Liu Shaopeng, Gao Robert X, John Dinesh, Staudenmayer John, Freedson Patty S
Electromechanical Systems Laboratory, Department of Mechanical Engineering, University of Connecticut, Storrs, CT 06269, USA.
Annu Int Conf IEEE Eng Med Biol Soc. 2011;2011:3188-91. doi: 10.1109/IEMBS.2011.6090868.
This paper presents a sensor fusion method for assessing physical activity (PA) of human subjects, based on the support vector machines (SVMs). Specifically, acceleration and ventilation measured by a wearable multi-sensor device on 50 test subjects performing 13 types of activities of varying intensities are analyzed, from which the activity types and related energy expenditures are derived. The result shows that the method correctly recognized the 13 activity types 84.7% of the time, which is 26% higher than using a hip accelerometer alone. Also, the method predicted the associated energy expenditure with a root mean square error of 0.43 METs, 43% lower than using a hip accelerometer alone. Furthermore, the fusion method was effective in reducing the subject-to-subject variability (standard deviation of recognition accuracies across subjects) in activity recognition, especially when data from the ventilation sensor was added to the fusion model. These results demonstrate that the multi-sensor fusion technique presented is more effective in assessing activities of varying intensities than the traditional accelerometer-alone based methods.
本文提出了一种基于支持向量机（SVM）的用于评估人类受试者身体活动（PA）的传感器融合方法。具体而言，对由可穿戴多传感器设备测量的50名测试受试者在进行13种不同强度活动时的加速度和通气情况进行了分析，并由此得出活动类型和相关能量消耗。结果表明，该方法在84.7%的时间内正确识别了13种活动类型，比单独使用髋部加速度计高出26%。此外，该方法预测相关能量消耗的均方根误差为0.43代谢当量，比单独使用髋部加速度计低43%。此外，融合方法在减少活动识别中受试者之间的变异性（各受试者识别准确率的标准差）方面是有效的，尤其是当将来自通气传感器的数据添加到融合模型中时。这些结果表明，所提出的多传感器融合技术在评估不同强度活动方面比传统的仅基于加速度计的方法更有效。
