Age estimation using biometric features among the black race
The estimation of a person’s age based on their face is usually done effortlessly by humans based on their intuition. However, age estimation from facial images has become one of the biggest challenges in the area of face recognition. This challenge is mainly due to the unavailability, underrepresentation and different forms of data that are used in training age estimation models. Data as a relevant component of artificial intelligence is key to the efficiency of age estimation models as well. An underrepresentation of face images belonging to people of the black race has raised concerns on algorithm biasedness and inaccuracies. With the motivation to rule out this biasedness, this study undergoes a data collection of face images of the black race, creating training and testing datasets for the proposed model. A hybrid KNN Classifier and Support Vector Regression (SVR) model approach is presented for age estimation using face images of the black race. The results of another hybrid model is compared to the KNN-SVR model on three datasets. It is concluded that the KNN-SVR fusion produces a highest accuracy on the newly created dataset.
Undergraduate thesis submitted to the Department of Computer Science, Ashesi University, in partial fulfillment of Bachelor of Science degree in / Computer Science, May 2020
age estimation, ethnicity, artificial intelligence