不同采样密度下体压分布特征

Body pressure distribution characteristics in different sampling densities

对压力座垫原始数据(采样密度为32×32)进行空域滤波处理剔除噪声,增强独立传感器之间数据分布的连续性.将体压分布采样密度降至24×24、16×16、8×8,提取不同采样密度下的平均压力、峰值压力、平均压力梯度和峰值压力梯度进行分析.单因素方差分析结果表明,32×32采样密度与24×24、16×16采样密度下的特征均值之间的差异较小(1.1、2.6 mmHg),与8×8采样密度下的特征均值之间的差异较大(9.0 mmHg).斯皮尔曼相关性分析结果表明,32×32与24×24、16×16、8×8采样密度下的平均压力、峰值压力、平均压力梯度和峰值压力梯度具有较高的相关性(P<0.05).其中相关性最高的为与24×24采样密度下的峰值压力(0.99,P<0.05),相关性最低的为与8×8采样密度下的平均压力梯度(0.55,P<0.05).试验结果表明采样密度为24×24和16×16的压力座垫可以提供精确的体压分布特征.

The raw data with sampling density of 32×32 was spatially filtered to eliminate noise,in order to enhance the continuity of data distribution between independent sensors.Then sampling density of data was decreased to24×24,16×16,and 8×8,respectively.Four common features(mean pressure,maximum pressure,mean pressure gradient,and maximum pressure gradient)were calculated at each sampling density.The one-way ANOVA analysis showed that the differences of mean values between 32×32 sampling density and 24×24 as well as 16×16 sampling densities were small(1.1 mmHg,2.6 mmHg),but the difference of mean value between 32×32 and 8×8 sampling densities was big(9.0 mmHg).Spearman correlation analysis revealed that the four common features of 32×32sampling density had high correlation with that of 24×24,16×16,and 8×8 sampling densities(P<0.05).The highest was the peak pressure correlation(0.99,P<0.05)between the 32×32 and 24×24 sampling densities,and the lowest was the mean pressure gradient correlation(0.55,P<0.05)between the 32×32 and 8×8 sampling densities.The test results showed that the pressure mat with the sampling density of 24×24 and 16×16 can provide accurate body pressure distribution characteristics.