Developing a real-time hand-gesture recognition technique for wheelchair control

Abstract

At present, significant number of people in the world are having motor disabilities. They need to use wheelchair for performing regular movements and activities. However, there are a lot of issues and challenges in using the conventional wheelchairs. Safe navigation, independent mobility and low cost are the key issues for wheelchair users. This paper presents a new mathematical model for an efficient and real-time hand gesture recognition technique for a smart wheelchair control system, in line with contemporary technologies. The model is developed using the positions of the significant hand landmarks, distances among them and some critical thresholds which are determined by good number of samples of hands. The proposed method offers more users’ flexibility through less hand movement in operation of the wheelchair. The model is tested for hands of different sizes, irrespective of any background in indoor or outdoor under sunlight. The experimental study demonstrates that it outperforms the existing methods in terms of success rate with good performance metrics.