Pre-Processing of the Remotely Sensed Images: Feature Extraction, Radiometric Corrections, Geometric Corrections, and Atmospheric Correction (Especially for GATE-Geospatial 2022)

Doorsteptutor material for CBSE/Class-6 is prepared by world's top subject experts: get questions, notes, tests, video lectures and more- for all subjects of CBSE/Class-6.

Examrace Books on Mapping, GIS, and Remote Sensing prepares you throughly for a wide range of practical applications.

When remotely sensed data is received from the imaging sensors on the satellite platforms it contains flaws and deficiencies. It must be processed using several steps:

Segmentation of Remote Sensing Imagery|Papers with Code

Pre-processing refers to those operations that are preliminary to the main analysis. Pre-processing includes a wide range of operations from the very simple to extremes of abstractness and complexity. These categorized as follow:

  • Feature Extraction
  • Radiometric Corrections
  • Geometric Corrections
  • Atmospheric Correction

The techniques involved in the removal of unwanted and distracting elements such as image/system noise, atmospheric interference and sensor motion from an image data occurred due to limitations in the sensing of signal digitization, or data recording or transmission process. Removal of these effects from the digital data is said to be “restored” to their correct or original condition, although we can, of course never know what are the correct values might be and must always remember that attempts to correct data what may themselves introduce errors. Thus, image restoration includes the efforts to correct for both radiometric and geometric errors.

Feature Extraction

Feature Extraction does not mean geographical features visible on the image but rather “statistical” characteristics of image data like individual bands or combination of band values that carry information concerning systematic variation within the scene. Thus, in a multi-spectral data, it helps in portraying the necessary elements of the image. It also reduces the number of spectral bands that have to be analyzed. After the feature extraction is complete the analyst can work with the desired channels or bands, but in turn, the individual bandwidths are more potent for information. Finally, such a pre-processing increases the speed and reduces the cost of analysis.

Radiometric and Atmospheric Corrections

Radiometric Corrections are carried out when an image data is recorded by the sensors, they contain errors in the measured brightness values of the pixels. These errors are referred to as radiometric errors and can result from the

  • Instruments used to record the data
  • From the effect of the atmosphere

Radiometric processing influences the brightness values of an image to correct for sensor malfunctions or to adjust the values to compensate for atmospheric degradation. Radiometric distortion can be of two types:

  • The relative distribution of brightness over an image in each band can be different to that in the ground scene.
  • The relative brightness of a single pixel from band to the band can be distorted compared with spectral reflectance character of the corresponding region on the ground.

The following methods define the outline the basis of the cosmetic operations for the removal of such defects:

Line-Dropouts and Corrections

A string of adjacent pixels in a scan line contains spurious DN. This can occur when a detector malfunction permanently or temporarily. Detectors are loaded by receiving sudden high radiance, creating a line or partial line of data with the meaningless DN. Line dropouts are usually corrected either by replacing the defective line with a duplicate of preceding or subsequent line or taking the average of the two.

- Line-Dropouts


Banding or striping occurs if one or more detectors go out of adjustment in each band. The systematic horizontal banding pattern seen on images produced by electro-mechanical scanners results in repeated patterns of lines with consistently high or low DN. Two reasons can be thus put forward in favour of applying a ‘de-striping’ correction:

  • The visual appearance and interpretability of the image are thereby improved.
  • Equal pixel values in the image are more likely to represent areas of the equal ground leaving radiance, other things being equal.

Two Different Methods of De-Striping

The two different methods of de-striping are as follow:

  • The first method entails a construction of histograms for each detector of the problem band, i.e.. , histograms generated from detectors: these histograms are calculated for, the lines 1,7, 13; lines 2,8, 14, etc. Then the means and standard deviation are calculated for each of the histograms. Assuming the proportion of pixels representing different soils, water, vegetation, cloud, etc. are the same for each detector, the means and standard deviations of the histograms should be the same. Stripes, however, are characterised by distinct histograms. De-striping then requires equalisation of the means and standard deviation of the detectors by forcing them to equal selected values - usually the mean and standard deviation for the whole image. The process of histogram matching is also utilized before mosaicing image data of adjacent scenes (recorded at different times) to accommodate differences in illumination levels, angles etc. A further application is resolution merging, in which a low spatial resolution image is sharpened by merging with high spatial resolution image.
  • The second method is a non-linear in the sense that the relationship between radiance (received at the detector) and rout (output by the sensor) is not describable in terms of single linear segments.

Random Noise

Odd pixels that have spurious DN crop up frequently in images - if they are particularly distracting, they can be suppressed by spatial filtering. By definition, these defects can be identified by their marked differences in DN from adjacent pixels in the affected band. Noisy pixels can be replaced by substituting for an average value of the neighbourhood DN. Moving windows of 3 x 3 or 5 x 5 pixels are typically used in such procedures.

Geometric Corrections

Geometric correction is undertaken to avoid geometric distortions from a distorted image, and is achieved by establishing the relationship between the image coordinate system and the geographic coordinate system using calibration data of the sensor, measured data of position and attitude, ground control points, atmospheric condition etc.

Steps in Geometric Correction

The steps to follow for geometric correction are as follows:

Illustration 2 for Steps_in_Geometric_Correction
  • Selection of method: After consideration of the characteristics of the geometric distortion as well as the available reference data, a proper method should be selected.
  • Determination of parameters: Unknown parameters which define the mathematical equation between the image coordinate system and the geographic coordinate system should be determined with calibration data and/or ground control points.
  • Accuracy check: Accuracy of the geometric correction should be checked and verified. If the accuracy does not meet the criteria, the method or the data used should be checked and corrected in order to avoid the errors.
  • Interpolation and resampling: Geo-coded image should be produced by the technique of resampling and interpolation. There are three methods of geometric correction as mentioned below.
    • Systematic correction: When the geometric reference data or the geometry of sensor are given or measured, the geometric distortion can be theoretically or systematically avoided. For example, the geometry of a lens camera is given by the collinearity equation with calibrated focal length, parameters of lens distortions, coordinates of fiducial marks etc. The tangent correction for an optical mechanical scanner is a type of system correction. Generally systematic correction is enough to remove all errors.
    • Non-systematic correction: Polynomials to transform from a geographic coordinate system to an image coordinate system, or vice versa, will be determined with given coordinates of ground control points using the least square method. The accuracy depends on the order of the polynomials, and the number and distribution of ground control points.
    • Combined method: Firstly, the systematic correction is applied, then the residual errors will be reduced using lower order polynomials. Usually the goal of geometric correction is to obtain an error within plus or minus one pixel of its true position.
Illustration 3 for Steps_in_Geometric_Correction

Developed by: