An in-depth explanation of Fotric ‘TWB’ and ‘IRedge’ function

Some recurring questions we receive from clients are: what does your imaging enhancement function (TWB or IRedge) do? And when are we supposed to use them?

We are here to answer them once and for all.

Before we dissect the individual functions, there is an important concept we must first get acquainted with: temperature span. Temperature span refers to the difference between the upper and lower temperature limit on the thermal imager screen. In a thermal imager, given a color palette, the larger the temperature span is, the fewer color representation units can be allocated to each discrete temperature interval that makes up the whole span. Therefore, a narrower temperature span enables the imager to expose more visual details and temperature variation on the object.

The images below were taken at the approximately same time by the same camera, the only difference being the temperature span. In both images, the temperature data are recorded by the thermal imager at the same level of precision. However, the grain and texture of the palm are visible under a narrower temperature span, while these visual details are more or less omitted in the image on the right.


The IRedge function strengthens the visual impact of object contour and edges to help users distinguish them from the background. Despite its name and effect, it is not a detail-enhancing algorithm that incorporates visible images. Instead, it operates solely based on temperature distribution recognition and color palette redistribution. When inspections take place in an environment of broader temperature span (when a high-temperature anomaly is present), objects of close temperature will likely be visually meshed together. If the inspection is under daylight, functions like image fusion might help. However, at night, it will take considerable effort to accurately distinguish one object from another and pinpoint the defect’s location.

Even though the visual details of a thermal image might have been compromised due to larger temperature span, the temperature measurement of each pixel is still accurate. Essentially, what IRedge does is looking for a temperature ‘step-function’ (abruptly large derivative) in an image. When this phenomenon occurs collectively, it constitutes an ‘edge’ (temperature distribution recognition). If the magnitude of the step-function is too small in comparison with the temperature span, the IRedge function would allocate more visual representation resources to the temperature range the edges are in (color palette redistribution). As a result, the temperature difference edges have would appear visually starker for inspectors.


In short, IRedge works exceptionally well in an environment of large temperature span to make objects more distinguishable from the background. However, it won’t make much of a difference in an environment of narrower temperature span.

When the TWB is not activated, the palette ribbon represents temperature in a linear manner (a certain length on the ribbon projects the same proportion of temperature difference on the overall temperature span). However, this representation is not always adequate. Extreme temperature points on the image can lift the upper limit of the palette ribbon, leaving large chunks of the ribbon to represent a few pixels. The elevated temperature range could easily blur out visual differences at lower temperature. TWB essentially re-scales the palette ribbon based on the number of pixels in representing each temperature range. Consequently, the temperature distribution of the entire image is more clearly laid out for the inspector.


The TWB mode is necessary when the presence of a temperature outlier obscures our judgment on temperature distribution by omitting subtle temperature changes. For example, in the metallurgy industry or house inspection with the presence of incandescent bulbs. However, on defect locating tasks, some inspectors would prefer anomalies to be as visually salient as possible. Having TWB turned on will likely normalize the abnormal area’s visual impact on the image, which is not a desirable outcome.

Here in Fotric, we strive to listen to our customers’ needs and solve their pain points. If you have any questions or suggestions for us, please don’t hesitate to reach out and make yourself heard at support@fotric.com or info@fotric.com.