摘要
For the automatic extraction of human body measures, an interactive algorithm based on PCL point clouds is proposed. The human body information is captured by a single Microsoft Kinect to obtain the point cloud data, and the neighborhood average method is used for denoising. The algorithm is adopted to estimate the human body measures. The method achieves a complete non-contact human-computer interaction, which has the characteristics of convenient operation and fast speed. The experimental results show that not only the algorithm can be applied to different groups of people, but also it improves the dimensional accuracy. The algorithm satisfies the accuracy requirements of GB/T 23698-2009.
- 单位