An evolutionary programming approach for securing medical images using watermarking scheme in invariant discrete wavelet transformation
Introduction
The advancement in the digital era leads to the utilization of huge multimedia contents in the recent years [[1], [2], [3], [4], [5]]. This development in the digital technology also brings some issues regarding the distribution and unauthorized access of the multimedia content. To overcome these issues, digital watermarking has been introduced. The watermarking procedure is responsible for hiding the information in terms of watermark into the multimedia [59]. In general, inserting of watermarking causes degradation of image quality. This degradation is not visible to the human eye. Furthermore, the watermarked images are affected with distortions caused by active or passive attacks [35]. The watermark procedure must be robust against all the attacks. The robustness in the watermarking denotes the idea of extracting the watermark after the attacks are employed. According to the visibility, domain and permanency, the watermarking procedures are classified in to different types.
As per the visibility is concerned, the digital watermarking is classified into visible watermarking [60] and invisible watermarking [61]. In the visible watermarking procedure, the owner of the multimedia content is easily recognized. The main theme of the visible watermarking procedure is immediate recognition of the owner’s identity in the multimedia content. The major drawback of the visible watermarking is that it can be easily tampered by the attackers using the image processing mechanisms [62]. In the invisible watermarking, the watermark is inserted in the multimedia at unknown places that are unnoticeable. The properties of the inserted watermark are similar to the original images. Therefore, it is difficult to identify the watermark in the original image.
In terms of domain, the watermarked techniques are classified into two types called as spatial domain and frequency domain. The spatial domain procedures are the oldest techniques that consider the modification of pixels in the image to the insertion of watermark. The commonly used approaches are spectrum [1] and least significant bit [2]. The advantages of spatial domain techniques are easy implementation procedure and low computation cost. However, the spatial domain techniques are not robust against all type of attacks. The frequency domain techniques are used to insert the watermark by transforming the original image with the help of discrete wavelet transform [[3], [4]], Ridgelet transform [[5], [6]], discrete cosine transform [[7], [8]], discrete Hadamard transform [[9], [10]], and discrete Fourier transform [[11], [12]]. After that the watermark insertion procedure is applied. These techniques are widely acceptable and more robust against all type of attacks.
While permanency is concern, the invisible watermarking approaches are divided into three types, such as fragile watermarking [[13], [14]], semi-fragile watermarking [[15], [16]] and robust watermarking [[17], [18]]. Fragile watermarking techniques are developed for identification of tampers in the watermarked images. These techniques are guaranteed the authenticity and integrity [65]. Semi-fragile watermarking techniques are the combination of both fragile and robust watermarking approaches. These techniques are robust against some attacks [63] and fragile against other attacks [64]. Robust watermarking techniques are developed to handle all type of attacks that are caused to tamper the watermarked images. The major types of attacks are compression, noise addition, filtering, image compression, geometric transformation and many more. The main theme of these techniques is to identify the origin of the attacks. The robust watermarking techniques are rich in providing the authenticity and verification of the images.
Since 1990, the machine learning approaches and artificial intelligence mechanisms such as neuro computing, fuzzy techniques and evolutionary algorithms are playing predominant role in solving real time problems in the applications. In this category, some of the algorithms [[41], [42], [56], [57]] like Genetic algorithm (GA), Ant Colony Optimization algorithm (ACO), Firefly algorithm and Particle Swarm Optimization algorithm (PSO) made numerous contributions to the watermarking field. In [58], the watermarking scheme was developed based on multiple scaling factors using the GA approach. Particle swarm optimization (PSO) is an efficient and easy optimization mechanism when compared to the other evolutionary approaches. This optimization mechanism is widely used in different domains as well as in watermarking [[19], [20], [21]], Apart from the PSO advantages, it has some drawbacks that the global convergence of the algorithm not satisfactory in all times. It is due to the population diversity evolved from the solution set [22].
