Vision Robot
Human Age-Groups Classification Using Appearance Images
There are many modern applications require the function of age classification, such as surveillance monitoring, forensic art, and cigarette vending machines. We propose a method to classify human age using appearance images and apply it to the human-robot interactions. After image preprocessing, we use Support Vector Machine (SVM), a machine learning algorithm, to train large database.
Our image database is from FG-NET and MORPH databases so that it has high degree of complexity and difficulty in recognition. To have higher recognition rate, we train RBF (radial basis function) and linear kernel models at the same time, and decide the final results by F1-decision method. The recognition results reach 95% for female and 97% for male of the proposed system. For the purpose of human-machine interaction, a user interface is designed to perform online age-group classification. The system could be applied on any computer or robot as long as it has a camera sensor. There are many modern applications require the function of age classification, such as surveillance monitoring, forensic art, and cigarette vending machines. This prospers some business such as advertising. By using this classification system, surveillance monitoring can count the age distribution of customer. Vending machines also analyze what is the most popular product in some age groups.