000000260 001__ 260 000000260 005__ 20260420125549.0 000000260 0247_ $$2DOI$$a10.6083/M4D21VJ4 000000260 037__ $$aETD 000000260 245__ $$aBiologically inspired visual models by sparse and unsupervised learning 000000260 260__ $$bOregon Health and Science University 000000260 269__ $$a2007 000000260 336__ $$aDissertation 000000260 502__ $$bPh.D. 000000260 502__ $$gComputer Science & Electrical Engineering (sunsetting) 000000260 520__ $$aThis research presents biologically inspired visual models and algorithms for automatic part extraction and object recognition from gray-scale images with approximate rotational invariance. Motivated by the efficiency of the primate visual system, the models employ sparse representation and unsupervised learning to achieve fault tolerance, low power consumption, and adaptive feature discovery without prior knowledge. A hierarchical architecture inspired by the primate ventral visual pathway was developed, including models of V1 and V2 for low-level feature extraction, V4 for parts-based shape representation, and IT for high-level object recognition. The models demonstrate efficient, transformation-tolerant visual processing, with V4 units exhibiting biologically consistent curvature and object-centered tuning. Using flexible constellations of rigid parts, the IT model achieves robust object recognition across viewpoints. Overall, this work integrates sparse coding and unsupervised learning into a biologically plausible framework with strong performance in visual recognition tasks. 000000260 540__ $$fCC BY 000000260 542__ $$fIn copyright - single owner 000000260 650__ $$aMachine Learning$$011449 000000260 650__ $$aVisual Pathways$$027863 000000260 650__ $$aVisual Perception$$027864 000000260 650__ $$aPrimates$$024593 000000260 650__ $$aNeural Networks, Computer$$029229 000000260 650__ $$aUnsupervised Machine Learning$$011455 000000260 6531_ $$ashape representation 000000260 6531_ $$asparse coding 000000260 6531_ $$atransformation invariance 000000260 6531_ $$aobject recognition 000000260 691__ $$aOGI School of Science and Engineering$$041365 000000260 692__ $$aDepartment of Computer Science and Electrical Engineering$$041404 000000260 7001_ $$aYang, Li$$uOregon Health and Science University$$041354 000000260 7201_ $$aPavel, Misha$$uOregon Health and Science University$$041354$$7Personal$$eAdvisor 000000260 8564_ $$96d552a59-e10e-4dee-a83d-3de94f00fb43$$s3921500$$uhttps://digitalcollections.ohsu.edu/record/260/files/260_etd.pdf$$ePublic$$21c1ca139bc76027d5836cb1b6759e8b2$$31 000000260 901__ $$a<p>These documents are archival records. They are retained for historical reference only. </p><p><b>Need an accessible version? Use the ‘Get Accessible Copy’ link above.</b></p> 000000260 905__ $$a/rest/prod/kh/04/dp/71/kh04dp71b 000000260 909CO $$ooai:digitalcollections.ohsu.edu:260$$pstudent-work 000000260 956__ $$aGet Accessible Copy$$uhttps://ohsu.libwizard.com/f/requestaccessibledocument 000000260 980__ $$aTheses and Dissertations