Restructured Eigenfilter Matching for Novelty Detection in Random Textures

@inproceedings{Monadjemi2004RestructuredEM,
  title={Restructured Eigenfilter Matching for Novelty Detection in Random Textures},
  author={A. Monadjemi and Majid Mirmehdi and Barry T. Thomas},
  booktitle={BMVC},
  year={2004}
}
A new eigenfilter-based novelty detection approach to find abnormalities in random textures is presented. The proposed algorithm reconstructs a given texture twice using a subset of its own eigenfilter bank and a subset of a reference (template) eigenfilter bank, and measures the reconstruction error as the level of novelty. We then present an improved reconstruction generated by structurally matched eigenfilters through rotation, negation, and mirroring. We apply the method to the detection of… CONTINUE READING
Highly Cited
This paper has 22 citations. REVIEW CITATIONS