Affine invariant image watermarking using intensity probability density-based Harris Laplace detector

Wang, Xiangyang; Niu, Panpan; Yang, Heng; Chen, Lili

doi:10.1016/j.jvcir.2012.05.008

Cited by 19 publications

(7 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the desynchronization attacks caused by the screen-cam process and user operations, we need to develop an appropriate synchronization method to locate the watermark. We test the feasibility of the Harris-Laplace, SIFT, and SURF operators, which are extensively employed to construct local scale-invariant feature regions (LFRs) as message embedding areas [37][38][39][40][41][42][43][44][45], in the screen-cam process. To select the most suitable operators for LFRs construction, the variations of feature point coordinates, feature scale, and feature direction are quantitatively analyzed under different shooting distances.…”

Section: Local Square Feature Region Constructionmentioning

confidence: 99%

Screen-Cam Robust Image Watermarking with Feature-Based Synchronization

et al. 2020

View full text Add to dashboard Cite

The screen-cam process, which is taking pictures of the content displayed on a screen with mobile phones or cameras, is one of the main ways that image information is leaked. However, traditional image watermarking methods are not resilient to screen-cam processes with severe distortion. In this paper, a screen-cam robust watermarking scheme with a feature-based synchronization method is proposed. First, the distortions caused by the screen-cam process are investigated. These distortions can be summarized into the five categories of linear distortion, gamma tweaking, geometric distortion, noise attack, and low-pass filtering attack. Then, a local square feature region (LSFR) construction method based on a Gaussian function, modified Harris–Laplace detector, and speeded-up robust feature (SURF) orientation descriptor is developed for watermark synchronization. Next, the message is repeatedly embedded in each selected LSFR by an improved embedding algorithm, which employs a non-rotating embedding method and a preprocessing method, to modulate the discrete Fourier transform (DFT) coefficients. In the process of watermark detection, we fully utilize the captured information and extract the message based on a local statistical feature. Finally, the experimental results are presented to illustrate the effectiveness of the method against common attacks and screen-cam attacks. Compared to the previous schemes, our scheme has not only good robustness against screen-cam attack, but is also effective against screen-cam with additional common desynchronization attacks.

show abstract

Section: Local Square Feature Region Constructionmentioning

confidence: 99%

Screen-Cam Robust Image Watermarking with Feature-Based Synchronization

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Harris-Laplace detector: A strong keypoint extraction method and the extracted features are invariant to image changes like rotation, scaling and translation. The Harris-Laplace detector first finds a set of images represented at different resolutions for reliable Harris detection, then for detecting keypoints an automatic scale selection procedure is applied to each corner by selecting the scale with a maximal LoG response (Wang, Niu, Yang, & Chen, 2012).…”

Section: Bag-of-features Modelmentioning

confidence: 99%

Bag-of-Feature model for sweet and bitter almond classification

Nasirahmadi

Ashtiani

2017

Biosystems Engineering

View full text Add to dashboard Cite

show abstract

“…The watermark embedding process should be synchronized with the extracting process. The principle of watermark synchronization are geometric correction methods [1][2][3], geometric invariant methods [4][5][6][7] and feature based methods [8][9][10]. Geometric correction methods include image registration and template watermarking schemes.…”

Section: Introductionmentioning

confidence: 99%