Multilevel Threshold Segmentation Method for Infrared Image of Power Equipment with Two-dimensional Entropy Based on Bat Algorithm

摘要

The region of interest (ROI) segmentation of infrared image is one of the key steps in intelligent fault diagnosis of power equipment. Subsequently, a two-dimensional entropy multilevel threshold based on bat algorithm is proposed. By analyzing the multilevel threshold segmentation principle with two-dimensional entropy, the bat algorithm is used to search the optimal segmentation threshold, and these thresholds are used to the threshold segmentation experiment on the infrared image of the power equipment. The results show that this method improves time consuming of the algorithm and the precision of image segmentation compared with the multi-threshold segmentation method with two-dimensional entropy based on particle swarm optimization algorithm, which effectively solves the problem of multilevel threshold segmentation of infrared image. It lays the foundation for the extraction and analysis of temperature field characteristics of follow-up equipment, and is more suitable for the intelligent diagnosis of infrared images of power equipment failure.