Crack defect detection model and implementation method thereof

The invention discloses a crack defect detection model and an implementation method thereof, the detection model is composed of an input module, a backbone network (ECSP-Darknet53), a thin neck feature fusion module (E-Neck) and a prediction output module, the backbone network ECSP-Darknet53 is composed of one CBS unit, four EBS units, four E-C3 modules and one SPPF, is responsible for reducing network model parameters and floating point operand, and is used for detecting the crack defect. Meanwhile, the extraction capability of the network on the deep feature information of the crack is enhanced; the thin neck feature fusion module E-Neck is composed of five EBS units, four ECACSP modules and a plurality of up-sampling and splicing operations, and is responsible for fusing features, enriching the expression ability of a feature map, improving the problem of loss of detail information in the feature fusion process and accelerating the network reasoning speed at the same time; and the E-Neck is responsible for