自然场景下交通标志立柱材料防腐性检测仿真
|
摘要
交通标志是保证交通安全的重要部分。针对传统交通标志立柱材料防腐性检测方法存在检测结果不准确、测时间过长等问题,提出基于GMM算法的交通标志立柱材料防腐性检测方法。根据对两种信号进行连续小波变换,计算小波模极大值以及信号的李普希兹指数,通过对信号状态进行识别,将得到的均值以及方差作为特征值。利用高斯分量个数的高斯混合模型对特征值分布进行较为准确的拟合分布,再利用朴素贝叶斯分类器对材料防腐性进行初步分类,通过分布概率综合评判自然场景下交通标志立柱材料防腐性能,完成检测。仿真结果证明,所提方法有效提高了检测准确度,减少了检测时间,提高了检测效率。
The traffic sign is an important part to ensure traffic safety. The traditional method to detect the corrosion resistance of column material of traffic sign is not accurate. Therefore, this paper presents a method to detect the corrosion resistance of column material of traffic sign based on GMM algorithm. According to the continuous wavelet transform for two kinds of signals, wavelet modulus maxima and Lipschitz exponent of signal were calculated. Meanwhile, the signal state was identified and the mean value and variance were obtained as feature values. In addition, Gaussian mixture model with Gaussian component was applied to accurate fitting distribution of feature values. Then, Naive Bayesian classifier was used to preliminarily classify the corrosion resistance of material. Finally, the distribution probability was used to comprehensively evaluate the corrosion resistance of traffic sign column material in natural scene and thus to complete the detection. Simulation results show that the proposed method can effectively improve the detection accuracy and reduce the detection time. Meanwhile, the detection efficiency is improved.
The traffic sign is an important part to ensure traffic safety. The traditional method to detect the corrosion resistance of column material of traffic sign is not accurate. Therefore, this paper presents a method to detect the corrosion resistance of column material of traffic sign based on GMM algorithm. According to the continuous wavelet transform for two kinds of signals, wavelet modulus maxima and Lipschitz exponent of signal were calculated. Meanwhile, the signal state was identified and the mean value and variance were obtained as feature values. In addition, Gaussian mixture model with Gaussian component was applied to accurate fitting distribution of feature values. Then, Naive Bayesian classifier was used to preliminarily classify the corrosion resistance of material. Finally, the distribution probability was used to comprehensively evaluate the corrosion resistance of traffic sign column material in natural scene and thus to complete the detection. Simulation results show that the proposed method can effectively improve the detection accuracy and reduce the detection time. Meanwhile, the detection efficiency is improved.
引文
[1] 章振华,等. 聚苯胺复合材料防腐性能研究进展[J]. 腐蚀科学与防护技术, 2017,29(1): 73-79.
[2] 高晓辉,等. 水分散型SiO2@PANI防腐材料的制备及性能[J]. 精细化工, 2017,34(4): 375-381.
[3] 郭兴魁,葛圣松. 纳米材料在水性防腐蚀涂料中的应用研究进展[J]. 涂料工业, 2017,47(8):83-87..
[4] 李政,等. 基于卷积神经网络的空心村高分影像建筑物检测方法[J]. 农业机械学报, 2017,48(9):160-165.
[5] 付建平,陈向东,陈明.基于湿度、水位和应变传感器的隧道突水模拟监测系统[J].电子设计工程, 2018,26(10):13-16.
[6] 郑亚光,潘久辉. 一种基于滑动分块的重复数据检测算法[J]. 计算机工程, 2016,42(2): 38-44.
[7] 李杨,等. UPDHES电泳涂层材料的制备及其防腐性能[J]. 表面技术, 2017,46(7): 161-167.
[8] 张兴田. 核电厂设备典型腐蚀损伤及其防护技术[J]. 腐蚀与防护, 2016,37(7): 527-533.
[9] 房亚楠,等. 氟碳涂料在防腐领域的研发现状和发展趋势[J]. 中国腐蚀与防护学报, 2016,36(2):97-106.
[10] 潘金山,郭秀云,谢恺. 高速铁路行车指挥系统信息检测仿真研究[J]. 计算机仿真, 2017, 34(5):168-171.
[2] 高晓辉,等. 水分散型SiO2@PANI防腐材料的制备及性能[J]. 精细化工, 2017,34(4): 375-381.
[3] 郭兴魁,葛圣松. 纳米材料在水性防腐蚀涂料中的应用研究进展[J]. 涂料工业, 2017,47(8):83-87..
[4] 李政,等. 基于卷积神经网络的空心村高分影像建筑物检测方法[J]. 农业机械学报, 2017,48(9):160-165.
[5] 付建平,陈向东,陈明.基于湿度、水位和应变传感器的隧道突水模拟监测系统[J].电子设计工程, 2018,26(10):13-16.
[6] 郑亚光,潘久辉. 一种基于滑动分块的重复数据检测算法[J]. 计算机工程, 2016,42(2): 38-44.
[7] 李杨,等. UPDHES电泳涂层材料的制备及其防腐性能[J]. 表面技术, 2017,46(7): 161-167.
[8] 张兴田. 核电厂设备典型腐蚀损伤及其防护技术[J]. 腐蚀与防护, 2016,37(7): 527-533.
[9] 房亚楠,等. 氟碳涂料在防腐领域的研发现状和发展趋势[J]. 中国腐蚀与防护学报, 2016,36(2):97-106.
[10] 潘金山,郭秀云,谢恺. 高速铁路行车指挥系统信息检测仿真研究[J]. 计算机仿真, 2017, 34(5):168-171.