Multi-scale multi-task neural network combined with transfer learning for accurate determination of the ash content of industrial coal flotation concentrate

•MSTNet is developed for rapid and accurate ash content determination of coal flotation concentrate.•The proposed MSTNet performs multi-scale and multi-task analysis on the froth images of coal flotation.•SimpleResizer is designed to allow MSTNet to accept image with different resolutions without modification.•DRAttention is designed to generate specialized feature vectors for input.•Transfer learning is innovatively used to further improve model accuracy for low-resolution inputs. Ash content is a key indicator to evaluate coal flotation concentrate quality and adjust flotation process parameters, which could be determined by analyzing froth images. In this research, a multi-scale multi-task neural network (MSTNet) was developed to realize accurate determination of the ash content of industrial coal flotation concentrate by analyzing froth images. Furthermore, transfer learning is used to further improve model accuracy for low-resolution images. Results obtained using industrial data show that MSTNet achieves a higher prediction accuracy while requiring less computations than previous models. It reaches the maximum R2 of 0.9063 with a processing time of 0.0035 seconds per image, while its competitors only reach the maximum R2 of 0.7231 with a processing time of 0.0038 seconds per image. This suggests that MSTNet surpassing its competitors in both accuracy and speed. Furthermore, MSTNet achieves the minimum MAPE of 0.0300, indicating that MSTNet has a mean relative prediction error of ± 3 %. This proves the high prediction accuracy of MSTNet. These results indicate that the proposed MSTNet holds great promise for practical applications. Its practical application will lead to more efficient and intelligent coal production.