The cracking of tunnel lining structure is a key disease that needs to be prevented and controlled in tunnel maintenance work. The development of intelligent tunnel management urgently requires a quantitative evaluation method for the cracking disease of tunnel lining structure that can be promoted and applied. In this paper, the automatic recognition technology of crack image is introduced, and the quantitative evaluation index system and threshold determination method of single crack disease and m ulti crack disease of lining structure section are established respectively by fusing the product scale method and the specification requirements. The length and width of lining crack are taken as the main evaluation indexes for the diagnosis method of single crack disease, and the depth, direction and development of lining crack are taken as the auxiliary evaluation indexes; The diagnosis method for multi crack diseases uses the length and width of lining cracks as the main evaluation indicators, and depth, direction, development, and distribution density as auxiliary evaluation indicators; The determination of indicator threshold not only considers the probability distribution characteristics of uniform distribution function but also combines the safety characteristics of tunnel structure under stress. The scores of the evaluation method have been systematically standardized without dimensionality, and software that can be promoted and applied has been developed. Case analysis shows that this method is scientifically reasonable, providing a new way to achieve intelligent identification of tunnel cracking diseases and automatic decision-making of maintenance plans.