Analysis of the GCR Communication Channel For Image Steganography
| Autor: | Ante Poljičak, Tomislav Cigula, Petar Branislav Jelusic, Davor Donevski, Tamara Tomašegović |
|---|---|
| Přispěvatelé: | Muštra, Mario, Vuković, Josip, Zovko-Cihlar, Branka |
| Rok vydání: | 2019 |
| Předmět: |
Computer science
Color image business.industry ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION 0211 other engineering and technologies Pattern recognition 02 engineering and technology Gray Component Replacement Image Quality Data hiding DFT DCT Color space symbols.namesake Fourier transform Frequency domain Information hiding Discrete cosine transform symbols RGB color model Artificial intelligence business Digital watermarking 021101 geological & geomatics engineering |
| Zdroj: | 2019 International Symposium ELMAR. |
| DOI: | 10.1109/elmar.2019.8918869 |
| Popis: | Main motivation behind this investigation is to determine how and to what extent Gray Component Removal (GCR) influences a color image. Specifically how GCR influences images in spatial domain, as well as in Fourier and Cosine domains. This is important to determine in order to develop a data hiding technique that can be used in GCR communication channel. To evaluate the influence of GCR on color images a large image dataset is first converted from RGB to CMYK color space with very aggressive GCR. Then the original and the converted images are compared in Fourier and Cosine domains. The results show that the GCR communication channel has a nonlinear impact on the image that depends exclusively on the content of the image. The impact in spatial domain is too severe for watermarking techniques in spatial domain. In contrast, the extent of the impact in frequency domain is low enough so GCR can be used for data hiding techniques based on both DFT and DCT transforms. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|