| 陈 瑶,张云伟,雷金辉,田泽薇,黎 丽.基于视觉的四足动物骨架及行走步态特征提取方法[J].电子测量与仪器学报,2022,36(2):68-77 |
| 基于视觉的四足动物骨架及行走步态特征提取方法 |
| Method for skeleton and gait parameters extraction ofquadrupeds walking based on vision |
| |
| DOI： |
| 中文关键词: 计算机视觉 深度学习 骨架提取 运动特征 |
| 英文关键词:computer vision deep learning skeleton extraction motion features |
| 基金项目:国家自然科学基金（51365019）项目资助 |
|
|
| Author | Institution |
| Chen Yao | 1. College of Information Engineering and Automation, Kunming University of Science and Technology |
| Zhang Yunwei | 1. College of Information Engineering and Automation, Kunming University of Science and Technology,2. Yunnan Key Laboratory of Artificial Intelligence, Kunming University of Science and Technology,3. Yunnan Key Laboratory of Computer Technology Application, Kunming University of
Science and Technology |
| Lei Jinhui | 1. College of Information Engineering and Automation, Kunming University of Science and Technology,3. Yunnan Key Laboratory of Computer Technology Application, Kunming University of Science and Technology |
| Tian Zewei | 1. College of Information Engineering and Automation, Kunming University of Science and Technology |
| Li Li | 1. College of Information Engineering and Automation, Kunming University of Science and Technology |
|
| 摘要点击次数: 1005 |
| 全文下载次数: 1209 |
| 中文摘要: |
| 为解决四足动物行走期间的骨架信息及步态特征提取问题,基于 HRnet 深度神经网络,通过上下文信息增强与多尺度
信息融合构建了四足动物骨架的提取模型。 在此基础上,建立了基于骨架信息对步态特征进行定量分析的方法。 通过在测试
数据集上对该模型的有效性进行验证,实验结果表明,该模型具有较好的精度,对四足动物关键点估计的平均相似度为
81. 04%,准确率为 92. 77%,召回率为 92. 75%。 基于骨架提取模型,以水牛、羊驼为实验对象,对其行走时的步频特征进行分析
计算,实验结果与人工统计结果相比,最大相对误差为 2. 73%。 通过对一个完整步态周期中水牛和羊驼的髋关节和膝关节角度
变化规律进行分析,提取了四足动物行走过程中的关节运动逻辑以及步态顺序。 最后,以犀牛为实验对象,验证了方法对拍摄
角度的变化具有一定适应性。 研究结果可为四足动物运动信息的智能感知提供参考。 |
| 英文摘要: |
| Due to the need of extracting skeleton information and gait parameters for walking quadruped, a skeleton extraction model of
quadruped is proposed by using context information enhancement and multi-scale information fusion based on HRnet, and a quantitative
analysis method for gait parameters is established. The validity of the model is verified on the test data set of images. Experiments show
that, in the key point estimation of quadruped, the model achieves good performance with the mean similarity of 81. 04%, the accuracy
of 92. 77%, and the recall rate of 92. 75%. Based on the skeleton extraction model, the frequency of buffalo and alpaca were analyzed
and calculated. Compared with the manual statistical results, the maximum relative error was 2. 73%. Through analyzing the angle
variations of buffalo’ s and alpaca’ s hip and knee joints during a complete gait cycle respectively, the joint motion logic and gait
sequence of walking quadruped can be extracted automatically. Finally, taking the rhinoceros as a sample, it is demonstrated that the
proposed method can work adaptively in a range of different shooting angle of images. The results can provide a reference for intelligent
perception of quadruped motion information. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
