A novel face recognition method based on one state of discrete Hidden Markov Model
Hameed R. Farhan, Mahmuod Hamza Al-Muifraje, Thamir Rashed Saeed · 2017
The trend for about twenty years, the research regarding the number of states in Hidden Markov Model (HMM) was mainly aimed at increasing it in order to ensure the robustness of the face recognition system. In this paper, a novel face recognition method is presented based on one state of discrete HMM, where it seemed impossible in the past. Contrary to other approaches that use the three parameters of the HMM to recognize faces, the proposed work uses only one parameter for discrimination, while the two others are fixed. The method starts with a pre-processing step to remove the influence of noise from the images using the median filter and the discrete wavelet transform (DWT), where at the same time; the DWT possesses a size reduction feature. The reduced size image is prepared to generate a sequence of overlapping blocks. The data manipulated by the model are produced by converting each block to an integer value using some statistical techniques and a quantization process, such that each image is described by a sequence of integers. The experimental results show that the use of one-state model highly reduces the computational complexity of the system and continues in achieving 100% recognition rate, which can be considered a practical breakthrough in the field of face recognition that has superseded the other research efforts in reducing the number of states of HMM, and consequently accomplished reduction of computational complexity and ultimate enhancement of system performance.