基于模糊粗糙集的睡姿压力图像识别

Sleep posture pressure image recognition based on fuzzy-rough set theory

睡姿作为评估睡眠质量和疾病预防治疗的指标之一,不仅影响人心理和生理健康的睡眠质量,而且对呼吸暂停综合症、褥疮等疾病诊断意义重大。为实现睡姿压力图像的自动识别,对基于模糊粗糙集算法进行了应用研究,首先采用柔性压力传感器阵列获取静态睡姿压力图像,并对其进行图像预处理以完成基于简单图像几何特征的提取;通过引入模糊粗糙集对图像条件属性离散化约简,剔除冗余信息推导模糊决策规则,进而实现睡姿类别有效分类。实验结果表明,与已有算法相比该算法的分类精确度可达92.9%,具有较好的准确度。

Sleep posture management,as one of the indicators of evaluating sleep quality and preventing pressure ulcer and respiration related diseases,is important for the nursing care of bedridden patients and breath disorder patients.In this paper,an unconstrained fuzzy-rough set algorithm is proposed and applied to recognize sleep posture.Firstly,the geometric features are extracted from the static pressure image obtained from a flexible pressure sensor array.Then,in the fuzzy rough set algorithm,the characteristic values for the recognition are analyzed and the continuous pressure distribution is discretized and converted to fuzzy decision table to identify sleep posture effectively by eliminating redundant information and inferring fuzzy decision rules.Experimental results show that the proposed algorithm can achieve a high accuracy of92.9% for the typical sleep posture recognition,which is much higher than the existing algorithms.