Jiang-Lung Liu

A robust watermarking scheme using self-reference image

In this paper, a robust watermark scheme for copyright protection is proposed. By modifying the original image in transform domain and embedding a watermark in the difference values between the original image and its reference image, the proposed scheme overcomes the weak robustness problem of embedding a watermark in the spatial domain. Besides, the watermark extraction does not require the original image so it is more practical in real application. The experimental results show that the proposed scheme provides not only good image quality, but is also robust against various attacks, such as JPEG lossy compression, filtering and noise addition.

Fault resilient and compression tolerant digital signature for image authentication

Many watermarking schemes have been proposed to protect the authority of the image producer. These schemes can help the image producer to assert the ownership when their images are illegal reproduced. A buyer may want to make sure that the received commodity is from the legal sender and prove its integrity. In this situation, digital signature schemes can meet these requirements. In this sender-receiver protocol, images are always compressed before transmission. So compression tolerant image-features have to be designed to replace the hash functions in conventional digital signature schemes. In this paper, a public-key based fault resilient and compression tolerant digital signature is proposed. The proposed method can not only verify the authenticity and the integrity of commodities, but also can locate the illegal modifications. Furthermore, the proposed method can provide the capability of recovery if images are tampered.

A copyright protection scheme for digital images using visual cryptography technique

A robust copyright protection scheme for digital image is proposed in the paper. In the proposed scheme, the watermark does not require to be embedded into the protected image but is used to generate a secret image and a public image by using the visual cryptography technique. Then the secret image is registered to certified authority (CA) for further protection. In the step of watermark extraction, the watermark can be acquired by performing exclusive-OR (XOR) operation between the secret image and the public image. The experimental results show that the proposed scheme not only can clearly verify the copyright of the digital image, but also is robust to withstand several image processing attacks such as JPEG lossy compression, cropping, noise adding, sharpening and blurring attacks.

Multiple layer data hiding scheme for medical images

Nowadays, the transmission of digitized medical information has become very convenient due to the generality of Internet. Regardless of the prevention of medical fault, the real-time detection of abnormal event, the support of clinical decision, even the model developing of medical service based on patient, Internet has created the biggest benefit to achieve the goals of promoting patient safety and medicine quality. However, it is easier that the hackers can grab or duplicate the digitized information on the Internet. This will cause following problems of medical security and copyright protection. Therefore, the information hiding techniques are developed for protection of medical information and copyright. This paper proposes a multiple-layer data hiding technique in spatial domain. It utilizes a reduced difference expansion method to embed the bitstream in the least significant bits (LSBs) of the expanded differences. By using the reduced difference expansion method, we can embed a large amount of data in a medical image whose quality can also be maintained. Moreover, the original image can be restored after extracting the hidden data from the stego-image. Experimental results show that the proposed scheme provides higher embedding capacity at the same level image quality compared with Tians difference expansion method.

Steganographic Method for Secure Communications

Cryptographic methods secure an important message by encrypting it to an unrecognized form of data which may arouse the interest of cryptanalysis for part of the recipients. Steganographic methods hide the encrypted message in cover carriers so that it cannot be seen while it is transmitted on public communication channels such as computer network. Many steganogrphic methods embed a large amount of the secret information in the first k LSBs of the pixels of the cover images. Because of the imperfect sensibility of the human visual system, the existence of the embedded secret information can be imperceptible. Unfortunately, the hidden secret information may be discovered by the common-cover-carrier attack if it has not been appropriately disposed. In this paper, an LSB-based steganographic method is proposed to resolve this problem. By using variable-size insertion and redundant Gaussion noise adding, the stego-images created with the proposed method can survive both the human visual system and the common-cover-carrier attack. Moreover, many cryptographic protocols are involved in the proposed method to resolve the problems of security and key management that may be encountered in other steganogrpahic methods. The proposed method is hence suitable for secure communications.

Adaptive digital watermarking using neural network technique

A novel image watermarking approach based on the human visual system (HVS) model and neural network technique is proposed. The human visual system model is utilized to generate the suitable strength of embedded watermark and can be described in terms of four properties of HVS model: entropy, frequency, luminance, and texture sensitivity. The neural network technique has been employed to obtain the local characteristics of image. In our experiments for each different image the watermark can be adjusted to provide a maximum and suitable strength subject to the imperceptibility constraint.

Digital watermarking using multiresolution wavelet transform

A novel technique for digital watermarking of still image based on the concept of wavelet transform is proposed. This algorithm takes advantages of multiresolution signal decomposition to obtain the frequency components of the image. Unlike most previous tree-based wavelet methods that embed a watermark into multiple subbands to enhance the robustness, the proposed method embeds a visually recognizable watermark pattern by modifying the single middle-frequency part of the original image. The experimental results show that the proposed method keeps image fidelity and is robust to common image processing operations, such as JPEG/JPEG 2000 lossy compression, filtering, and incorporating attacks. Some quantitative measures are adopted to objectify performances. Compared with other known watermarking schemes, the proposed method demonstrates its potential for reliable and secure copyright protection technique of multimedia data.

A robust watermarldug scheme using filter and ROI

In this paper, a robust watermarking technique for copyright protection is proposed. The proposed scheme is different from most existing schemes. By using the concept of filter technique to embed the watermark into the R01 (region-of-interest) images, the proposed scheme overcomes the problem that watermark embedded into the spatial domain can be destroyed easily after image processing attacks. The experimental results show that besides the invisibility and security, the scheme is also robust against signal processing attacks.

Robust copyright protection scheme for digital images using the low-band characteristic

A copyright protection technique without using the original im- age in the process of copyright verification is demanded in most appli- cations. In addition, robustness, imperceptibility, security, and unambigu- ity are also essential requirements for a copyright protection technique. In this paper, a robust copyright protection scheme for digital images, based on the low-band characteristic, is proposed to meet these require- ments. Experimental results show that the proposed scheme not only can clearly verify the copyright of the digital images, but also is more robust than existing schemes against several attacks such as JPEG lossy compression, noise adding, sharpening, and blurring.

Adaptive DWT-based watermarking scheme using neighboring mean value relationship

A watermarking algorithm performed in the wavelet domain is proposed. For the purpose of security and robustness, a watermark sequence is embedded by selectively modifying the middle-frequency parts of the image. Human visual system (HVS) is adopted to further ensure the watermark invisibility. The original image is not required for watermark extraction. Simulation results show that the proposed scheme is superior to quantization-based methods in terms of transparency, robustness, and computation complexity.