Automatic detection of image orientation using Support Vector Machines

[摘要] ENGLISH ABSTRACT:In this thesis, we present a technique for the automatic detection of image orientation using SupportVector Machines (SVMs). SVMs are able to handle feature spaces of high dimension and automaticallychoose the most discriminative features for classification. We investigate the use of variouskernels, including heavy tailed RBF kernels. We compare the classification performance of SVMswith the performance of multilayer perceptrons and a Bayesian classifier. Our results show that SVMsout perform both of these methods in the classification of individual images. We also implement anapplication for the classification of film rolls in a photographic workflow environment with 100%classification accuracy.

[发布机构] Stellenbosch University