A Machine Vision based Rock Slag Monitoring System of Tunnel Boring Machine

In this paper, a rock slag monitoring system of tunnel boring machine (TBM) is established with machine vision technique to guarantee the safety and efficiency of TBM operation. An industrial camera is used to catch the slag image, and a system is design to monitor the rock slag image. Firstly, a fractional image enhancement algorithm is studied for the unclear rock slag image, which is caused by the motion blur. Secondly, Hough transform method is used to distinguish and recognize the boundary line on the threshold image. The maximum size, the minimum size and the average size of rock slag are estimated based on the identified area. Thirdly, K-means clustering algorithm is used to classify the types of the surrounding rocks. The monitoring system designed in this work can monitor the real-time working condition and surrounding rock of a TBM. The specific functions include monitoring of the sizes of the rock slag pieces, the load of the conveyor belt and the classification of surrounding rock. Finally, experiments are performed to show the effectiveness of the proposed system.