Region of Interest
It is a challenging work to solve the geometric attack in the field of digital watermarking. In order to solve the synchronization between the host image and the watermark, image normalization is introduced. Firstly, the geometrically invariant space of image is constructed by using image normalization, and a region of interest (ROI) is obtained from the normalized image by utilizing the invariant centroid theory. Then, the contourlet transform is performed on the ROI. Low-pass sub-band coefficients are divided into non-overlapping blocks.
- Categories:
125 Views
70 images with a total of more than 2 GB of data have been employed for the experimental evaluation. All images are acquired at 24 bpp with 8 bits per pixel per component (bpppc). They depict various tissues of different sizes and stained with Hematoxylin and Eosin (H\&E) stain. The tissues employed in the experiments are skin fibroblast (SKNF), endometrial (END), lung (LNGF), embryonic stem cells (ES), kidney clear cell carcinoma (KIRC), pancreas (PANC), brain glioblastoma multiforme (GBM), colon adenocarcinoma (COAD), and lymphatic (LYMP).
- Categories:
316 Views