The development of more electric aircrafts (MEA) and all electric aircrafts (AEA) inevitably implies an increase in electric power and a consequent increase in distribution voltage levels. Increased operating voltages coupled with low pressure in some areas of the aircraft greatly increase the chances of premature insulation failure. Insulation failure manifests itself as surface discharges, arc tracking, arcing, and disruptive or breakdown discharges, in order of increasing severity. Unfortunately, on-board electrical protections cannot detect discharges at an early stage, so other strategies must be explored. In their early stage, insulation faults manifest as surface and corona discharges. They generate optical radiation, mainly in the near-ultraviolet (UV) and visible spectral regions. This paper focuses on a method to detect the discharges, locate the discharge sites, and determine their intensity to facilitate predictive maintenance tasks. It is shown that by using small size and low-cost image sensors, it is possible to detect, locate, and quantify the intensity of the discharges. This paper also proposes and evaluates the behavior of a discharge severity indicator, which is based on determining the intensity of digital images of the discharges, so it can be useful to apply predictive maintenance tasks. The behavior and accuracy of this indicator has been tested in the laboratory using a low-pressure chamber operating in the pressure range of 10–100 kPa, which is characteristic of aircraft applications, analyzing a needle-plane air gap geometry and using an image sensor. The proposed method can be extended to other applications where electrical discharges are an issue.