Application of convolutional neural network in ore sorting technology based on dual-energy x-ray transmission image

Automatic sorting technology based on dual-energy x-ray transmission images has become an indispensable technology in the field of ore sorting, due to its advantages of large processing capacity, no pollution, and high accuracy. Traditional dual-energy x-ray image sorting uses dual-energy curve method, which requires complex image processing algorithm and pixel value extraction algorithm. In this paper, the convolutional neural network is used to replace the traditional method for image classification, and the calculation process is simpler. Validation experiment shows that the accuracy of the convolutional neural network is slightly higher than that of the dual energy curve method, and the computation time is shorter than that of the traditional method.