This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

Search for Publication

Year(s) from:  to 
Keywords (separated by spaces):

Multi-scale Indices for Content-based Image Retrieval

H. Rehrauer, K. Seidel and M. Datcu
IEEE Intern. Geoscience and Remote Sensing Symposium IGARSS'99


For automatic remote-sensing image interpretation it is important to give specific consideration to the resolution of the data. The performance of content-based retrieval systems can be enhanced significantly if the scale of spatial features is used explicitly as a meta feature. This allows to use low-dimensional feature vectors at each scale instead of a high-dimensional feature vector for all scales. We developed a system where the user can query for signal properties, like texture characteristics. He is encouraged to restrict the search space by indicating the scale he is interested in. The signal oriented search is done using indices that are computed completely unsupervised. These indices represent the characteristic signal classes of the data. The system does no interpretation of the classes. It is up to the user to ``name'' the contents according to his application interest.

Download in postscript format
  author = {H. Rehrauer and K. Seidel and M. Datcu},
  title = {Multi-scale Indices for Content-based Image Retrieval},
  booktitle = {IEEE Intern. Geoscience and Remote Sensing Symposium IGARSS'99},
  year = {1999},
  keywords = {remote sensing,scale space,database,texture}