Title :
An experimental comparison of preprocessing methods for age classification
Author :
Yoon, Youngwoo ; Yoon, Hosub ; Lee, Jaeyeon
Author_Institution :
Robot Res. Div., Electron. & Telecommun. Res. Inst. (ETRI), Daejeon, South Korea
fDate :
Sept. 27 2009-Oct. 2 2009
Abstract :
This paper investigates image transformations as preprocessing steps that can be applied toward a state-of-the-art age classification framework of manifold learning. We report on the experimental results of four different preprocessing methods in terms of classification accuracy using a large training and test face database. The use of histograms of oriented gradient (HOG) descriptors increases the classification accuracy from 46 to 75%. Robustness to the rotation and translation is also tested.
Keywords :
face recognition; gradient methods; image classification; learning (artificial intelligence); statistical analysis; HOG; age classification; experimental comparison; histogram of oriented gradient descriptor; image rotation; image transformation; image translation; manifold learning; preprocessing method; test face database; Cognition; Face detection; Face recognition; Histograms; Human robot interaction; Image databases; Machine vision; Robustness; Service robots; Testing;
Conference_Titel :
Robot and Human Interactive Communication, 2009. RO-MAN 2009. The 18th IEEE International Symposium on
Conference_Location :
Toyama
Print_ISBN :
978-1-4244-5081-7
Electronic_ISBN :
1944-9445
DOI :
10.1109/ROMAN.2009.5326053