| Summary: | 碩士 === 銘傳大學 === 資訊管理學系碩士班 === 97 === Least Significant Bit (LSB), a kind of information hiding techniques, can embed a secret into a cover image to prevent the revelation of secrets to malicious people. Since the method is quite simple and it will destroy cover images, the substitution matrix is proposed to improve the image quality of cover images and the security of secrets. Basically, the construction of optimal substitution matrices is intrinsically a combinatorial optimization problem. Nowadays, some optimization methods have been applied to the construction of substitution matrices, such as greedy algorithms and genetic algorithms. Ant Colony Optimization (ACO) Algorithm, which is also a kind of optimization methods, was derived from the observation of real ants’ behavior. Researchers found that real ants use the chemicals called “pheromones” to communicate with each other and form the shortest path connecting a nest of ants and a food source. It has been shown that ACO Algorithm is effective to solve many combinatorial optimization problems and its efficiency is better than that of genetic algorithm and simulated annealing. Therefore, in this thesis, the ACO algorithm will be exploited to construct optimal substitution matrices for efficiency consideration. In addition, we also propose position substitution matrices to further improve the image quality. Thus, instead of using only the color substitution matrix, we combine color and position substitution matrices to conceal secrets in cover images. Experimental results show that the addition of position matrices is effective in improving the image quality of cover images.
|
|---|
