Year: 2012
Type: article
Abstract: Automated visual classification becomes increasingly practical as cameras and computational power become more accessible.Already, many existing automated visual inspection methods are competitive with human levels of performance.This dissertation presents a semi-supervised visual classification algorithm based on Independent Component Analysis, Yan Karklin's Hierarchical Bayesian Model algorithm and Bayes' Theorem.Several problems of current interest, both in industry and the research literature, are used as example applications: mobile robotics, industrial inspection and mammography.Results are given demonstrating the effectiveness of this method.
Author: Richard Earl Hudson
Institution: Case Western Reserve University
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2526362167