Bolts are widely used as fasteners in transmission lines

and their defect images have the characteristics of small intra-class differences and large inter-class differences. Aiming at the problem that large models with high complexity and excellent performance consume a lot of computing resources in analyzing bolt defect images

a knowledge distillation technology is introduced into the classification of bolt defect images in transmission lines

and a classification method of transmission line bolt defect images is proposed based on dynamic supervised knowledge distillation. The adaptive weighting method is used in the output layer of the network to improve the accuracy of the small model in learning bolt defect labels; moreover

the attention transfer mechanism is used in the hidden layer of the network to improve the bolt feature expression ability of the small model. In order to fully improve the bolt defect classification ability of small model

the adaptive weighting method of the network output layer is combined with the attention transfer mechanism of the network hidden layer. Finally

the effectiveness of the large model using the distillation method proposed in this paper to guide the small model training is verified by the self-built bolt defect image classification data set

and the experimental results show that the classification accuracy of the small model is improved by 2.17%

the classification accuracy of the small model and the large model is only 0.63% difference

and the parameter amount of the small model is only 7.8% of the parameter amount of the large model. The study realizes the efficient classification of bolt defects and compromises the precision and resource consumption.