Class Notes (1,100,000)
CA (620,000)
Brock U (10,000)
ERSC (200)
Lecture 4

ERSC 3P07 Lecture Notes - Lecture 4: Convolution, Charleston, South Carolina, Fuselage


Department
Earth Sciences
Course Code
ERSC 3P07
Professor
Marilyne Jollineau
Lecture
4

This preview shows pages 1-2. to view the full 6 pages of the document.
Lecture 4: Data preprocessing and transformations of image data
Radiometric Correction
Atmospheric Errors
Instrument Errors
Correction for Atmospheric Errors
Do nothing (e.g., land-use and land-cover
mapping where surface features are normally
quite different spectrally)
Take in-situ measurements
Model the atmosphere (correct for date, sun
altitude, latitude and longitude)
Model the atmosphere and take in-situ measurements
Use multiple looks or multiple bands
Absolute Radiometric Correction
A. Image containing substantial haze prior to atmospheric correction.
B. Image after atmospheric correction algorithm is applied.
Relative Radiometric Correction

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

Instrument Errors
Random bad pixels (AKA shot noise)
Line-start/stop problems
Line or column drop-outs
Line or column striping
Line Dropout
Solution: take the average of the line above and below the drop-out line.
Line Striping
Solution: compute histograms over homogeneous area and apply corrections.
You're Reading a Preview

Unlock to view full version