?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Accurate+and+robust+image+superresolution+by+neural+processing+of+local+image+representations+&rft.creator=Miravet%2C+Carlos&rft.creator=Rodriguez%2C+Francisco+B.&rft.subject=Machine+Vision&rft.subject=Neural+Nets&rft.description=Image+superresolution+involves+the+processing+of+an+image+sequence+to+generate+a+still+image+with+higher+resolution.+Classical+approaches%2C+such+as+bayesian+MAP+methods%2C+require+iterative+minimization+procedures%2C+with+high+computational+costs.+Recently%2C+the+authors+proposed+a+method+to+tackle+this+problem%2C+based+on+the+use+of+a+hybrid+MLP-PNN+architecture.+In+this+paper%2C+we+present+a+novel+superresolution+method%2C+based+on+an+evolution+of+this+concept%2C+to+incorporate+the+use+of+local+image+models.+A+neural+processing+stage+receives+as+input+the+value+of+model+coefficients+on+local+windows.+The+data+dimension-ality+is+firstly+reduced+by+application+of+PCA.+An+MLP%2C+trained+on+synthetic+se-quences+with+various+amounts+of+noise%2C+estimates+the+high-resolution+image+data.+The+effect+of+varying+the+dimension+of+the+network+input+space+is+exam-ined%2C+showing+a+complex%2C+structured+behavior.+Quantitative+results+are+presented+showing+the+accuracy+and+robustness+of+the+proposed+method.&rft.publisher=Springer-Verlag&rft.date=2005&rft.type=Conference+Paper&rft.type=PeerReviewed&rft.format=application%2Fpdf&rft.identifier=http%3A%2F%2Fcogprints.org%2F4567%2F1%2Fmiravet_rodriguez_arxiv_05.pdf&rft.identifier=++Miravet%2C+Carlos+and+Rodriguez%2C+Francisco+B.++(2005)+Accurate+and+robust+image+superresolution+by+neural+processing+of+local+image+representations.++%5BConference+Paper%5D+++++&rft.relation=http%3A%2F%2Fcogprints.org%2F4567%2F