Previous work has shown Gaussian mixture vector quantization (GMVQ) based classifiers to be effective in classifying image blocks, image regions and whole images. A significant attraction of GMVQ for whole image classification is that simple local features can be used, thereby avoiding time-consuming feature design and selection. Unfortunately, however, this approach does not work so well when the artifact of interest occupies a small area relative to the size of the image. We propose a simple weighting approach to focus the classifier's attention on the artifact blocks. This extends the usefulness of whole image GMVQ classification without compromising the simplicity of feature selection. The algorithm is motivated by difficulties in classifying pipeline images. Results on this dataset show the weighted GMVQ approach to be effective for classifying images with small artifacts of interest.
Gaussian mixture model classifiers for small objects in images
2005-01-01
131439 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Gaussian Mixture Model Classifiers for Small Objects in Images
British Library Conference Proceedings | 2005
|Gaussian mixture model for edge-enhanced images
British Library Online Contents | 2004
|Spatiotemporal Gaussian mixture model to detect moving objects in dynamic scenes
British Library Online Contents | 2007
|Image Denoising With Gaussian Mixture Model
IEEE | 2008
|A self-adaptive Gaussian mixture model
British Library Online Contents | 2014
|