The method of discriminating the man’s age was proposed by using edge information and NN. The rate of an average accurate classification is 61.5%, which shows the classification of each age by using NN.
色情報をベースに,顔画像から輪郭やしわなどのエッジ情報を抽出し,その情報を基に年代推定のコンピュータシミュレーションによる実験を行い,その有効性を述べた.
共同発表者:Miyoko Nakano, Fumiko Yasukata, Minoru Fukumi