Laurent Nana

A New Approach against Color Attacks of Watermarked Images

We present a new approach for modeling color attacks of RGB-color watermarked images. We have used a based interpolation watermarking algorithm and supposed that the attacks are simulated by a scaling of the colors followed by a translation. We give bounds for the extracted watermark depending on the original image, the watermark and the attack. Different attacks like LSB, embedding another watermark, Stirmark have been simulated and the quality of the extracted watermark is discussed in each case.

A Novel Cryptographic Encryption Technique for Securing Digital Images in the Cloud Using AES and RGB Pixel Displacement

In todays cyberspace, Information security plays an active role in ensuring confidentiality, integrity, and authenticity of data transmission or storage. One of the foremost fears with the rapid increment of adoption of data storage in the cloud environment is data security. Media contents such as images and video recordings needed to be highly encrypted without any loss of information during the encryption and the decryption process. Contents such as medical images, surveillance images, personal photos, etc needed to be well protected. This paper proposes a hybrid approach of encryption of images using an Advanced Encryption Standard (AES) and an RGB cryptographic technique. The Method made use of AES for the generation of a shared secret key which is then used by the visual cryptographic algorithm for the encryption process. The encryption is done solely based on RGB pixel shuffling and displacement. At the end of the process, there was no pixel expansion hence there was no loss in image quality. The programming and simulation of the processes as well as the analysis of the results were done using MATLAB.

Feature Based Encryption Technique for Securing Forensic Biometric Image Data Using AES and Visual Cryptography

With the current emergence of biometric image data applications on devices such as smart phones, security cameras, personal computers etc, there is a need for securing the image templates obtained from crime scenes as well as such devices before storing them in locations such as the cloud etc. In this paper, we proposed an encryption technique of securing the biometric image data collected from devices with an approach of feature based encryption technique. The biometric image data was encrypted using advanced encryption process and visual cryptography method. The method engaged the encrypted feature extracted from the plain image in the encryption process and used in to encrypt the image based on a visual cryptographic technique. Analysis of the plain and ciphered image was done and there was no pixel expansion in the process. The implementation was done using MATLAB.

A Hybrid Encryption Technique for Securing Biometric Image Data Based on Feistel Network and RGB Pixel Displacement

Biometric data in a form of images collected from biometric devices and surveillance devices needed to be protected during storage and transmission. Due to the Nature of biometric data, information of the evidence or content of the images need to be preserve after encryption of the plain image and the decryption of the ciphered image. This has to be achieved with a good level of hardness encryption algorithm. Hence this work has proposed a hybrid encryption technique for securing biometric image data based on Feistel Network and RGB pixel displacement. The implementation was done using MATLAB.

A robust associative watermarking technique based on frequent pattern mining and texture analysis

Nowadays, the principle of image mining plays a vital role in various areas of our life, where numerous frameworks based on image mining are proposed for object recognition, object tracking, sensing images and medical image diagnosis. Nevertheless, the research in the image authentication based on image mining is still confined. Therefore, this paper comes to present an efficient engagement between the frequent pattern mining and digital watermarking to contribute significantly in the authentication of images transmitted via public networks. The proposed framework exploits some robust features of image to extract the frequent patterns in the image data. The maximal relevant patterns are used to discriminate between the textured and smooth blocks within the image, where the texture blocks are more appropriate to embed the secret data than smooth blocks. The experiments result proves the efficiency of the proposed framework in terms of stabilization and robustness against different kind of attacks. The results are interesting and remarkable to preserve the image authentication.

