In order to monitor the surface condition of the conveyor belt in the process of running, a method of beam structure light irradiation based on machine vision is adopted. A spring-type mechanical vibration damping device is designed to improve the focusing quality of the camera, and an algorithm is proposed to solve the selection of spring parameters under different flutter amplitudes. Yolov7 deep learning algorithm was adopted and ACmix attention mechanism was introduced to identify the surface cracks of conveyor belt. The experimental results show that the improved YOLOV7-ACmix algorithm can effectively improve the accuracy and generalization ability of image recognition.

Cracks are one of the main causes of structural failure and they develop in the structures due to various reasons such as fatigue, temperature variation, excessive load, cyclic load, environmental effects, impact loading etc. Thus, structural health monitoring is necessary to avoid risks, damages and failures. So, in order to avoid an extensive failure or accident, the early prognosis of crack in structures is necessary. Visual inspection and some non-destructive testing (NDT) methods for detection of crack are difficult as it requires time, expenses and are quite inefficient. So the alternative methods are motivated to be developed. In this study, vibration analysis, finite element analysis (FEA) and an alternative way which is artificial neural network (ANN) is used to predict, detect and identify the damages in structures. It is found that the theoretical, experimental, finite element analysis and artificial neural network have good accuracy in predicting the crack characteristics.