Sign Language Classification using Boosted Cascade of Classifiers
Hand gesture recognition is an important component in applications such as human computer interaction, robot control, and disable people assistance systems in which performance and robustness are the primary requirements. In this paper, we propose a hand gesture classification system able to efficiently recognize 24 basic signs of American Sign Language. In this system, computational performance is achieved though the use of a boosted cascade of classifiers that are trained by AdaBoost and informative Haar wavelet features. A new type of feature to adapt to complex representation of hand gesture is also proposed. Experimental results show that the proposed approach is promising.