一种基于图像特征融合的动态手势识别方法  

A Dynamic Gesture Recognition Method Based on Image Feature Fusion

作　　者：陈茜 李蔚清[1] CHEN Xi;LI Weiqing(School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing 210094)

机构地区：[1]南京理工大学计算机科学与工程学院,南京210094

出　　处：《计算机与数字工程》2023年第6期1381-1386,共6页Computer & Digital Engineering

摘　　要：相较于依赖其他传感器的手势交互,基于视觉的手势交互不需要额外穿戴外部设备、成本较低,成为了最自然的人机交互方式之一。针对增强现实电子沙盘中特定的手势识别要求,论文使用深度相机获取视频帧,利用CNN提取的特征和Hu矩融合,并结合LSTM对动态手势进行识别。实验显示,方法较单特征手势识别方法有更高的准确性,并且在不同光照情况下具有鲁棒性。针对增强现实电子沙盘中特定的九种动态手势,在不同环境下平均实时识别率达到91.57%。Compared with gesture interactions that rely on other sensors,visual-based gesture interactions are one of the most natural ways of human-computer interaction because they do not require the need to wear additional external devices and are less ex-pensive.In view of the specific gesture recognition requirements in augmented reality electronic sand table,this paper uses a depth camera to obtain video frames,combines features extracted by CNN with Hu moment,and identiies dynamic gestures in conjunction with LSTM.Experiments show that the method is more accurate than the single-feature gesture recognition method,and it is robust in different lighting situations.For the nine dynamic gestures in the augmented reality electronic sand table,the average real-time recognition rate has reached 91.57%under different environments.

关 键 词：增强现实 动态手势识别 HU矩 特征融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]