Age Estimation on Human Face Image Using Support Vector Regression and Texture-Based Features

This paper proposed a framework for estimating human age using facial features. These features exploit region information, such as wrinkles on the eye and cheek, which are then represented texture-based feature. Our has several steps: preprocessing, feature extraction, estimation. In this research, three extraction methods their combination performed, Local Binary Pattern (LBP), Phrase Quantization (LPQ), Binarized Statistical Image Feature (BSIF). After extracting feature, Principle Component Analysis (PCA) was performed to reduce size. Finally, Support Vector Regression (SVR) method used predict age. evaluation, estimation error will be based mean average (MAE). experiment, we utilized well-known public dataset, face-age.zip, UTK Face datasets, containing 15,202 image data. The data were divided into training of 12,162 images testing 3,040 images. experiments found that combining BSIF LPQ with PCA achieved lowest MAE 9.766 9.754. results show could image.