摘要
针对目前内分泌代谢护理理疗器智能化较低,提出一种基于表情识别的内分泌代谢护理智能理疗器系统,以此通过表情识别的智能交互,自动控制理疗器的脉冲幅值,最终更好地作用于内分泌患者的治疗。首先通过上位机实时识别脸部表情信息,将其传输到下机位的STM32微控制器中,以实现对脉冲幅值大小的调节;为提高表情识别的准确度,以更好地通过患者的表情控制理疗脉冲幅值,对mini-Xception模型进行改进,以提高对患者表情识别的准确率。结果表明:改进后的MC_mini-Xception模型在数据集中进行测试,准确率可达到99%;在系统调试中,通过识别表情信息,在表情信息对理疗波形幅值大小的控制中,融入中位值平均滤波算法,使其可更好控制滤波大小,使得智能理疗器系统在实际的内分泌代谢护理的运用中更加稳定。
In view of the low intelligence of current physiotherapy equipment,an intelligent physiotherapy system based on ex-pression recognition for endocrine and metabolic care is proposed.The system recognizes the facial expression information in real time through the upper computer and transmits it to the STM32 microcontroller of the lower computer to adjust the pulse amplitude.In order to further improve the accuracy of expression recognition and better control the amplitude of physiotherapy pulse,the mini xception al-gorithm is improved.The simulation experiment shows that the improved MC_The mini xception model is tested in the data set,and the accuracy rate can reach 0.990,with a high accuracy rate.In the system debugging,the recognized expression information is clas-sified,and the median value average filtering algorithm is incorporated in the process of controlling the physiotherapy waveform by the expression information,which makes the intelligent physiotherapy instrument system more stable in the actual endocrine and metabolic nursing application.
作者
刘旭斌
LIU Xubin(Chengdu Third People’s Hospital,Chengdu Sichuan 610000,China)
出处
《自动化与仪器仪表》
2023年第6期275-280,共6页
Automation & Instrumentation
基金
四川省卫生健康委科技项目《1,25(OH)2D3介导的PI3K/Akt/mTOR信号通路在糖尿病足创面愈合中的作用研究》(17PJ384)。
关键词
智能交互
表情识别
内分泌护理
理疗器
intelligent interaction
expression recognition
endocrine care
physiotherapy instrument
