الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 86 |  |  |
| | | from | 86 | | |
Abstract
Human faces convey a lot of useful information that can be employed effectively in many aspects in our life. Therefore, many researchers have been attracted to the human facial analysis, research field and huge amounts of research efforts have been done in its research sub-fields, including face recognition, gender recognition, facial expression recognition, future face prediction, face image reconstruction, and facial age estimation.
Recently, the research area of automated facial age estimation has gained an increasing attention from the researchers due to its relevance to several daily life applications such as such, as surveillance monitoring and security control, electronic customer relationship management (ECRM), biometrics, and entertainment. In this work, we have proposed an automated facial age estimation method based on several well-known classifiers famous in the machine learning domain, such as deep belief network, SVM and K-NN which able to estimate the human age based on the face image. The proposed approach consists of four steps: image preprocessing, feature extraction using PCA, feature reduction using PCA, and DBN based classification process. The proposed approach has been evaluated via a dataset that includes Morph II database images. The experimental results have shown that the proposed approach has a promising performance by achieving classification accuracy up to 100% compared with SVM and K-NN.
4.7 Future Work
For future works, different modifications are to be added to the proposed approach such as employing wavelet transform and linear discriminant analysis for feature extraction and nature-inspired algorithms for feature selection and classifier’s parameter optimization.