Airffic Aircraft InstituteAircraft, evidence and the market
HomeFrom flight to workDoes a red area in a thermal image mean a fault?

From flight to work / 03

Does a red area in a thermal image mean a fault?

Thermal patterns are clues; diagnosing a fault requires context and other evidence.

5 minResearch reviewed through 5 November 20264 sources
On this pageColor is a way to display measurementsCapture conditions belong with the imageCompare AI results on comparable evidenceFinish with the maintenance handoffFollow the cause and effectSources for this lesson
Sir Airffic Go
Sir Airffic Go

Let us examine what a thermal image measures and what still needs interpretation.

From the evidence

Color is a way to display measurements

A thermal camera records infrared radiation and displays the measurements as an image. Its colors are a visual scale, rather than the visible colors of the object. Red or white therefore has meaning only within that display and the conditions of the measurement; it is not a universal fault label.

In solar-panel inspection, unusual thermal patterns can identify areas worth investigating. Sunlight, reflections and wind can also influence the image. The question is not simply whether an area looks different, but what could explain the difference under the conditions in which it was captured.

Our interpretation

Capture conditions belong with the image

An inspection plan should specify the required detail, useful viewing angles and the area that must be covered. Broad coverage can still miss the evidence needed for a particular decision. Enlarging an inadequate image later does not recreate information that the camera failed to collect.

Visible photographs can provide context about surfaces, objects and dirt. Connecting them to the corresponding thermal views helps an inspector compare explanations. Close-quarter inspection research illustrates this integration challenge: carrying two cameras is only the beginning; their observations must refer to the same location and usable evidence.

Turn thermal differences into decision evidence
Record conditions
Find thermal differences
Compare images and locations
Assess the cause

A conceptual workflow, not a replacement for asset-specific procedures or expert judgment.

From the evidence

Compare AI results on comparable evidence

A review of photovoltaic inspection found considerable attention to automated fault detection and classification. It also found that studies used different datasets, fault categories and evaluation measures. A headline accuracy percentage alone therefore cannot provide a fair ranking of methods for a new site.

For illustration, identifying one obvious anomaly is a different task from distinguishing several possible causes. A meaningful result should explain what was classified and what images were used. The buyer then needs evidence relevant to the intended work, rather than a number detached from its test.

Different evidence answers different questions
EvidenceQuestion it helps answer
Thermal imageWhere is the thermal pattern unusual?
Visible imageWhat does the surface and context look like?
Location recordWhich part of the asset is this?
Capture conditionsAre the observations comparable?
Further checksIs there enough evidence for a repair decision?
Our interpretation

Finish with the maintenance handoff

Flyability included thermal and visible imaging in Elios 2, and later connected inspection points to 3D models with Elios 3. These announcements show a direction of product development toward useful asset evidence. They do not establish equivalent performance across solar plants and the indoor facilities those products target.

A useful handoff identifies the image, location, relevant conditions and what has actually been checked. Maintenance staff can then apply the asset’s decision criteria. The aircraft helps collect evidence; the completed inspection must still explain how that evidence supports a practical action or a need for further investigation. Asset identifiers and capture dates also make it easier to find information about the same equipment at the next inspection.

The idea to keep

Inspection connects a thermal anomaly with its likely cause and the correct asset.

Try the idea yourself

What additional information would help explain an unusual thermal pattern?

Sources for this lesson