Hand Gesture Recognition Based on Auto-Landmark Localization and Reweighted Genetic Algorithm for Healthcare Muscle Activities

Due to the constantly increasing demand for automatic localization of landmarks in hand gesture recognition, there is a need more sustainable, intelligent, and reliable system recognition. The main purpose this study was develop an accurate recognition that capable error-free auto-landmark any dateable RGB image. In paper, we propose based on landmark extraction from images regardless environment. gestures performed via two methods, namely, fused directional image methods. method produced greater extracted accuracy. proposed system, (HGR) done several different (1) HGR point-based features, which consist (i) distance (ii) angular (iii) geometric features; (2) full are composed SONG mesh geometry active model. To optimize these applied gray wolf optimization. After optimization, reweighted genetic algorithm used classification Experimentation five challenging datasets: Sign Word, Dexter1, Dexter + Object, STB, NYU. Experimental results proved auto with feature technique efficient approach towards developing robust system. were compared Artificial Neural Network (ANN) decision tree. developed plays significant role healthcare muscle exercise.