In this paper, we present a theoretically and computationally simple but efficient approach for rotation invariant texture classification. This method is based on new texture signatures extracted from spectrum. Rotation invariant texture features are obtained based on the extension of the derived signatures. The features are tested with 1000 randomly rotated samples of 20 Brodatz texture classes. Comparative study results show that our method is highly efficient in rotation invariant texture classification.
|Title of host publication||Proceedings of Texture 2002|
|Subtitle of host publication||the 2nd international workshop on texture analysis and synthesis with ECCV 2002|
|Place of Publication||Edinburgh|
|Number of pages||6|
|Publication status||Published - 2002|
|Event||The 2nd International Workshop on Texture Analysis and Synthesis - Copenhagen, Denmark|
Duration: 1 Jun 2002 → 2 Jun 2002
|Conference||The 2nd International Workshop on Texture Analysis and Synthesis|
|Period||1/06/02 → 2/06/02|
Zhang, J., & Tan, T. (2002). New texture signatures and their use in rotation invariant texture classification. In M. Chantler (Ed.), Proceedings of Texture 2002: the 2nd international workshop on texture analysis and synthesis with ECCV 2002 (pp. 157-162). Heriot-Watt University.