Thermal Imaging Specifications Explained

Thermal imaging cameras are an incredibly useful tool in conservation and wildlife monitoring, however they come with a complex set of technical specifications that are not always fully explained, making it difficult to choose the best model for your purposes.

In this article we aim to clearly explain what each of the key technical specifications mean, and hopefully enable you to make an informed decision.

Detection, Recognition and Identification ranges (DRI)

The distance at which, using a thermal device, you can detect, recognise and identify an object is based on the Johnson Criteria, a set of military standards used to compare the capabilities of different thermal devices. These criteria are not perfect and advances in technology since their inception in the 1950s have exacerbated some of their flaws, however despite limitations they remain the current standard used by manufacturers. So if you are looking to purchase a thermal camera it is worth understanding what they mean.

One thing to note with DRI is that the range values are based on the assumption that you will have ideal conditions (i.e. good weather). In most applications this will not be the case and so it is safe to assume that these distances will be reduced.

It is also important to understand that these distance measurements are based on a 50% probability that you will successfully be able to discern the animal at each of these distances. This, of course, cannot be guaranteed and so in many instances these distances will be smaller.

Illustrative example of DRI in thermal imaging devices

1. Detection Range

The detection range is the maximum distance that you can perceive a 1.8m tall object when looking through the device. At this distance the object would cover approximately 2 or more pixels on the device. Manufacturers will list this in metres in the specifications.

At this distance you will not be able to discern what you are looking at, but you will be able to detect the presence of something warm against the background. Providing there is sufficient light you can use a pair of binoculars or a spotting scope to inspect the heat source in greater detail and quickly identify what you are looking at.

Thermal image of Red Deer captured on an XP50 at 1500m This image shows two Red Deer in the Scottish Highlands, photographed on a Pulsar XP50 at a distance of approximately 1500m. It illustrates detection distance, as the heat sources do stand out in the landscape but it is not possible to confidently discern what they are. We used a pair of binoculars to confirm these were two Red Deer.

2. Recognition Range

The recognition range is the distance at which the critical mass on your subject covers around 6 or more pixels. This value is not listed by manufacturers, but it can be considered to be approximately 40% of the detection range for most thermal devices in good conditions.

At this distance you will be able to discern what type of animal you are looking at (e.g. deer, dog, bird), but you will not be able to see enough detail to identify anything closer to the family or species level. Again, providing it is light you can use a pair of binoculars to get a closer look at your subject animal.

3. Identification Range

The identification range is the distance at which the critical mass on your subject covers at least 12 pixels. Again this value is not listed by manufacturers, but it can be considered to be approximately 20% of the detection range for most thermal devices in good conditions.

At this distance you will be able to discern more details about your subject. In some cases this will allow you to identify to the species level, but this is highly context-specific. In most cases, you will only be able to identify the animal to the family and/or class level.

Sensor Resolution

This value defines the number of pixels that make up the sensor, also called a microbolometer, in a thermal device. You will see this presented as two values (e.g. 320 x 240) and these each define the number of pixels vertically and horizontally respectively.

In simple terms, if you increase the number of pixels on a microbolometer you will get more temperature points recorded and consequently, you will get a more detailed image. More expensive thermal devices do exactly this, and feature sensors with much larger resolutions than more economic ones.

It is important to avoid confusing sensor resolution with display resolution or recording resolution.

  • Display resolution is the resolution of the built-in screen that you see when you look through the device. This will often be greater than the sensor resolution. It is worth making sure you don't mistake these two terms.
  • Recording resolution is the resolution at which images and photos are captured. On economic or entry-level devices, this can be smaller than the display resolution, and consequently the quality of captures will be slightly poorer than what you see when looking through the device.
Comparison Image of wading birds on an Axion XM30F and Axion 2 XG35 Side by side comparison of images of waterfowl taken on Axion XM30F and Axion 2 XG35. Sensor resolutions are 320x240 and 640x480 respectively. Image is illustrative of differences in quality between different resolutions, however the lower recording resolution of the XM30F means images taken with it will look poorer than what you see when looking through the viewfinder.

Pixel Pitch

This is the distance between the centres of two adjacent pixels on the sensor. This is measured in micrometres (e.g. 12µm) and is normally listed after the sensor resolution.

Devices with smaller pixel pitches are more detailed as they have a higher density of pixels on the sensor. In other words, having a smaller pixel pitch means the device can take more temperature point measurements than one with a larger pixel pitch, and thereby give a more accurate and detailed image.

At the time of writing, you will typically see either a 12µm or 17µm pixel pitch on most handheld thermal devices.

Illustrative example of Pixel Pitch in thermal imaging devices An illustrative example of how lower pixel pitch allows for greater density of pixels and consequently more detailed images.