To address the issues of multi scaling factors in watermarking, this paper proposed the image watermarking algorithm in wavelet transformation (IDWT) using the singular value decomposition (SVD) and particle swarm optimization (PSO). The improved DWT is applied to the medical image to retrieve the invariant wavelet domain. Then, the low frequency sub-band is selected and divided in to four non-overlapping blocks. The suitable region is selected from the blocks using the entropy for watermark insertion. The watermark is inserted in to the selected region by modifying the values of the coefficients in the image using threshold function. The scaling factors are optimized using the PSO algorithm. The performance of the proposed model is evaluated using the existing schemes similar to the properties of the proposed model. The Normalized Coefficient (NC) and Peak Signal to Noise Ratio (PSNR) are considered to evaluate the similarity between the original image and watermarked image. The attacks are employed to test the robustness of the proposed approach. The novelty of the proposed watermarking approach represented as follows:
- i)
The proposed watermarking scheme utilized improved discrete wavelet transformation (IDWT) referred from Li et al. [23] to retrieve the invariant wavelet domain.
- ii)
The entropy mechanism is used to identify the suitable region for insertion of watermark. This will improve the imperceptibility and robustness of the watermarking procedure.
- iii)
The application of SVD in other domains like DWT, DCT and DFT improves the watermarking procedure. Therefore, SVD is applied to the invariant wavelet domain.
- iv)
The scaling factors such as PSNR and NC are considered for evaluation of the proposed method and the PSO is employed to optimize the scaling factors.
The rest of the paper is organized as follows: Section 2 deals with the related work regarding the different approaches employed medical image watermarking. Section 3 deals with preliminaries used in the paper. Section 4 explains about the proposed watermarking algorithm. Section 5 evaluates the experimental results and finally Section 6 concludes the research work.
Section snippets
Related work
This section deals with the existing watermarking approaches based on singular value decomposition (SVD) along with artificial intelligence based watermarking approaches are explained. In [44], the authors proposed watermarking approach based on the combination of SVD and discrete wavelet transformation (DWT). After the transformation is applied to the host image, the image is divided in to LL, LH, HL, HH sub-bands. The SVD is applied to the original image to replace with the SVD values of the
Preliminaries used in the algorithm
This section deals with the concepts that are used to transform the medical image in to wavelet transformation and then selects the blocks for watermark insertion.
Proposed watermarking procedure
In this section, the explanation about proposed watermarking scheme based on singular value Decomposition (SVD) and Improved DWT (IDWT) are given as follows:
Experimental evaluation
This section deals with the performance evaluation of the proposed medical image watermarking scheme and the comparison is performed with existing schemes proposed by Lin et al. [39] and Singh et al. [40]. For simplification, these methods are called as SVD and Entropy-SVD. To make the proposed method more effective, the algorithms proposed by Mishra et al. [41] and Aslantas et al. [42] are considered for comparison and the mechanisms are named as SVD-Firefly and SVD-GA. Table 1 represents the
Conclusion
In this paper, a robust watermarking approach is developed based on the objectives to improve the imperceptibility and robustness. The watermarking approach considered low frequency blocks to insert the watermark image. The entropy is employed for finding the suitable blocks to insert the watermark. The parameters changing the matrices for insert and exchange are optimized using the PSO algorithm. The algorithm uses invariant wavelet domain which helps in increasing the performance to handle
References (66)
- et al.
Robust watermarking based on DWT and nonnegative matrix factorization
Comput. Electr. Eng.
(2009) - et al.
Perceptually adaptive spread transform image watermarking scheme using Hadamard transform
Inform. Sci.
(2011) - et al.
Improving the performance of DCT-based fragile watermarking using intelligent optimization algorithms
Opt. Commun.
(2009) - et al.
An intelligent watermarking method based on particle swarm optimization
Expert Syst. Appl.
(2011) - et al.
A novel image watermarking in redistributed invariant wavelet domain
J. Syst. Softw.
(2011) - et al.
An optimized watermarking technique based on self-adaptive DE in DWT–SVD transform domain
Signal Process.
(2014) - et al.
