　and　reconstruction　algorithm　is　used　to　denoise　the　magnetic　flux　leakage　(MFL)　signals,　then　peak　value,　trough　value,　peak-to-peak　value　and　energy　parameters　are　extracted　as　fault　feature　vectors,　and　a　SVM　classifier　is　designed　and　used　to　conduct　the　fault　pattern　recognition.　The　experimental　results　show　that　the　accuracy　of　the　SVM　classification　achieves　to　100%.　©　(2014)　Trans　Tech　Publications,　Switzerland.