(2006) Edge-Directed Invariant Shoeprint Image Retrieval. In: IEE Visual Information Engineering Confence, September 2006, Bangalore, India.
|
PDF (Peer-Reviewed Paper)
VIE-With-DC-AB-2006-Wave-Maxima.pdf Download (2MB) | Preview |
Abstract
In this paper, we propose the use of image feature points for the classification of shoeprint images in a forensic setting. These feature points are quantified using wavelet maxima points extracted from a nonorthogonal wavelet decomposition of the shoeprint images. Wavelet transforms have been shown to be an effective analysis tool for image indexing, retrieval and characterization. This effectiveness is mainly attributed to the ability of the latter transforms to capture well the spatial information and visual features of the analyzed images using only few dominant subband coefficients. In this work, we propose the use of a nonorthogonal multiresolution representation to achieve shift-invariance. To reduce the content redundancy, we limit the feature space to wavelet maxima points. Such dimensionality reduction enables compact image representation while satisfying the requirements of the "information-preserving" rule. Based on the wavelet maxima representations, we suggest the use of a variance-weighted minimum distance measure as the similarity metric for image query, retrieval, and search purposes. As a result, each image is indexed by a vector in the wavelet maxima moment space. Finally, performance results are reported to illustrate the robustness of the extracted features in searching and retrieving of shoeprint images independently of position, size, orientation and image background.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Uncontrolled Keywords: | Scene of crime evidence, shoeprint images, wavelet maxima, moment-invariant features, Hu and geometric moments. |
| Subjects: | Computer Electrical |
| Department: | College of Computing and Mathematics > Information and Computer Science |
| Depositing User: | LAHOUARI GHOUTI |
| Date Deposited: | 18 Jun 2008 08:02 |
| Last Modified: | 01 Nov 2019 13:46 |
| URI: | http://eprints.kfupm.edu.sa/id/eprint/9149 |