Application research of contact network image detection based on support vector
The complexity of the contact network parts and dramatic changes in the background environment make it difficult to detect the process. Based on this, this paper builds a corresponding support vector algorithm based on the kernel function to detect the image of the contact network. Meanwhile, this study establishes a support vector training model based on linear kernel, radial basis kernel and polynomial kernel. It is a single texture type training model based on linear kernel support vector method. In this study, the effectiveness of the kernel-based support vector method for insulator identification is verified by comparing the time and recognition performance of each model. At the same time, this study verifies the validity of the model by experimentally analyzing the spatial positioning accuracy of the mapping model. This provides a theoretical reference for subsequent related research.
L. Zhang, "Application research of contact network image detection based on support
vector.", Engineering Intelligent Systems, vol. 26 no. 2-3, pp. 75-82, 2018.