Research On ADCN Method for Damage Detection of Mining Conveyor Belt

Belt conveyor is considered as a momentous component of modern coal mining transportation system, and thus it an essential task to diagnose monitor the damage belt in real time accurately. Based on deep learning algorithm, this present study proposes method detection based ADCN (Adaptive Deep Convolutional Network). A convolution network with unique adaptability built extract different scale features visible light image damage, target classified located form anchor boxes. data set diversity collected according actual working conditions belt. After training regression, model can perfectly capture classify damaged video running. Compared SVM method, better meet real-time reliability requirements detection, has positioning function which does not have.