摘要
研究了汽车车位的地磁法检测。针对车辆在长时间停放下出现了基线漂移的现象,致使传统的地磁检测算法在车辆长时间停放时容易出现漏检、误检等情况,提出了一种基于径向基函数(RBF)神经网络的地磁车位检测优化算法,该算法对长时间停车状态下的基线进行补偿,以获得更加准确的基线值从而提高检测的精度。实验结果表明,通过该算法得到的基线值能快速逼近真实值,对基线漂移有较好的补偿效果,经过RBF神经网络对基线进行补偿后,漏检率降低了6.65%,准确率提高了7.31%。
Cars＇ geomagnetic parking place detection is studied. Considering that the baseline drift occurs under cars＇ long parking,thus the traditional geomagnetic detection algorithm is prone to missed detection and misleading during long parking,an algorithm for geomagnetic parking detection based on radial basis function（ RBF） neural networks is proposed. The algorithm compensates the baseline under long time parking condition to obtain more accurate baseline values to improve the detection accuracy. The experimental results show that the baseline value obtained by this algorithm quickly approximates the real value,and the compensation effect on the baseline drift is improved. After compensating the baseline by RBF neural networks,the false detection rate is reduced by 6. 65% and the accuracy rate is improved by 7. 31%.
作者
顾夫挺
郭海锋
何德峰
彭明洋
Gu Futing;Guo Haifeng;He Defeng;Peng Mingyang(College of Information and Engineering,Zhejiang University of Technology, Hangzhou 310023)
出处
《高技术通讯》
EI
CAS
北大核心
2018年第3期227-232,共6页
Chinese High Technology Letters
基金
国家自然科学基金(61374111)
浙江省自然科学基金(LY14F030012)
浙江省教育科学规划(2016SCG241)资助项目
关键词
地磁检测算法
基线漂移
径向基函数(RBF)
车位检测
geomagnetic detection algorithm
baseline drift
radial basis function （RBF）
parking place detection
