基于数据捕捉的舞蹈动作轨迹自动跟踪方法  

Automatic tracking method of dance movement trajectory based on data capture

作　　者：吴红[1,2] WU Hong(Humanities Academy,Xi’an International University,Xi’an 710077,China;Sehan University,Mokpo 526-702,Republic of Korea)

机构地区：[1]西安外事学院人文艺术学院,西安710077 [2]世翰大学,韩国木浦市526-702

出　　处：《信息技术》2023年第12期116-120,129,共6页Information Technology

摘　　要：由于当前已有方法未能对舞蹈视频图像进行滤波处理,导致轨迹跟踪结果不准确。提出一种基于数据捕捉的舞蹈动作轨迹自动跟踪方法。通过小波多尺度滤波方法减弱背景杂波以及高频相干斑噪声,将经过滤波处理的数据集中各个舞蹈视频进行等分,对各个分段视频进行累加边缘特征,同时将全部视频边缘特征累加到同一幅图像中,通过数据捕捉方法提取方向梯度直方图特征。同时将Harris角点检测和模板特征匹配相结合,对舞蹈旋转动作轨迹进行自动跟踪。仿真实验结果表明,所提方法可以快速准确实现轨迹自动跟踪。Since the existing methods fail to filter the dance video images,resulting in the inaccurate trajectory tracking results.An automatic tracking method of dance movement trajectory based on data capture is proposed.The background clutter and high-frequency coherent speckle noise are reduced by the wavelet multi-scale filtering method,and each dance video in the filtered data set is divided into equal parts.The edge feature of each segmented video is accumulated,and the all the edge features are accumulated into one image,and the directional gradient histogram feature is extracted by the data capture method.At the same time,Harris corner detection and template feature matching are combined to automatically track the dance rotation trajectory.Simulation results show that the proposed method can quickly and accurately realize automatic trajectory tracking.

关 键 词：数据捕捉 舞蹈动作 轨迹自动跟踪 图像滤波 HARRIS角点检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]