• Journal of Information Processing Systems
• /
• v.11 no.3
• /
• pp.406-420
• /
• 2015

We present a secure and robust image watermarking scheme that uses combined reversible DWT-DCT-SVD transformations to increase integrity, authentication, and confidentiality. The proposed scheme uses two different kinds of watermarking images: a reversible watermark, $W_1$, which is used for verification (ensuring integrity and authentication aspects); and a second one, $W_2$, which is defined by a logo image that provides confidentiality. Our proposed scheme is shown to be robust, while its performances are evaluated with respect to the peak signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR), normalized cross-correlation (NCC), and running time. The robustness of the scheme is also evaluated against different attacks, including a compression attack and Salt & Pepper attack.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.2
• /
• pp.41-48
• /
• 2019

Digital image watermarking techniques have been studied in various fields in order to protect the copyright of multimedia contents mostly composed by images. JPEG compression is a representative copyright attack for image watermarking. The CRT-based watermarking methods in DCT domain have been studied, because these are robust for JPEG compression attack. In this paper, we propose a new method with improved performance compared with the existing CRT based methods in DCT domain. The proposed method selects the DCT block according to the denseness at the watermark insertion phase, and iteratively adjust the CRT Conditional coefficients at the watermark extraction. This paper shows that it is more robust for the JPEG compression attacks through experimental results.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.1
• /
• pp.27-35
• /
• 2022

In this paper we propose white noise adding method to prevent missclassification of deep learning system by adversarial attacks. The proposed method is that adding white noise to input image that is benign or adversarial example. The experimental results are showing that the proposed method is robustness to 3 adversarial attacks such as FGSM attack, BIN attack and CW attack. The recognition accuracies of Resnet model with 18, 34, 50 and 101 layers are enhanced when white noise is added to test data set while it does not affect to classification of benign test dataset. The proposed model is applicable to defense to adversarial attacks and replace to time- consuming and high expensive defense method against adversarial attacks such as adversarial training method and deep learning replacing method.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.3
• /
• pp.341-351
• /
• 2003

Digital image watermarking is the method that can protect the copyright of the image by embedding copyright information, which is called watermark. Watermarking must have robustness to intentional or unintentional data changing, called attack. The conventional watermarking schemes are robust to waveform attacks such as image compression, filtering etc. However, they are vulnerable to geometrical attacks such as rotation, scaling, translation, and cropping. Accordingly, this paper proposes new watermarking scheme that is robust to geometrical attacks by using invariant centroid. Invariant centroid is the gravity center of a central area in a gray scale image that remains unchanged even when the image is attacked by RST including cropping and proposed scheme uses invariant centroids of original and inverted image as the template. To make geometrically invariant domain, template and angle compensated Log -Polar Map(LPM) is used. Then Discrete Cosine Transform(DCT) is performed and the watermark is embedded into the DCT coefficients. Futhermore, to prevent a watermarked image from degrading due to interpolation during coordinate system conversion, only the image of the watermark signal is extracted and added to the original image. Experimental results show that the proposed scheme is especially robust to RST attacks including cropping.

• Asia pacific journal of information systems
• /
• v.31 no.3
• /
• pp.335-357
• /
• 2021

As reproduction of images can be done with ease, copy detection has increasingly become important. In the duplication process, image modifications are likely to occur and some alterations are deliberate and can be viewed as attacks. A wide range of copy detection techniques has been proposed. In our study, content-based copy detection, which basically applies DCT-based features for images, namely, pixel values, edges, texture information and frequency-domain component distribution, is employed. Experiments are carried out to evaluate robustness and sensitivity of DCT-based features from attacks. As different types of DCT-based features hold different pieces of information, how features and attacks are related can be shown in their robustness and sensitivity. Rather than searching for proper features, use of robustness and sensitivity is proposed here to realize how the attacked features have changed when an image attack occurs. The experiments show that, out of ten attacks, the neural networks are able to detect seven attacks namely, Gaussian noise, S&P noise, Gamma correction (high), blurring, resizing (big), compression and rotation with mostly related to their sensitive features.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.2 s.34
• /
• pp.39-47
• /
• 2005

