Many　JPEG　steganography　techniques　are　used　to　communicate　secret　messages　by　terrorists　that　threaten　the　security　of　a　nation.　So　steganalysis　is　very　important.　Avcibas　proposed　a　steganalysis　based　on　binary　similarity　measures　which　only　work　well　on　LSB-based　steganography　and　derived　features　from　the　spatial　domain　of　images.　This　paper　proposes　a　novel　usage　of　binary　similarity　measures　in　JPEG　steganalysis.　The　method　captures　the　seventh　and　eighth　bit　planes　of　the　non-zero　DCT　coefficients　from　JPEG　images　and　computes　14　features　of　each　image　based　on　binary　similarity　measures.　These　features　are　used　to　construct　a　support　vector　machine　classifier　which　can　distinguish　between　stego　images　and　cover　images.　The　experiment　results　are　presented　to　demonstrate　that　the　proposed　scheme　has　lower　computational　complexity　and　the　same　high　detecting　accuracy.