A novel chaos-based fragile watermarking for image tampering detection and self-recovery

Tong, Xiaojun; Yang, Liu; Zhang, Miao; Chen, Yue

doi:10.1016/j.image.2012.12.003

Cited by 103 publications

(59 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…(1) For the blocks whose 2 LSBs are embedded by 8 bits watermark information, the expectation value of MSE is 1.5 2 , then the PSNR's expectation can be computed by: PSNR = 10 log 10 ( 255 2 1.5 2 ) = 44.61 (dB), (12) (2) For the blocks whose 3 LSBs are embedded by 12 bits watermark information, the expectation value of MSE is 3.5 2 , then the PSNR's expectation is calculated by: PSNR = 10 log 10 ( 255 2 3.5 2 ) = 37.25 (dB), (13) In practical application, if the PSNR value of a processed image is larger than 35 dB, people cannot tell the difference between the original image and its processed version. Table 2 gives the analysis of watermark imperceptibility for the test images shown in Figure 5.…”

Section: Imperceptibility Analysismentioning

confidence: 99%

“…Besides, we compare the presented method with other two self-recovery fragile watermarking methods in [13,25]. Reference [13] is a watermarking scheme with a fixed watermark capacity, and reference [25] is variable capacity watermarking scheme. Figure 6 shows the performance of tamper detection and recovery for text addition attacks.…”

Section: Performance Of Tamper Detection and Self-recoverymentioning

confidence: 99%

“…In addition, the recovered image could be further used for the subsequent image processing. In fragile watermarking scheme with recovery ability, the recovery information for an image block is usually a content representation of the block, like average pixel value [12,13], vector quantization (VQ) index [14], the discrete cosine transform (DCT) coefficients [15,16], etc.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Self-Recovery Fragile Image Watermarking with Variable Watermark Capacity

2018

View full text Add to dashboard Cite

Currently, the watermark capacity of most self-recovery fragile image watermarking schemes is fixed. That means for smooth regions and texture regions, the length of watermark information is always the same. However, it is impractical since more recovery information is needed for the recovery of texture regions. In this paper, a self-recovery fragile image watermarking with variable watermark capacity is proposed. Based on the characteristic of singular value decomposition (SVD), a new block classification method is introduced. The image blocks are classified into smooth blocks and texture blocks. For smooth blocks, the average pixel values are adopted as the recovery information to recover the tampered blocks, while for texture blocks, the quantized and coded DCT coefficients are adopted as the recovery information. After encrypted by binary pseudo-random sequence, the recovery watermark of each block is embedded into its mapping block. In the detection side, the three-level detection mechanism is applied to detect and locate the tampered regions. The experimental results prove that the proposed method achieves good tamper detection results, and the recovered image has better image quality than other self-recovery fragile watermarking methods.

show abstract

Section: Imperceptibility Analysismentioning

confidence: 99%

Section: Performance Of Tamper Detection and Self-recoverymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Self-Recovery Fragile Image Watermarking with Variable Watermark Capacity

2018

View full text Add to dashboard Cite

show abstract

“…A set of minute changes can inadvertently modify it. The key objective of providing sensitivity is established by using stipulated procedures like the ones used in [5][6][7][8][9][10][11]. Authors in ref.…”

Section: Fragile Watermarkingmentioning

confidence: 99%

“…Authors in ref. [5,9] highlight chaotic maps which themselves by property are sensitive to certain initial values manipulated on them, thereby providing fragility. Wei-Che Chen, MingShi Wang in [6], describe a fuzzy c-means clustering, wherein fuzzy constraints defined over certain range of pixel bound to unique clusters initiating sensitivity factor.…”

Section: Fragile Watermarkingmentioning

confidence: 99%

Detection of Image Tampering over Diverse information Security Schemata: A State-of-the-Art

Pande¹,

Patil²,

Bhattacharya³

2014

IJCA

View full text Add to dashboard Cite

Many recent technologies in the field of image processing have necessitated the attention to the field of image forensics. Increase in cyber communication system and availability of advanced digital processing tools, in the past decades has given birth to forgery attempts. Irrespective of various approaches used to protect the Image, proving integrity of the image received in communication is a difficult issue. Under such circumstances, no image can be treated secure against breaches. Moreover, knowledge of the manipulation model is a must for detecting a certain type of tampering.The aim of this paper is to highlight new developments regarding detection of tampering in comparison of various schemata used in the past decades for forgery detection. An assortment of various models used for providing information security to image based on authentication, integrity and confidentiality is presented. Methods of tamper detection have been assessed over the type of attack. An in depth classification of types of image security has been proposed which emphasizes total security issues. The paper puts forward chief developments in schemata of tampering detection. General Terms

show abstract

Image forgery detection in contourlet transform domain based on new chaotic cellular automata

Barani

Ayubi

Jalili

et al. 2015

Security Comm Networks

View full text Add to dashboard Cite

Image authentication includes techniques that detect unauthorized changes in digital pictures. Image authentication is generally categorized into two primary categories, active and passive methods. In this paper, a new active image authentication method is proposed for tamper detection based on contourlet transform. Cellular automata and ‘Game of Life’ are also used to improve the detection of tampered area. A pseudo‐random number generator based on cellular automata has been used to improve security of proposed technique. We evaluates security and randomness of proposed pseudo‐random number generator using by NIST statistical test suite. Experimental results show that the proposed method is sensitive to any tampers, and it can detect forged areas of the images. Visual quality of image after embedding process is desirable due to Peak Signal‐to‐Noise Ratio (PSNR) measure. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

A novel chaos-based fragile watermarking for image tampering detection and self-recovery

Cited by 103 publications

References 22 publications

A Self-Recovery Fragile Image Watermarking with Variable Watermark Capacity

A Self-Recovery Fragile Image Watermarking with Variable Watermark Capacity

Detection of Image Tampering over Diverse information Security Schemata: A State-of-the-Art

Image forgery detection in contourlet transform domain based on new chaotic cellular automata

Contact Info

Product

Resources

About