Conveyor belt speed detection via the synergistic fusion of optical flow and feature matching

This paper studies a conveyor belt speed detection method based on the fusion of optical flow and feature matching. This method combines the ability of the optical flow method in calculating the motion information of image pixels and the advantage of the feature matching method in accurately identifying feature points. Through the mutual assistance and fusion of the two, the accurate detection of the conveyor belt speed is realized. An optimized Bayesian algorithm is adopted to distribute the weights according to the accuracy and reliability of the optical flow method and the feature matching method, as well as the light intensity, and the final belt speed value is obtained through comprehensive calculation. This research method has significant advantages in improving the performance of conveyor belt speed detection, and provides support for the operation monitoring and fault diagnosis of the conveyor belt.