It　is　difficult　to　realize　the　quantitative　detection　of　defects　by　using　the　time　domain　or　frequency　domain　features　of　Lamb　wave　signals　due　to　their　dispersion　and　multi-modal　characteristics.　The　study　of　defect　detection　quantitatively　in　aluminum　plates　is　carried　out　by　means　of　wavenumber　analysis　based　on　the　full-optical　laser　ultrasonic　testing　system.　The　time-space　wavefield　signals　are　obtained　by　pulse　laser　in　the　fixed　location　excitation　and　continuous　laser　one-dimensional　scanning　acquisition,　where　the　propagation　characteristics　of　Lamb　wave　signal　and　the　regular　pattern　of　interaction　between　Lamb　wave　signal　and　defect　can　be　visually　displayed.　The　two-dimensional　Fourier　transform　is　used　to　convert　the　wavefield　signal　from　the　time-space　domain　to　the　frequency-wavenumber　domain　where　individual　modes　included　in　the　signals　are　well　discernible.　Based　on　the　idea　of　short-time　Fourier　transform,　for　retaining　the　space　information,　a　short-space　two-dimensional　Fourier　transform　is　adopted　to　obtain　the　distribution　of　wavenumber　along　the　scanning　path,　where　the　location　and　size　of　defect　can　be　visually　displayed.　Furthermore,　the　thickness　of　the　aluminum　plates　in　the　defect　area　can　be　calculated　based　on　the　relationship　between　the　wavenumber　and　the　frequency-thickness　product.　Experimental　results　show　that　the　method　can　effectively　evaluate　the　location,　size　and　depth　of　the　defects.　©　2018　Journal　of　Mechanical　Engineering.