Multilevel Blocks and Pixels Colour Values Scrambling Based Shifting Image Technique
Ahmed Bashir Abugharsa, Hamida Almangush · 2014
With the rapid increase in the use of digital images, it is important to protect confidential image data from illegal access. Image scrambling is a useful approach to protect image data by confusing the image which can thus prevent it from being misused. The issues concerning the relationship among the adjacent pixels of the scrambled image, the weakness of visual leakage and produce a similar histogram plot, all affect the image scrambling process. This paper is aimed to at determining the iteration level that is extracted from the secure key and the key mask for building a scrambling table which is used in the scrambling processes and the pixel colour values. The technique begins by dividing the original image that has to be scrambled into blocks whereby each block contains a specific number of pixels. These blocks are then scrambled by shifting the block positions in the image by using the vertical and horizontal scrambling method and then the pixel colour values are modified based on a scrambling table. The experimental results of the paper show that the block scrambling technique has reduced the relationship among the image elements. This is happened because of the entropy value of the scrambled images has increased and the correlation value is lower. Moreover, the scrambling degree of the scrambling technique has larger values for the scrambled image. Subsequently, the pixel colour value in the technique helps to confuse and amend the pixel colour values of the scrambled image to produce a different histogram plot when compare with the original images. The study also proves that a smaller block size yields higher entropy, low correlation, a high scrambling degree and have the ability to restore the scrambled images to their original form.