Many　methods　can　be　used　to　detect　structural　defects.　In　general,　vibration　modes　are　very　sensitive　to　the　existence　of　defects.　The　information　of　modes　includes　the　local　feature　information　of　investigated　structure　and　the　inherent　properties,　so　it　can　be　used　to　detect　the　presence　of　defects　and　to　identify　their　locations.　However,　when　the　defects　are　small,　it　is　difficult　to　determine　the　existence　of　the　defects　by　relying　on　the　vibration　modal　shapes.　A　new　method　is　proposed　to　detect　the　weak　defect　information　in　cantilever　beam　by　using　the　Stationary　Wavelet　Transform　(SWT)　method　to　analyze　the　vibration　modal　shapes.　Finite　element　modal　analysis　is　employed　to　mimic　the　modal　shape　of　a　cracked　cantilevered　beam　with　various　defect　depths,　widths　and　positions.　SWT　is　applied　to　obtain　the　first　four　modal　shapes.　Results　obtained　demonstrate　that　the　proposed　method　can　be　used　for　the　identification　of　small　defects.　Meanwhile,　the　peak　of　stationary　wavelet　detail　coefficient　curve　would　increase　with　the　increase　of　the　defect　depth　and　defect　width.　Furthermore,　in　the　engineering　applications,　it　is　suggested　to　integrate　several　vibration　modal　shapes　to　identify　defects　to　minimize　the　vibration　node　effect　©　2016,　Editorial　Department　of　CJAM.　All　right　reserved.