融合改进RF算法的人体姿态识别方法在现代智能化工程中的应用

Application of human posture recognition method integrated with improved RF algorithm in modern intelligent engineering

摘要:

对人体姿态识别及现代智能化工程设计成为人机交互领域的重要研究方向进行了研究。在实现更高效、智能的人体姿态识别中,采用了基于密度的带有噪声的应用空间聚类与随机森林(Density-Based Spatial Clustering of Applications with Random Forest,DBSCAN-RF)的分类训练器,该方法的技术创新和独特之处在于结合了密度聚类和随机森林的优点,能够有效地处理带有噪声的数据集,并具有较高的计算效率和可扩展性。通过实验测试,DBSCAN-RF算法的识别召回率最高达到了98.64%,相比于传统的RF算法、K-means-RF以及Mean-shift-RF算法,其数值分别增加了6.37%、4.28%、3.95%。同时,DBSCAN-RF算法在跌倒和正常走路的识别召回率分别达到了95.31%和96.48%。此外,DBSCAN-RF算法的测试时间均低于62 ms。经实际应用满足了现代智能化的人体姿态识别工程上的应用,为现代智能化的人体姿态识别提供了可靠的技术支持。

Abstract:

Human body pose recognition and modern intelligent engineering design have become an important research direction in the field of human-computer interaction. In the realization of more efficient and intelligent human posture recognition, using the density based application with noise of spatial clustering and random forest (Density-Based Spatial Clustering of Applications with Random Forest, DBSCAN-RF) classification trainer, the technical innovation and unique method combines the advantages of density clustering and random forest, can effectively deal with noisy data sets, and has high computational efficiency and scalability. Through experimental testing, the recognition recall rate of DBSCAN-RF algorithm reached the highest level of 98.64%, which increased by 6.37%, 4.28%, and 3.95% compared with the traditional RF algorithm, K-means-RF and Mean-shift-RF algorithm. Meanwhile, the recognition recall rate of the DBSCAN-RF algorithm reached 95.31% and 96.48% for falls and normal walking, respectively. Moreover, the test time of the DBSCAN-RF algorithm was all lower than 62ms. The practical application meets the application of modern intelligent body posture recognition engineering and provides reliable technical support for modern intelligent body posture recognition.