Human age detection is on the most important study while considering the physical features of any human being. The difficulty of age detection task originates from many reasons such as the lack of enough labeled samples to model the aging patterns of subjects, as well as uncontrolled conditions in data collection such as illumination, pose, occlusions and other environmental variables. To eliminate all the issues faced in traditional system, the method of human age classification using filter techniques and wrinkle analysis has been designed. Our system proposes a novel age classification by analyzing human skin detected through skin texture. We extract facial features using wrinkle analysis. By using the facial feature extraction methods, human age is classified into four groups namely infants, young adults, adults, senior adults. The facial features are extracted using Artificial Neural Network & wrinkle analysis using sufficient large amount of database to increase the credibility of results. An attempt is made to eliminate the external factors generated due to photography, hence giving a high accuracy. Our methodology provides a more accurate means of human age classification and eliminates the tedious task of maintaining and managing the facial database which has to be updated at regular intervals. In this methodology we use age estimation classifiers for each age group. Our results indicate that machines can estimate the age of a person almost as the reliably as humans.
Keywords : Wrinkle Features, Artificial Neural Network, Filter Techniques and Wrinkle Analysis.