Analysis of Key Technologies of Bridge Damage Detection Based on Visual Recognition

In the identification of bridge damage in series deflection-affecting lines, noise signals due to axle coupling often occur. Removing these has become a key technique for effectively identifying damage. Taking main line overpass project Fuzhou City Fujian Province as an example, we collected deflection data by using HPON-X target-free deflectometer and used YOLOv3 algorithm deep learning vehicle load position. The were measured studied DB9 wavelet de-noising method. research shows that this method can greatly reduce influence interaction on line, enhance accuracy speed detection.