JPEG bitstream based integrity with lightweight complexity of medical image in WMSNS environment

The study aims to preserve integrity of JPEG bitstream transmitted in Wireless Multimedia Sensor Networks (WMSNs) environment, lightweight complex process that is proposed to parse bitstream of encoded medical image and formulated the encryption process for that represented bit of Diff value in DC coefficients rather than appended bits of ACs coefficients. Our choice deals with low down complexity by encrypting limited numbers of DC coefficients versus ACs coefficients per block, or even all image blocks. This model may be utilized in building a secure framework of remote image analysis and archiving center for that transmitted medical images from body sensors and clinician, or even other remote imaging center with high integrity and low complexity. The experiments result shows that the proposed approach gives an interesting and remarkable result to preserve the medical image integrity. The obtained results are discussed in details.

A Security Technique for Authentication and Security of Medical Images in Health Information Systems

Medical images stored in health information systems, cloud or other systems are of key importance. Privacy and security needs to be guaranteed for such images through encryption and authentication processes. Encrypted and watermarked images in this domain needed to be reversible so that the plain image operated on in the encryption and watermarking process can be fully recoverable due to the sensitivity of the data conveyed in medical images. In this paper, we proposed a fully recoverable encrypted and watermarked image processing technique for the security of medical images in health information systems. The approach is used to authenticate and secure the medical images. Our results showed to be very effective and reliable for fully recoverable images.

A Novel Hybrid Discrete Cosine Transformation and Visual Cryptographic Technique for Securing Digital Images

Security of digital images in todays cyber is a major concern. Transmission of compressed and uncompressed multimedia data is necessary for communication between different devices. Devices of different screen features and resolution need very efficient and fast compression techniques in order to maintain visual features of transmitted media. Robustness is of a key consideration and importance in image compression techniques. Full recovery of images after compression is a challenge for image compression algorithms. It is practically closely impossible for most encrypted-compressed-images or compressed-encryptedimages to be efficiently and fully reconstructed. In this paper we proposed a robust, efficient and fully recoverable encrypted-compressed image based on hybrid discrete cosine transformation and visual cryptographic technique for securing digital images. We first encrypted the image without pixel loss and then compressed the image which resulted into pixel loss during the compression phase. We decrypted the compressed image by obtaining perfect visuals but losses in pixel values. The analysis of the process proved to be an efficient way of encrypting images and compressing them for other channels with less data capacity without the worry of visual loss as well as ensuring security of the data during transmission. The programming and implementation was done using MATLAB.

Feature Based Encryption Technique for Securing Digital Image Data Based on FCA-Image Attributes and Visual Cryptography

Lossless pixel value encrypted images still maintains the some properties of their respective original plain images. Most of these cryptographic approaches consist of visual cryptographic techniques and pixel displacement approaches. These methods of cryptography are useful in cases as medical image security where pixel expansion is avoided in both the encryption and decryption processes. In this paper we propose a hybrid cryptographic encryption approach by using features generated from digital images based on Galois lattice theory and a visual cryptographic technique based on RGB pixel displacement. The features extracted from a plain image and a lattice was generated which was then used to generate a key used to encrypt the plain image. At the end of the process, there was no pixel expansion and the arithmetic mean, the entropy as well as the Galois lattice of both ciphered and plain image remained the same. The features extracted from the plain image were the same as that of the ciphered image irrespective of pixel displacement that occurred, this makes our approach a suit-able basis for image encryption and storage as well as encrypted image indexing and searching based on pixel values. The implementation was done using Galicia, Lattice Miner and MATLAB..

A Cryptographic and Watermarking Encryption Technique for Securing and Authentication of Digital Images

Identity management is crucial in cyberspace where infringement of one’s privacy, copy right disputes, theft and other malicious activities are very rampant. There exists a need of ownership identification and authentication of digital images. In solving and contributing to this field, we proposed a cryptographic and watermarking encryption technique for securing and authentication of digital images for identity management. In our approach, we encrypt the data to be embedded into the image before embedding it to the image. We make sure we select a random position in the image based on an embedding key. We make sure the message to be embedded was converted into values that falls in between 0255. This is to avoid it being detected using forensic or steganalytic software. General Terms Security, Cryptography, watermarking, image processing