Sensitivity analysis of hand movement classification technique using motion templates

This paper presents the sensitivity analysis of a new technique for automated classification of human hand gestures based on Hu moments for robotics applications. It uses view-based approach for representation, and statistical technique for classification. This approach uses a cumulative image-difference technique where the time between the sequences of images is implicitly captured in the representation of action. This results in the construction of temporal history templates (THTs). These THTs are used to compute the 7 Hu image moments that are invariant to scale, rotation, and translation. The recognition criterion is established using K-nearest neighbor (K-NN) Mahalanobis distance. The preliminary experiments show that such a system can classify human hand gestures with a classification accuracy of 92%. This research has been conducted for medical and robotics framework. The overall goal of our research is to test for accuracy of the recognition of hand gestures using this computationally inexpensive way of dimensionality-reduced representation of gestures for its suitability for medical and robotic applications