摘要
为了辅助裁判对健美操难度动作给出正确的分数,设计了基于大数据和动作识别算法的健美操辅助评审系统。首先,文中通过扩充标准动作数据建立了评分比对标准数据库。另外,进一步给出了系统架构,并给出了关键的基于3D数据的采集模块设计。文中将通过傅里叶金字塔过滤的深度特征与骨骼特征进行融合,对融合的数据进行基于支持向量机的数据分类,从而实现了动作识别模块的设计。系统测试结果表明,文中的设计具有较为精确的识别结果,为裁判进行打分提供了有力的工具。
In order to help the referee give the correct score for the aerobics difficulty action,this paper designs aerobics assistant evaluation system based on big data and motion recognition algorithm. First, this paper establishes a standard database of scoring comparisons by extending standard action data. In addition,the system architecture is further given,and the key 3D data-based acquisition module design is given. At the same time,this paper integrates the depth features and the bone features filtered by the Fourier pyramid,and classifies the data of the fusion data based on the support vector machine,thus realizing the design of the motion recognition module. The system test results show that the design of this paper has more accurate identification results and provides a powerful tool for scoring the judges.
作者
吕默
万连城
LV Mo;WAN Lian-cheng(Xi'an Physical Education University,Xi'an 710068,China;Xidian University,Xi'an 710071, China)
出处
《电子设计工程》
2019年第16期6-10,共5页
Electronic Design Engineering
基金
国家自然科学基金(61701380)
关键词
大数据
深度特征
骨骼特征
支持向量机
big data
depth feature
skeletal feature
support vector machines
