摘要
分析了数字图像采集过程中可能受到的影响,提出了沥青混凝土路面数字图像的采集方法。利用数字图像技术对沥青混凝土路面的构造深度进行了分析,建立了数字图像矩阵信息与路面构造深度的关系模型,提出了利用广义回归神经网络减少环境误差的方法。实验证明数字图像技术不仅能降低沥青混凝土路面构造深度的检测成本,也能提高其检测速度,具有较好的应用价值。
The influence that may be caused during the procedure of digital image capture is analyzed,and digital image capture methods for asphalt concrete pavement are put forward. The analysis on the structure depth of asphalt concrete pavement is made by utilizing the digital image technology,the relational model of digital image matrix information and pavement structure depth is built,and the method of utilizing generalized regression neural network to reduce the environmental error is put forward. The experiment shows that the digital image technology cannot only reduce the measurement cost of asphalt concrete pavement structure depth,but also improve the measurement speed,which has better application value.
作者
何力
HE Li(Chongqing Jiaotong University,Chongqing 400074,Chin)
出处
《北方交通》
2018年第6期78-81,共4页
Northern Communications
关键词
图像采集
图像处理
构造深度
神经网络
Image capture
Image processing
Structure depth
Neural network
