利用激光图像技术检测路基压实度

为了检测路基压实度,设计了路基压实度激光图像检测装置,并对该检测方法进行了研究。从特征图像中分别提取平均灰度级、标准方差、平滑度、三阶矩、一致性、熵、能量、对比度、相关性和均匀性等10个基于灰度共生矩和基于直方图的纹理特征参量,以及含水量参数;通过PCA(主成分分析)降维处理,从原始数据中优选出4个主成分因子参量;最后,利用BP神经网络预测密实度。实验结果表明:(1)BP神经网络经过11次学习后达到了要求的误差;(2)利用系统预测的密实度数值与环刀法检测值相比较,平均绝对和相对误差在2.1%左右。因此,该检测方法检测路基压实度是可行的。

Detection of Subgrade Compaction Using Laser Image Technology

In order to test subgrade compactness, a laser image measurement system of soil compactness was established and thenondestructive measurement method of soil compactness was investigated.ten texture features (i.e.,mean, standard deviation,smoothness, third moment, uniformity, entropy, homogenity, energy,correlation and contrast) based on the statistical moment andhistogram were extracted from each image, and the water content parameter,The principal component analysis (PCA) was performed toselect four Characteristic parameters. Finally, soil compactness was predicted by BP neural networks. The experimental results showthat BP neural networks reached the required error after eleven loops, compared with the measurement values by using round knifemethod, the avervage absolute error and relative error of compactness were less than 2.1%. Therefore, the measurement of the systemis feasible.