Sag Monitoring of Overhead Power Lines based on Image Processing

Inspection of power line infrastructures must be periodically carried out by utilities in order to ensure reliable and safe power distribution. Research efforts are focused on automating the power line inspection process by looking for methods that satisfy the different requirements, e.g. simultaneous detection of transmission towers, checking for defects, and analyzing security distances. Following this direction, and since the dynamic line rating is a hot topic in power transmission, this paper proposes an approach based on computer vision to measure the sag and deduce the distance between the line and ground. This method uses the hyperbolic cosine function to draw different templates tagged with according calculated sag. Then each template will be matched to the test image later using template matching. Evaluations show highly encouraging results and therefore. This shows that a computer vision-based approach is a promising technique for real-time, non-contact power line monitoring.