Abdullah Aziz Laftah Al-Azeerjawi

The paper proposes a new general method for producing a multilevel permutation functioning as an m-tree traversal. It is composed of two basic steps: a random number generator of period length equal m to determine which child to traverse, and recursive permutation in which permutated the subtree if found. The test results proved that the suggested method of permutation is successful depending on the correlation measure.

Permutation is reordering for a set of objects with taking into consideration the important of order of their locations. In order to increase security performance, this paper presents a new method of permutation for sequence of order 2^8. In this propose method, any information can be converted to the sequence of approximate values between 0 and 255 as a set of blocks. It is needed to build the so called lookup table so that it contains the frequency and positions of the original sequence values. In the experiments, twenty-six image files have been experimented as evidence to the proposed method. In addition, the correlation and entropy measures are computed to test the quality of permutation. All the tests observed that method have significantly effective in reducing the correlation and thereby decreasing the perceptual information of the sequence. Hence, the security is improved.

Support vector machine can determine the global finest solutions in many complicated problems and it is widely used for human face classification in the last years. Nevertheless, one of the main limitations of SVM is optimizing the training parameters, especially when SVM used in face recognition domains. Various methodologies are used to deal with this issue, such as PSO, OPSO, AAPSO and AOPSO. Nevertheless, there is a room of advancements in this kind of optimization process. Lately, an improved version of PSO is developed which is called Modified PSO. In this paper, a new technique based on Modified PSO, called (Modified PSO-SVM) is proposed to optimize SVM parameters. The proposed scheme utilizes Modified PSO to seek the finest parameters of SVM Two human face datasets: SCface, CASIAV5 and CMU Multi-PIE face datasets are used in the experiments. Then, a comparison was done with the PSO-SVM, OPSO-SVM and AOPSO-SVM and it showed promising results in terms of accuracy.