A　method　operating　at　an　outdoor　swimming　pool　is　presented　in　this　paper.　In　a　swimming　pool,　one　key　problem　we　faced　is　that　the　SNR　(signal　to　noise　ratio)　will　be　lower　than　general　surroundings　such　as　bubbles,　shadow,　water　waves,　and　lower　contrast.　In　order　to　eliminate　noise　and　preserve　the　edge　and　texture　detail　in　the　image　that　we　obtained,　we　select　the　wavelet　denoising　algorithm.　Because　the　contrast　is　lower　and　the　wavelet　transform　will　make　the　image　fuzzy,　firstly,　we　use　the　canny　edge　detection　operator　to　detect　the　feature　of　the　image.　generally,　water　waves　are　in　the　special　location　for　the　swimming　pool,　we　ignore　the　water　waves　edge　information.　Secondly,　we　use　wavelet　transform　to　process　the　processed　edge　image　and　the　original　image,　then　we　add　these　wavelet　coefficients　at　the　corresponding　level,　we　will　reserve　the　first　level　wavelet　transform　coefficients.　Finally,　we　can　use　the　Donoho　hard　threshold　denoise　technology　to　eliminate　the　noise.　For　our　actual　situation,　our　algorithm　makes　some　improvement　in　two　aspects.　The　experiment　shows　that　image　denoise　perform　better　than　that　using　traditional　method.　©2010　IEEE.