DN2R: A tool to correct distortions of satellite remote sensing images due to atmospheric problems by converting digital numbers to reflectance

Abstract

Processing of digital satellite images is a collection of techniques and algorithms for the manipulation of digital images by computers. The great impediments in acquiring satellite images are objects in the atmosphere such as clouds, dusts and surfaces, which scatter the electromagnetic energy. This study aims at developing a tool for correcting the distortions of satellite images due to atmospheric effects. It is intended to develop an automated tool.

For this study, the past research work on the atmospheric correction of satellite images were reviewed, and based on the outcome of the past researches, advanced algorithms were used for the development. The development of the tool was done using Visual C++. This tool was developed for satellite images in Geotiff format, since Geotiff format is a non proprietary satellite image format. Satellite images from Landsat Pathfinder and Indian Remote Sensing Satellites were considered in this study.

The tool was tested using commercially available satellite image analysis software. Excel worksheets were used to evaluate the outcome of the development. Samples of pixels from the source and resulting images were selected, and the results were plotted and analysed.

This tool was evaluated by some of the researchers and scientists who use satellite images for analysis purposes. The results are satisfactory.