Manuscript received November 12, 2025; revised December 8, 2025; accepted March 2, 2026; published September 29, 2026.

Abstract—Thermal imaging is one of the best modern methods used to detect faults in electrical circuits and electronic maps and identify short circuits, unlike traditional methods. However, it is rarely available, expensive, and does not have open source software for modification and development. In an effort to provide cost-effective technical solutions, this research presents an innovative methodology for manufacturing high-precision alternative thermal cameras by modifying traditional USB cameras at a cost of no more than $50 while maintaining similar performance, as the economic cost of thermal cameras is estimated at approximately $2000. A convolutional neural network is also designed to improve results by training this neural network on images of faults and predicting faults in electronic circuits using artificial intelligence and comparing different algorithms to obtain the best results, such as (edge detection, Sabel, Gaussian, Laplacian, Canny, Fourier transform), where a fault diagnosis prediction rate of 91.3% was obtained in this innovative system, which is a very high rate without human intervention and predicting errors before they occur using convolutional neural networks. The fault diagnosis time does not exceed 2.3 s, and the system has the ability to automatically recognize new fault patterns without the need to reset the model. These results open up new horizons in the field of predictive maintenance, as the proposed system can be integrated with smart monitoring systems to create highly efficient preventive maintenance solutions, especially in sensitive electronic systems

Keywords—self-diagnostic system, electronic faults, infrared thermal camera, convolutional neural networks, image processing, Printed Circuit Boards (PCB) testing, short circuit

Cite: Abdulrafa Hussain Maray, Mohammed T. Majeed, and Najm Abdullah Abdulqader Tahhan, "Intelligent Diagnostic System for Electronic Board Faults Using an Infrared Thermal Camera and Convolutional Neural Networks," Journal of Image and Graphics, Vol. 14, No. 5, pp. 808-816, 2026.

Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).