一种基于MASK R-CNN的铁路隧道裂缝识别方法

A Method of Crack Identification in Railway Tunnel Based on MASK R-CNN

铁路隧道裂缝会影响隧道的稳定性,严重时危及列车运行安全,因此,隧道裂缝检测成为避免隧道重大灾害和日常养护的重要工作。传统检测方法以人工为主,检测效率和质量低下,无法满足实际需求,同时基于图像处理的裂缝识别算法准确度不高,使用场景局限,完成任务相对单一。针对于此,提出了一种基于深度学习神经网络Mask R-CNN模型对裂缝进行智能检测的方法,通过调整算法参数,优化模型检测结果,获得适用于隧道裂缝检测的Mask R-CNN模型,从而实现铁路隧道裂缝快速、准确地检测与定位。

Cracks in the railway tunnel will affect the stability of the tunnel and seriously endanger the safety of train operation. Therefore, tunnel crack detection has become an important work to avoid major disasters and daily maintenance of the tunnel. The traditional detection methods are mainly manual, and the detection efficiency and quality are low, which can not meet the actual needs. At the same time, the accuracy of the crack recognition algorithm based on image processing is not high, the use of the scene is limited, and the task is relatively simple. In view of this, a method of intelligent crack detection based on deep learning neural network Mask R-CNN model is proposed. By adjusting the algorithm parameters and optimizing the model detection results, the Mask R-CNN model suitable for tunnel crack detection is obtained, thus the railway tunnel cracks can be detected and located quickly and accurately.