On the security of a hybrid watermarking algorithm based on singular value decomposition and Radon transform
AEU–Int. J. Electron. Commun.
(2011) - et al.
A blind dual color images watermarking based on singular value decomposition
Appl. Math. Comput.
(2013) - et al.
A novel SVD and VQ-based image hiding scheme
Pattern Recogn. Lett.
(2001) - et al.
Moment-based local image watermarking via genetic optimization
Appl. Math. Comput.
(2014)
Color image blind watermarking scheme based on QR decomposition
Signal Process.
Optimized gray-scale image watermarking using DWT–SVD and Firefly Algorithm
Expert Syst. Appl.
A new logo watermarking based on redundant fractional wavelet transform
Math. Comput. Modell.
A digital watermarking scheme based on singular value decomposition and tiny genetic algorithm
Digital Signal Process.
An improved SVD-based watermarking scheme using human visual characteristics
Opt. Commun.
Secure spread spectrum watermarking for multimedia
IEEE Trans. Image Process.
Embedding secret information into a dithered multi-level image
Proc. IEEE. Military Commun. Conf.
A new image watermarking algorithm based on wavelet transform
IEEE Can. Conf. Electr. Comput. Eng.
Robust digital water-marking in the Ridgelet domain
LIEEE Signal Process. Lett.
A robust watermark algorithm based on ridgelet transform and fuzzy c-means
Int. Symp. Information Engineering and Electronic Commerce
DCT-domain watermarking techniques for still images: detector performance analysis and a new structure
IEEE Trans. Image Process.
Hidden digital watermarks in images
IEEE Trans. Image Process.
Improved image watermarking scheme using fast Hadamard and discrete wavelet transforms
J. Electron. Imaging
Circularly symmetric watermark embedding in 2-D DFT domain
IEEE Trans. Image Process.
Colour image watermarking using multidimensional fourier transforms
IEEE Trans. Inform. Forens. Secur.
A sharing-based fragile watermarking method for authentication and self-recovery of image tampering
EURASIP J. Adv. Signal Process.
Authentication and recovery for wavelet-based semi fragile watermarking
Opt. Eng.
A semi-fragile watermarking scheme using weighted vote with sieve and emphasis for image authentication
IEICE Trans. Fundam. Electr. Commun. Comput. Sci.
Robust digital watermarking in the Ridgelet domain
IEEE Signal Process. Lett.
Robust image watermarking based on generalized radon transformations
IEEE Trans. Circuits Syst. Video Technol.
Watermark extraction optimization using PSO algorithm
Res. J. Appl. Sci. Eng. Technol.
An intelligent watermarking approach based particle swarm optimization in discrete wavelet domain
Int. J. Comput. Sci. IJCSI
Quantum-behaved particle swarmoptimization with immune operator
Found Intell Syst. Lect. Notes Comput. Sci.
Cited by (73)
A reversible fragile watermarking technique using fourier transform and Fibonacci Q-matrix for medical image authentication
2024, Biomedical Signal Processing and ControlA wavelet based medical image watermarking scheme for secure transmission in telemedicine applications
2022, Microprocessors and MicrosystemsAn enhanced time efficient technique for image watermarking using ant colony optimization and light gradient boosting algorithm
2022, Journal of King Saud University - Computer and Information SciencesCitation Excerpt :Despite these advantages, the major limitation is large memory requirements to store the observational data. Gangadhar et al. (2018) and Makbol et al. (2017), they proposed an image watermarking wavelet transformation method using SVD and particle swarm optimization for addressing multi scaling factors in image watermarking. Both of these approaches are robust and imperceptible against most of the attacks.
An adaptive watermarking for remote sensing images based on maximum entropy and discrete wavelet transformation
2024, KSII Transactions on Internet and Information SystemsAn Extensive Analysis of Digital Image Watermarking Techniques
2024, International Journal of Intelligent Systems and Applications in EngineeringAn Optimized Dual Image Watermarking Scheme based on Redundant DWT and Randomized SVD with Henon Mapping Encryption
2024, Circuits, Systems, and Signal Processing