Half-face based recognition using principal component analysis

<span lang="EN-US">Face recognition is a considerable problem in the field of image processing. It used daily various applications from personal cameras to forensic investigations. Most provides solutions proposed based on full-face images, are slow compute and need more storage. In this paper, we propose an effective way reduce features size database face method thus get increase speed discrimination by using half face. Taking advantage symmetry, first step divide into two halves, then left processed principal component analysis (PCA) algorithm, results compared Euclidian distance distinguish person. The system was trained tested ORL database. found that accuracy reached up 96%, minimized 46% running time decreased 120 msec 70 with 41.6% reduction.</span>