An Image Processing Based Approach Towards Recognition of Underground Cable Defects

This paper presents a novel approach for detecting possible faults in underground cables at the manufacturing industry using image processing techniques. With the increasing adoption of underground cables to minimize faults and outages in power transmission systems, ensuring the quality of these cables becomes crucial. Detecting defects at the manufacturing stage can save significant costs and effort by identifying faulty cables before they are deployed in the field. We propose a system that utilizes image processing algorithms to analyze cable images and identify potential faults, such as unbalanced currents, cable sheath faults, and short circuits. The system also calculates the area of each core to ensure equal current flow, and estimates the total cross-sectional area of the cable to assess the adequacy of insulation. The proposed approach offers a non-destructive and efficient method to enhance quality control in the manufacturing process of underground cables.