Failure Detection on Electronic Systems Using Thermal Images and Metaheuristic Algorithms

Abstract

Segmentation is considered an important part of image processing. There are commonly used segmentation techniques to improve the thresholding process such as Otsu and Kapur. The use of these techniques allows us to find the regions of interest in an image by accurately grouping the pixel intensity levels. On the other hand, the use of thermal images makes it possible to obtain information about the temperature of an object and to capture the infrared radiation of the electromagnetic spectrum through cameras that transform the radiated energy into heat information. The segmentation of this kind of images represents a challenging problem that requires a huge computational effort. This work proposes the use of metaheuristic algorithms, combined with segmentation techniques and applied to thermal images, to detect faults and contribute to the preventive maintenance of electronic systems.