论文部分内容阅读
通过对人体运动点的三维跟踪识别,结合人体的运动特征实现人体运动信息提取和图形重构,分析运动人体的规律性特征,提高运动训练水平,提出一种基于三维散乱点特征融合提取和图像重构的人体运动点激光跟踪技术。根据人体运动特征部位,采用激光三维扫描方法进行人体运动过程建模和图像采集,对采集的人体运动图像进行运动特征点标定和排列,提取人体骨架的三维散乱运动点集并表征为一个高维向量,形成动作向量库,对动作向量库中的运动点进行图像融合处理,实现运动图像重构和运动动作跟踪识别。仿真结果表明,采用该方法进行人体运动三维重构和跟踪识别的实时性和准确性较好,对人体运动点的三维重构和跟踪识别具有较强的自适应能力。
Through the three-dimensional tracking of human movement point recognition, combined with human movement characteristics of human motion information extraction and graphics reconstruction, analysis of the regularity of the human body movement characteristics and improve the level of exercise training, a three-dimensional scattered point features based on fusion extraction and image Reconstruction of the human body movement point laser tracking technology. According to the parts of the human body movement, laser three-dimensional scanning method was used to model the movement process of the human body and image acquisition. The moving human body movement images were calibrated and arranged for the movement feature points. The three-dimensional scattered movement point set of the human skeleton was extracted and characterized as a high- Vector to form a motion vector library, the motion vector library motion points for image fusion processing, to achieve the reconstruction of the moving image and motion tracking recognition. The simulation results show that the proposed method has good real-time performance and accuracy in three-dimensional reconstruction and tracking recognition of human movement, and has strong self-adaptability to the three-dimensional reconstruction and tracking recognition of human motion points.