?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Coupled+dictionary+learning+for+multimodal+image+super-resolution&rft.creator=De+Castro+Mota%2C+JF&rft.creator=Song%2C+P&rft.creator=Deligiannis%2C+N&rft.creator=Rodrigues%2C+MRD&rft.description=Real-world+data+processing+problems+often+involve+multiple+data+modalities%2C+e.g.%2C+panchromatic+and+multispectral+images%2C+positron+emission+tomography+(PET)+and+magnetic+resonance+imaging+(MRI)+images.+As+these+modalities+capture+information+associated+with+the+same+phenomenon%2C+they+must+necessarily+be+correlated%2C+although+the+precise+relation+is+rarely+known.+In+this+paper%2C+we+propose+a+coupled+dictionary+learning+(CDL)+framework+to+automatically+learn+these+relations.+In+particular%2C+we+propose+a+new+data+model+to+characterize+both+similarities+and+discrepancies+between+multimodal+signals+in+terms+of+common+and+unique+sparse+representations+with+respect+to+a+group+of+coupled+dictionaries.+However%2C+learning+these+coupled+dictionaries+involves+solving+a+highly+non-convex+structural+dictionary+learning+problem.+To+address+this+problem%2C+we+design+a+coupled+dictionary+learning+algorithm%2C+referred+to+sequential+recursive+optimization+(SRO)+algorithm%2C+to+sequentially+learn+these+dictionaries+in+a+recursive+manner.+By+capitalizing+on+our+model+and+algorithm%2C+we+conceive+a+CDL+based+multimodal+image+super-resolution+(SR)+approach.+Practical+multispectral+image+SR+experiments+demonstrate+that+our+SR+approach+outperforms+the+bicubic+interpolation+and+the+state-of-the-art+dictionary+learning+based+image+SR+approach%2C+with+Peak-SNR+(PSNR)+gains+of+up+to+8.2+dB+and+5.1+dB%2C+respectively.&rft.subject=coupled+dictionary+learning%2C+multimodal+data%2C+sparse+representation%2C+sequential+recursive+optimization%2C%0D%0Amultispectral+image+super-resolution%2C+Dictionaries%2C%0D%0AImage+resolution%2C+Signal+resolution%2C+Data+models%2C+Training%2C%0D%0AOptimization%2C+Sparse+matrices&rft.publisher=Institute+of+Electrical+and+Electronics+Engineers+(IEEE)&rft.date=2017-04-24&rft.type=Proceedings+paper&rft.publisher=IEEE+Global+Conference+on+Signal+and+Information+Processing+(GlobalSIP)&rft.language=eng&rft.source=+++++In%3A++2016+IEEE+Global+Conference+on+Signal+and+Information+Processing+(GlobalSIP).++(pp.+pp.+162-166).++Institute+of+Electrical+and+Electronics+Engineers+(IEEE)%3A+New+York%2C+USA.+(2017)+++++&rft.format=text&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F1529221%2F1%2FCoupledDLforImgSR_FinalVer.pdf&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F1529221%2F&rft.rights=open