Noise Equivalent Temperature Difference (NETD)

In simple terms, this value describes the minimum difference in temperature that the detector is able to distinguish. This value is displayed in milli-Kelvins (mK) (e.g. <40mK).

Thermal imagers with smaller NETD values produce a more detailed image and in particular they make it easier to pick out objects from the background. This is particularly useful in situations where your subject is a similar temperature to its surroundings, such as in rainy or foggy conditions.

This image shows the detail visible on an Axion 2 XG35. This device has a <40mK NETD value which offers great detail for a pocket sized monocular. More advanced devices such as the Helion 2 XP50 Pro have a much lower NETD value of <25mK which is superb for resolving small temperature differences.

Refresh rate

This describes the frequency that the image produced by a thermal device is updated/refreshed per second. This will be listed as a Hertz (e.g. 30Hz).

Higher refresh rates (>30Hz) will offer smoother video and ensure you do not miss any fast moving animals. This is particularly important for monitoring bats, when it is essential to use a device with a refresh rate of at least 30Hz (Fawcett Williams, 2021).

Almost all thermal monoculars now have a sufficient refresh rate for most wildlife monitoring, however, you should double-check this if considering a very small or inexpensive thermal device. Thermal cameras for smartphones often have very low refresh rates of 5-10Hz and this can make for very choppy viewing.

Magnification

Magnification describes how many times larger the object you see through a detector appears than when observed with the naked eye. You will see this displayed as a ‘value x’ (e.g. 3x).

Because thermal imaging devices are typically used for scanning and detection, high magnification is uncommon and most models are between 1x - 5x.

Thermal devices with larger base magnifications are useful for observing subjects at greater distances, however this comes at the cost of having a narrower field of view. You will need to consider what is a good compromise between these two parameters and select a device that is best suited to your needs overall.

Field of View / Angle of View

This is a slightly more complicated specification as the terms Field of View and Angle of View are used interchangeably and this can confuse things. A quick look online will lead down a bit of a rabbit hole!

In very simple terms Field of View (FOV) is a measurement of how large an area can be seen at a certain distance when looking through a device. For thermal devices you will see this listed as a horizontal and vertical distance measurement in meters, at a set distance, typically 100m or 100yd (e.g. Field of view (HxV), m@100m, 21.8 x 16.3).

Angle of View (AOV) on the other hand is the extent of the scene visible measured angularly. This is a slightly more complicated term as it is affected by both the sensor size and the focal length of the device. It is beyond the scope of this article to explain focal length in detail, but all you need to know in practice is:

  • Devices with shorter focal lengths (measured in mm) will have a wider AOV and consequently you will see a larger area than you would on a device with a longer focal length.
  • Devices with larger sensors will capture more and the resultant AOV will be wider than those with smaller sensors.

AOV is rarely listed separately by manufacturers, and instead you will see this listed in degrees (°) alongside FOV. Most manufacturers will only list the horizontal angle, however some will also provide the vertical angle as well (e.g. Field of view (HxV), degrees 12.4° x 9.3°).

You need to consider is what sort of FOV/AOV would be best suited for your target species. If you know you are looking for small fast-moving species, a device with a wider FOV/AOV (such as the Pulsar Axion 2 XG35) would be best suited as this will make following the animal or scanning the area easier. If you need to get close-up images of distant or small subjects then consider a device with a greater magnification and a narrower FOV (such as the Pulsar XP50 Pro).

Comparison thermal Image of cattle on an Axion XM30F and Axion 2 XG35 Side by side comparison of images of cattle taken on Axion XM30F and Axion 2 XG35. Overall you can see how the specifications for these two monoculars result in a very different FOV/AOV. The sensor resolutions are 320x240 and 640x480, and it is worth noting that the base magnifications are 3.0x and 2.5x respectively.

Digital Zoom

Most thermal devices will allow you to zoom in on the picture. You will often see this displayed as separate values for each level of zoom, ‘value x / y’ (e.g. 3x / 4x). Sometimes the digital zoom is instead listed at the end of the magnification (e.g. 3.2x - 12.0x, the latter value is the base magnification with a 4x digital zoom).

This zoom will increase the size of your subject however because the resolution does not increase it will appear more pixelated.

Thermal Image of Geese captured on an Axion 2 XG35 This image shows a picture-in-picture zoom function of the Pulsar Axion XG35. The 10x zoom can help with focusing on a part of the image, however it does not increase its quality meaning it can appear more grainy/pixelated. The content of this article is copyright of NatureSpy. All rights reserved. You may print and/or download its contents for your own personal, non-commercial use only. Re-sale, commercial exploitation, third-party use and any redistribution or reproduction of part or all its contents in any form is prohibited without express written permission. Nor may you transmit or store any of its contents in any other website or other form of electronic retrieval system.