Semi-fragile self-recovery watermarking scheme based on data representation through combination

Rhayma, Hanen; Makhloufi, Achraf; Hamam, Habib; Hamida, Ahmed Ben

doi:10.1007/s11042-019-7244-x

Cited by 13 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of these transforms, for example the DFT, have limitations, since they can give only frequency without any time information, which will restrict us to deal only with stationary (frequency doesn't change in time) signals. Images of our concern here are considered as non-stationary signals; therefore, an alternative frequency domain transform that can give both frequency and time information should be considered, and wavelet is such a type [10]. Wavelet transform depends on two filter types: low and high pass filters.…”

Section: Wavelet Transformmentioning

confidence: 99%

“…As a result of these compression operations, there will actually be some loss in the data and certainly, among them, some watermark bits. As a consequence, the watermarked image can have high imperceptibility (because Human Visual System can't notice the changes that take place on high frequency bands while easily pick up such changes in the low frequency bands), as well as achieving a high robustness against unintentional change through compression [10]. Discrete wavelet transform (DWT) low-level sub bands can be used in order to embed watermark bits [9].…”

Section: Wavelet Transformmentioning

confidence: 99%

See 1 more Smart Citation

Image Watermarking Based on IWT and Parity Bit Checking

Awad

Majeed²

2021

eijs

View full text Add to dashboard Cite

Various document types play an influential role in a lot of our lives activities today; hence preserving their integrity is an important matter. Such documents have various forms, including texts, videos, sounds, and images. The latter types' authentication will be our concern here in this paper. Images can be handled spatially by doing the proper modification directly on their pixel values or spectrally through conducting some adjustments to some of the addressed coefficients. Due to spectral (frequency) domain flexibility in handling data, the domain coefficients are utilized for the watermark embedding purpose. The integer wavelet transform (IWT), which is a wavelet transform based on the lifting scheme, is adopted in this paper in order to provide a direct way for converting image pixels' integer values to integer coefficient values rather than floating point coefficients that could be produced by the traditional wavelet transform. This direct relation can enhance the processed image quality due to avoiding the rounding operations on the floating point coefficients. The well-known parity bit approach is also utilized in this paper as an authentication mechanism, where 3 secret parity bits are used for each block in an image which is divided into non-overlapped blocks in order to enforce a form of fragile watermark approach. Thus, any alteration in the block pixels could cause the adopted (even) parity to be violated. The fragile watermarking is achieved through the modification of least significant bits ((LSBs) of certain frequency coefficients' according to the even parity condition. In spite of this image watermarking operation, the proposed method is efficient. In order to prove the efficiency of our proposed method, it was tested against standard images using measurements like peak signal to noise ratio (PSNR) and structural similarity index (SSIM). Experiments showed promising results; the method preserves high image quality (PSNR≈ 44.4367dB, SSIM≈ 0.9956) and good tamper detection capability.

show abstract

Section: Wavelet Transformmentioning

confidence: 99%

Section: Wavelet Transformmentioning

confidence: 99%