This paper proposes and analyzes the MAC(Message Authentication Code) algorithm that is used for the transition integrity and the entity authentication of message. The MAC algorithm uses the DES algorithm which has 64-bit block and 56-bit key and we compare the security according to 64-bit and 32-bit length of MAC value. Moreover, we use the SEED algorithm which has 128-bit block and 128-bit key and compare the security according to 128-bit and 64-bit length of MAC value. We analyze the security the forgery attack according to length of message and length of MAC value. this paper, a coarse-to-fine optical flow detection method is proposed. Provided that optical flow gives reliable approximation to two-dimensional image motion, it can be used to recover the three-dimensional motion. but usually to get the reliable optical flows are difficult. The proposed algorithm uses Horn's algorithm (or detecting initial optical flow, then Thin Plate Spline is introduced to warp a image frame of the initial optical flow to the next image frame. The optical flow for the warped image frame is again used iteratively until the mean square error between two image sequence frames is lowered. The proposed method is experimented for the real moving Picture image sequence. The proposed algorithm gives dense optical flow vectors.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.3
• /
• pp.59-72
• /
• 2004

In this paper, we classify various attacks that may damage the embedded watermark signal into waveform attack and synchronization attack. And for the specialized defense against the two kinds of attacks we propose a new watermarking scheme named as two layer watermarking method. It consists of a synchronization layer to prepare for the geometrical attack such as rotation and shift and a marking layer 0 embed the watermark signal actually. Namely, the synchronization layer only determines the target region where the watermark signal will be embedded or extracted. And the marking layer spreads the watermark signal over the image as the conventional watermarking methods do. Using the layered structure, the proposed method overcomes the problem that the conventional watermarking methods that do not use the original image at the verification side are most vulnerable to geometrical attacks.

• Journal of Korea Society of Industrial Information Systems
• /
• v.22 no.6
• /
• pp.17-22
• /
• 2017

In this paper, a password recognition system that can overcome a shoulder-surfing attack is developed. During the time period of password insertion, the developed system can prevent the attack and enhance the safety of the password. In order to raise the detection rate of the password image, the mopology technique is utilized. By adapting 4 times of the expansion and dilation, the niose from the binary image of the password is removed. Finally, the mobile phone application is also developed to recognize the one time password and the detection rate is measured. It is shown that the detection rate of 90% is achieved under the dark light condition.

• Journal of Information Processing Systems
• /
• v.9 no.2
• /
• pp.271-286
• /
• 2013

Recently, a pixel-chaotic-shuffling (PCS) method has been proposed by Huang et al. for encrypting color images using multiple chaotic systems like the Henon, the Lorenz, the Chua, and the Rossler systems. All of which have great encryption performance. The authors claimed that their pixel-chaotic-shuffle (PCS) encryption method has high confidential security. However, the security analysis of the PCS method against the chosen-plaintext attack (CPA) and known-plaintext attack (KPA) performed by Solak et al. successfully breaks the PCS encryption scheme without knowing the secret key. In this paper we present an improved shuffling pattern for the plaintext image bits to make the cryptosystem proposed by Huang et al. resistant to chosen-plaintext attack and known-plaintext attack. The modifications in the existing PCS encryption method are proposed to improve its security performance against the potential attacks described above. The Number of Pixel Change Rate (NPCR), Unified Average Changed Intensity (UACI), information entropy, and correlation coefficient analysis are performed to evaluate the statistical performance of the modified PCS method. The simulation analysis reveals that the modified PCS method has better statistical features and is more resistant to attacks than Huang et al.'s PCS method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.2-4
• /
• 2021

In a distributed machine learning system, sharing gradients was considered safe because it did not share original training data. However, recent studies found that malicious attacker could completely restore the original training data from shared gradients. Gradient Leakage Attack is a technique that restoring original training data by exploiting theses vulnerability. In this study, we present the image transformation method based on Discrete Cosine Transform to defend against the Gradient Leakage Attack on the federated learning setting, which training in local devices and sharing gradients to the server. Experiment shows that our image transformation method cannot be completely restored the original data from Gradient Leakage Attack.