Method and system for detecting running state of underground belt conveyor based on optical flow estimation

The invention provides an underground belt conveyor operation state detection method and system based on optical flow estimation. The method comprises the steps that image data of a belt conveyor are collected; inputting the acquired image data into a preset optical flow estimation network model to obtain optical flow image data; wherein the optical flow estimation network model is obtained by training a first preset neural network based on a first training data set; inputting the optical flow image data into a preset motion state prediction network model to obtain prediction of the motion state of the belt conveyor; wherein the motion state prediction network model is obtained by training a CNN-Transform network based on the second training data set. According to the method, the start-stop state and the loading state of the belt can be effectively identified, and the problem that an inter-frame difference method of a pure vision scheme cannot well identify the running state of the no-load belt of the belt co