Thermal Imaging: A Colorful Introduction
Lessons Learned – Part 1 of 4
A series of insights brought to you by the Consortiq team
To understand thermal imaging and its applicability in our world, we need to start with the key question: What is a thermal sensor?
To understand this, we need to look at some basic – but quantum – physics.
In essence, we can see because light reflects off an object. That light then enters the eye where rods and cones (receptors) on the back of the eye collect the inputs, and send them to the brain to be processed via the visual pathway.
Therefore, we can ‘see’ this light, as it becomes visible.
However, “visible” light makes up a small part of the much broader electromagnetic spectrum illustrated here:
Thermal imaging image courtesy of wikipedia on a Creative Commons Attribution-Share Alike 3.0 Unported license
All of these differing wavelengths — from very long waves, such as radio, to very short waves, such as gamma radiation (the shorter the wavelength the more frequently they arrive, hence low to high frequency) — make up the entire spectrum range.
The key fact in all of this is that, of the entire electromagnetic spectrum, only 0.0035% is the light we actually see – “visible” light.
So, back to thermal sensors.
They’re capable of ‘seeing’ electromagnetic radiation in the infrared band of the spectrum. Infrared radiation is closely linked to the temperature of the object, so there is a correlation between the object’s temperature and the electromagnetic radiation it emits.
As that object’s heat increases, it emits higher frequencies on the spectrum until it enters the visible light band and glows red hot.
To get technical, everything has a natural ‘emissivity’ level. In other words, objects have varying levels of effectiveness as it pertains to emitting thermal energy as thermal radiation.
A thermal sensor uses those levels of emissivity to create a picture. By assigning colors or shades to different thermal radiation values, you get the familiar thermal image.
It’s not true to say that a thermal camera can measure the temperature of an object, per se. It’s possible, but only if the camera is set up and calibrated correctly. Additionally, the emissivity values for the substance being measured must be known.
Simply put, if you know that steel has an emissivity value of “X” at a given temperature, then you can calibrate your system so that, when it ‘sees’ that emissivity value on a steel object, it displays that temperature value. However, if you use the sensor on something other than steel without recalibrating it, the system will display an incorrect temperature reading.
Now, thermal sensors are very good at displaying temperature differentials… not necessarily by amount, but they do show that there is a difference.
By adjusting the scale — or gain — you can make these differences even more evident, which is particularly useful for certain types of surveys.
For example, electrical components that are heating differently to the same component under the same load elsewhere may be indicative of a fault. Moisture leakage may give an artificially cooler result than would be expected.
The proper use of thermal sensors can provide greater insight into a survey area than just a visual inspection, but interpreting the resulting data is crucial. Unlike a visual inspection, there is much analysis that can be done with thermal imagery if captured in a radiometric format.
In effect, this is capturing the thermal data in what a photographer may refer to as “raw footage.” The ‘picture’ is therefore not just an image, but an accumulation of data. That data can then be manipulated and analyzed to produce a detailed analysis of the properties of the area, item or, indeed, building which has been surveyed.