Digital Image Processing Tutorial-Image Enhancement Part-III (Spectral Enhancement)

Spectral Enhancement of a Satellite Image

Spectral enhancement is directly representing spectral properties of sensor or satellite image. This spectral property of satellite image is described by spectral resolution. 

Spectral Resolution refers to the number of bands (wavelength) in the spectrum measured by sensor. It is capability of sensor board on satellite to record reflectance of earth surface feature in different wavelength/multi band

Spectral resolution describes the ability of a sensor to define fine wave length intervals.

This property of sensor is very significant in collecting information of earth surface, as earth surface react with incident energy (Electromagnetic radiation from Sun) and reflectance occur due to this process recorded by sensor. Earth surface feature react with radiated energy in three different way
  • Transmission
  • Absorption
  • Reflectance
Reflectance is recorded by sensor and it is completely depend on incident energy from Sun. In other words depend on portion of electromagnetic energy. (The complete details of electromagnetic energy and its bifurcation is already provide in earlier article)

Reflectance will change with change in incident energy such as water body on earth surface will have different reflectance in blue wave length, red wavelength and infrared wave length. 

So it is important to collect complete reflectance from surface feature, therefore deploy different sensors which can record re-radiated energy from earth surface feature.

So commonly used sensors are for Blue, Green and red wavelength from visible part and near infrared from Infrared portion of electromagnetic radiation. When spectral enhancement is applied, it is understood that one is modifying the spectral values of satellite image.

This Spectral enhancement process in Erdas imagine described by below steps.

Apply Spectral Enhancement to an image

In the Image Interpreter menu, click Spectral Enhancement.

Use Tasseled Cap

The Tasseled Cap  is a method of image band conversion to enhance spectral information of a satellite image. It mainly is used to interpret vegetation using the derived information from different bands or wavelengths. 

In the Spectral Enhancement menu, select Tasseled Cap. The Tasseled Cap dialog opens.

Under Input File, enter jaipur.img. That image is a Landsat TM image of Jaipur.
Enter tasseled.img in the directory of your choice as the Output File name.

Under Output Options, turn on the Stretch to Unsigned 8 bit checkbox by clicking on it.

Click Set Coefficients.The Tasseled Cap Coefficients dialog opens.
For this exercise, you use the default entries, although you may change these entries at any time.

Click OK in the Tasseled Cap Coefficients dialog. Click OK in the Tasseled Cap dialog to start the function. A Job Status dialog opens to report the state of the job.

When the Job Status dialog indicates that the job is Done, click OK.

Check Results
Open a Viewer and display jaipur.img.

Open a second Viewer and then open the Select Layer To Add dialog by clicking on the Open icon in the Viewer toolbar.

In the Select Layer To Add dialog, enter the name of the directory in which you saved tasseled.img, press Enter on your keyboard, and then click tasseled.img in the file list to select it.

Click the Raster Options tab at the top of the Select Layer To Add dialog. Under Layers to Colors, use layer 1 as Red, layer 2 as Green, and layer 3 as Blue. Click OK in the Select Layer To Add dialog.

The image, tasseled.img, shows a degree of brightness, greenness, and wetness, as calculated by the Tasseled Cap coefficients used.

Layer 1 (red) = the brightness component (indicates areas of low vegetation and high reflectors)
Layer 2 (green) = the greenness component (indicates vegetation)
Layer 3 (blue) = the wetness component (indicates water or moisture)

Importance of Tasseled cap in Image enhancement

Tasseled cap describe all three aspects of image specially for vegetation such as greenness, brightness and wetness where as vegetation index (such as NDVI) describes only greenness.



Aerial Photographs based Height Measurement using Parallax

GIS/Geoinformatics Educational Institute and Courses in India

How to Download Free Satellite Images From Bhuvan Portal of NRSC