Texture discrimination using discrete cosine transformation shift-insensitive (DCTSIS) descriptors

@article{Frye2000TextureDU,
  title={Texture discrimination using discrete cosine transformation shift-insensitive (DCTSIS) descriptors},
  author={Richard Eugene Frye and Robert S. Ledley},
  journal={Pattern Recognition},
  year={2000},
  volume={33},
  pages={1585-1598}
}
Many of the numerous texture measurements are based on space-frequency signal decomposition; these include Gabor "lters and wavelet-based methods. The discrete cosine transformation (DCT) extracts spatial-frequency (SF) components from a local image region. It is the basis for the JPEG image compression standard and has many fast algorithmic implementations. By using a sliding DCT we derive a SF representation for a region of interest (ROI) surrounding each image pixel. We show that the DCT coe… CONTINUE READING