Improvement of Dynamic Hand Gesture Recognition Based on HMM Algorithm

Abstract

In order to enhance efficiency and accuracy of dynamic hand gesture recognition based on HMM method. we propose a new HMM algorithm combined with the DTW for the computation's high complexity of HMM method in the training stage. The new HMM can establish the relationship of fuzzy closeness degree between the DTW algorithm and the HMM algorithm. Meanwhile, it can combine with the template adaptive strategy, so that improving the recognition accuracy of the algorithm, and enhance the accuracy and timeliness in human-robot interaction.