Show simple item record

dc.contributor.advisorAlan Bovik
dc.creatorReaves, David
dc.date.accessioned2011-09-01T20:36:54Z
dc.date.available2011-09-01T20:36:54Z
dc.date.created2008-05
dc.date.issued2011-09-01
dc.identifier.urihttp://hdl.handle.net/2152/13371
dc.description.abstractRapidly advancing technologies o er a greater volume of people the possi- bility to both create and consume information. And, with this widening of opportunity, the volume of digital information has increased in mammoth proportion. Indeed, this age of information is marked by quantity, but what of quality? It has become necessary to formulate a systematic method to sift through the vast amount of data. This paper presents an algorithm that seeks to emulate the manner by which a human might judge an image's aesthetic value. The notion that a machine could imitate human thought processes is not necessarily novel, and, as such, a fair amount of work has been done regarding algorithmic aesthetic digital image rating. Most of these proposed algorithms, however, have been unable to satisfactorily mimic ac- tual human ratings. This paper builds on these past works and yet goes further by signi cantly improving on these prior accomplishments. The re- sult of our focus on the discovery of an optimal vector of image features is a highly accurate emulation of human ratings.en_US
dc.language.isoengen_US
dc.subjectCollege of Natural Sciencesen_US
dc.subjectalgorithmen_US
dc.subjectaesthetic valueen_US
dc.subjectaesthetic image ratingen_US
dc.titleAesthetic image rating (AIR) algorithmen_US
dc.typeThesisen_US
dc.description.departmentComputer Sciencesen_US


Files in this item

Icon

This item appears in the following Collection(s)

Show simple item record