Abstract
One of the most popular tool implemented in face recognition issues is Principal Component Analysis (PCA) which is successfully used in machine learning and data analysis. However, if the images are not regular with some factors that affect the image recognition accuracy such as variation of facial expressions, different poses or lighting problems, this technique may show some deficiencies. In this work, different kinds of methods were implemented by combining different preprocessing techniques to evaluate and compare them under different lighting conditions of images. In order to have the same lighting conditions for every image, the methods were applied to them after PCA processing. As a result, the face recognition accuracy was improved by means of implementing the techniques separately or in combination.