Highway tunnel maintenance technical specifications clearly pointed out that the investigation of tunnel cracks is one of the special inspection items. At present, manual detection is often used and the missing detection is inevitable. In order to overcome this shortcoming, automatic detection with machine vision has become the main method in this field at home and abroad. In this paper, the background in machine vision methods, the present research on crack of tunnel concrete lining detection algorithms at home and abroad are reviewed. Lining image preprocessing, crack detection, interference elimination, crack width measurement and error analysis are included, while the advantages and disadvantages of different algorithms are compared. Finally, the conclusion and future vision are introduced.