基于无线射频识别的手术器械自适应分类仿真

Adaptive Classification and Simulation of Surgical Instruments Based on Radio Frequency Identification

针对当前无线射频识别的手术器械自适应分类时,普遍存在着分类时间过长、能量消耗过大等问题。提出基于线性组合模型的手术器械自适应分类方法。通过对无线射频识别的手术器械自适应分类进行分析,引用高斯滤波对读写器读取到的手术器械设备标签信号强度进行预处理,减少周围环境因素对信号的干扰,使手术器械设备标签信号强度值与标签实际位置相符,引用粒子群算法找出最优标签信号强度值。将监督预学习和监督微调相结合,构建支持向量机线性组合模型,将手术器械标签的信号强度值输入到支持向量机中,获取手术器械设备标签信号的后验概率,根据后验概率进行手术器械自适应分类。实验结果表明,所提方法分类时间较短、能量消耗较小。

In the current adaptive classification for surgical instrument based on radio frequency identification, classification time is too long and energy consumption is too high. Therefore, an adaptive classification method for surgical instruments based on linear combination model was proposed. By analyzing the adaptive classification of surgical instruments based on radio frequency identification, we used Gaussian filter to preprocess the signal intensity of surgical instrument equipment tag read by the reader, so as to reduce the interference of environmental factors on the signal, so that the signal strength value of surgical instrument equipment label was consistent with the actual position of tag. Moreover, we used the particle swarm algorithm to find the strength value of best tag signal. We combined the supervised pre-learning with the supervised fine tuning to construct the linear combination model of support vector machine. Then, we inputted signal intensity value of surgical instrument label into the support vector machine to obtain the posterior probability of surgical instrument label signal. According to the posterior probability, we performed the adaptive classification of surgical instruments. Simulation results prove that the proposed method has shorter classification time and less energy consumption.