IV - FROM DATA TO INFORMATION
2- HOW CAN WE IMPROVE THE RENDERING OF AN IMAGE?
2.2- True and false colour composites
The display of a digital image on a computer screen is obvious: the "red", "green" and "blue" channels of the display system are associated with the image captured by the "red", "green" and "blue" channels of the digital camera or sensor, respectively. The resulting image therefore faithfully corresponds to what an observer's eye has seen on the spot. We speak of a true colour image in this case.
With remote sensing systems, however, it is also possible to detect and capture parts of the electromagnetic spectrum that are not detectable by the naked eye, such as infrared. To visualise this information, we associate the spectral bands of the sensing system with display colours of a screen (red-green-blue) that do not necessarily match them. In the process, colourful compositions that we call "false colour images" are created..
Such false colour views are very useful for analysing at a glance data coming from three spectral channels. After all, our eyes are powerful analysis tools that allow us to compare the colour of different objects, and on a false colour image, we can thus identify spatial patterns in the spectral variation of objects.
A commonly used false-colour representation is the so-called "false-colour infrared" composition. This combines the near-infrared, red and green bands of the sensor with the red, green and blue colours of the screen, respectively. This composition is very efficient for analysing vegetation because its spectrum has a clear peak in the near-infrared. In an image in "false infrared colours", vegetation with high photosynthetic activity appears bright red (near-infrared peak) while water appears practically black because it absorbs almost all wavelengths. In contrast, mineral surfaces (bare soil, concrete) appear in shades of blue to white.
|
True colours (Red-Green-Blue) |
False infrared colours (NIR-Red-Green) |
Colour composite (SWIR4-SWIR3-NIR). |
The MSI sensors on the Sentinel-2 satellites record radiation in 13 different spectral channels. By choosing different combinations of 3 of these channels, colour compositions can be created to highlight desired features. The images above show the southeast of the Brussels agglomeration and the Sonian Forest, captured by Sentinel-2B on 1 May 2024. |
||
Many other combinations are of course possible, depending on the spectral bands available. Whether they are useful depends on the purpose of the analysis and the objects or biophysical properties of the Earth's surface we want to highlight.
Now it's your turn! Create your own original colourful composition of the area of your choice at Sentinel Hub EO Browser.
