Quantcast

Decomposition algorithm for depth image of human health posture based on brain health

Research paper by Bowen Luo, Ying Sun, Gongfa Li, Disi Chen, Zhaojie Ju

Indexed on: 23 Mar '19Published on: 23 Mar '19Published in: Neural Computing and Applications



Abstract

At this stage, brain health can be directly expressed in the human hand posture estimation. Therefore, model estimation of healthy human hand posture can also be used as a criterion for brain health. The recognition algorithm of healthy human hand gesture based on global feature extraction of depth map sequence is not enough to analyze the motion correlation of healthy human hand posture, which leads to the need to improve the accuracy of human body hand gesture description and the change of movement speed of robustness. After analyzing the characteristics of healthy human hand movement in detail, this paper proposes a hand posture decomposition algorithm based on depth map sequence. The goal is to find information that plays a key role in hand gesture recognition in the depth map sequence. The algorithm can remove redundant information and improve the robustness of the recognition algorithm.