Spectral Overlap Correction Using Weighted Deconvolution for Improved Dark Object Technique for Correction of EO1-Hyperi
- PDF / 772,935 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 59 Downloads / 197 Views
RESEARCH ARTICLE
Spectral Overlap Correction Using Weighted Deconvolution for Improved Dark Object Technique for Correction of EO1Hyperion Data Shailesh S. Deshpande1,2 • Arun B. Inamdar2 • Krishna Mohan Buddhiraju2
Received: 25 February 2016 / Accepted: 22 August 2017 Ó Indian Society of Remote Sensing 2017
Abstract We present a critical modification to improved dark object technique for correcting hyperspectral data (EO1-Hyperion). The modification is required in improved dark object technique as the original method does not take into account overlap of spectral response functions of two adjacent bands of hyperspectral sensor. We used weighted deconvolution for correcting the original overlap affected path radiance correction propagation factors. Further, we compared the reduction in correction factors—in different conditions—because of the overlap. We calculated the path radiance for April 22 Hyperion image and compared it with other methods such as 6SV. We found noticeable difference in corrected and uncorrected path radiance propagation factors with ‘‘clear’’ to ‘‘very clear’’ atmospheric models. For the other models (‘‘moderate’’, ‘‘hazy’’, ‘‘very hazy’’), the difference is negligible and can be ignored and improved dark object technique can be applied without any overlap correction. Keywords EO1-Hyperion Improved dark object technique Overlapping bands Spectral deconvolution
& Shailesh S. Deshpande [email protected] 1
Tata Research Development and Design Centre (A Division of Tata Consultancy Services), 54-B, Hadapsar Industrial Estate, Pune 411013, India
2
Centre of Studies in Resources Engineering, Indian Institute of Technology (IIT), Bombay, Powai, Mumbai 400076, India
Introduction Removing path radiance is an important initial step in processing digital images that are acquired remotely. Identifying a dark object (pixel/s) within the image and then removing radiance recorded from such object/s (additive component of atmospheric scattering) from every pixel is one of the common correction methods used extensively. Irrespective of simplifying assumptions (Vincent 1972; Chavez 1996), dark object method is computationally simple, and all the information can be derived from the image itself. For example, deep clear water body should not reflect infrared radiation and hence radiance from water-body pixel/s can be assumed to be a response of haze. Alternatively, one can use histogram to identify initial path radiance: use the offset in the histogram of each band as a path radiance and subtract the offset from each band respectively to correct the radiance. The assumption of the histogram method is that there is at least one (or a few) pixel/s having zero radiance given the histogram of a particular band contains sufficiently large number of pixels (Chavez 1988; Richards and Jia 2006). Generally, histogram would show some shift and/or the graph would rise sharply after some initial values. The initial point of inflection (or offset) can generally be taken as path radiance quantity for fur
Data Loading...
