Smart Healthcare Hand Gesture Recognition using CNN-based Detector and Deep Belief Network

Gesture recognition in dynamic images is challenging computer vision, automation and medical field. Hand gesture tracking between both human must have symmetry real world. With advances sensor technology, numerous researchers recently proposed RGB techniques. In our research paper, we introduce a reliable hand model that accurate despite any complex environment, it can track recognise gestures. Firstly, videos are converted into frames. After light intensity adjustment noise removal, passed through CNN for extraction. Then from the extracted hand, features full hand. Neural gas locomotion thermal mapping to make feature vector. The vector then fuzzy optimiser reduce uncertainties fuzziness. optimised classifier Deep Belief Network (DBW) classification of Egogesture Jester datasets used validation systems. experimental results over demonstrate overall accuracies 90.73% 89.33% respectively. experiments proves system readability suitability with other state arts model.