?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=WAVELET+BASED+NONLINEAR+SEPARATION+OF+IMAGES&rft.creator=Almeida%2C+Mariana+S.+C.&rft.creator=Almeida%2C+Lu%C3%ADs+B.&rft.subject=Statistical+Models&rft.subject=Machine+Vision&rft.subject=Machine+Learning&rft.subject=Artificial+Intelligence&rft.description=This+work+addresses+a+real-life+problem+corresponding%0Ato+the+separation+of+the+nonlinear+mixture+of+images+which%0Aarises+when+we+scan+a+paper+document+and+the+image+from%0Athe+back+page+shows+through.%0AThe+proposed+solution+consists+of+a+non-iterative+procedure%0Athat+is+based+on+two+simple+observations%3A+(1)+the+high%0Afrequency+content+of+images+is+sparse%2C+and+(2)+the+image%0Aprinted+on+each+side+of+the+paper+appears+more+strongly+in%0Athe+mixture+acquired+from+that+side+than+in+the+mixture+acquired+from+the+opposite+side.%0AThese+ideas+had+already+been+used+in+the+context+of+nonlinear+denoising+source+separation+(DSS).+However%2C+in+that+method+the+degree+of+separation+achieved+by+applying+these+ideas+was+relatively+weak%2C+and+the+separation+had+to+be+improved+by+iterating+within+the+DSS+scheme.+In+this+paper+the+application+of+these+ideas+is+improved+by+changing+the+competition+function+and+the+wavelet+transform+that+is+used.+These+improvements+allow+us+to+achieve+a+good+separation+in+one+shot%2C+without+the+need+to+integrate+the+process+into+an+iterative+DSS+scheme.+The+resulting+separation+process+is+both+nonlinear+and+non-local.%0AWe+present+experimental+results+that+show+that+the+method%0Aachieves+a+good+separation+quality.&rft.date=2006&rft.type=Preprint&rft.type=NonPeerReviewed&rft.format=application%2Fpdf&rft.identifier=http%3A%2F%2Fcogprints.org%2F4915%2F1%2FDenoising.pdf&rft.identifier=++Almeida%2C+Mariana+S.+C.+and+Almeida%2C+Lu%C3%ADs+B.++(2006)+WAVELET+BASED+NONLINEAR+SEPARATION+OF+IMAGES.++%5BPreprint%5D+++++&rft.relation=http%3A%2F%2Fcogprints.org%2F4915%2F