Lightweight network based on residual information for foreign body classification on coal conveyor belt

The coal in the mine must be transported long distance by coal belts before reaching the ground. Plenty of studies on the safe and efficient transportation of coal in mines reveal that the belts often suffer some hazards caused by foreign objects such as large gangue, bolts and other foreign bodies scratching, tearing the belt, and blocking the coal discharge point in the process of coal transportation. If the early warning, sorting and linkage control are not timely, it will seriously affect the coal transportation efficiency. To overcome the problems of large amount of network parameters, poor real-time performance, and low recognition accuracy in the current classification and recognition of belt foreign objects, a lightweight network that integrates residual information is proposed. Firstly, the residual block is adopted as the basic feature extraction unit of the network, and the activation function is removed between the convolution blocks in the residual block. Then, the cross-learning mechanism and feature splicing method are used to fuse the feature information of different scales, resulting in the enhanced expressiveness of the features. Furthermore, the structure of the information fusion network is simplified and the number of information fusion networks is increased, which improves the scalability of the model. Moreover, the loss function is thresholded during the forward propagation of the model, which can solve the problem of elevated test set loss function and improve the generalization of the model. By conducting the experiments on the Cifar10, Cifar100 and the mining dataset, the recognition accuracy of the proposed network model can reach as high as 94.1%, 73.9% and 85.1%, respectively. Compared with the ShufflenetV2, MobileNetV2, ResNet50, ResNeXt50, W-ResNet50 and ResNet110 algorithms on the mining dataset, the accuracy rates proposed are 4.2%, 4.3%, 0.7%, 0.5%, 0.3% and 0.8% higher than those respectively. In addition, compared with ResNet50, ResNeXt50, W-ResNet50 and ResNet110, whose classification accuracies are similar to the network proposed, the FPS can be increased by 28, 26, 34 and 46, respectively. The results demonstrate that while improving the classification and identification accuracy of foreign objects, the calculation speed of the proposed algorithm in this paper has also been accelerated, which can significantly improve the transportation efficiency of coal and promote the deep integration of computer vision and coal mine safe production.