压缩感知在隧道病害识别中的应用研究

Study on Application of Compressed Sensing in Recognition of Tunnel Fault

该文通过分析隧道病害识别的数据采集过程,提出关于压缩感知理论的隧道病害识别系统的数据采集方法。对采集的隧道病害识别图像进行小波变换,将隧道病害识别图像变换成易传输,数据量较小的图像数据,使用正交匹配追踪算法,以及小波逆变换对来自传输设备的隧道病害识别图像进行重建,并实现原始图像精确和近似重构。由仿真可知,压缩感知理论应用于隧道病害识别的数据采集传输过程,降低了传输数据量,并易于图像数据传输,且能以较小误差实现隧道病害识别的重构,同时有相应的硬件使用环境。从而为加快隧道的检测提供新方法。

The article analyzes the data collection process of tunnel fault recognition, and puts forward the data collection method of tunnel fault recognition system with regard to the compressed sensing theory. The collected tunnel fault recognition image is carried out of the wavelet transformation. The recognition image of tunnel fault is transformed into the easy transmission and smaller data size of image data. The orthogonal matching pursuit algorithm and the inverse wavelet transformation are used to rebuild the recognition image of tunnel fault from the conveying equipment, and realize the accurate and approximate reconstitution of the original image. The simulation shows that the application of compressed sensing theory in the data collection transmission process of tunnel fault recognition can reduce the data value of transmission, easy transmit the image data and realize the reconstitution of tunnel fault recognition in the smaller error, and at the same time, there is the relative hardware using environment, which can provide the new method to speed up the detection of tunnel.