3D openjpeg2000 with integer klt
Loading...
Date
2018-06
Authors
Shahrul Azmam Mamat
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
A hyperspectral image processing system especially remote sensing technology is
very important as it able to provide extremely useful information for the various type of
Earth monitoring system – terrains, weather, oceanology, pollution control and etc.
Hyperspectral image which includes the spatial (inter-pixel) and spectral (multi-component images) redundancy. The amount of acquired data of the hyperspectral image
greatly increased since the spectral and spatial resolution of sensors which monitoring
Earth terrains increase as well. Increased size of imagery data will require an image
compression as there are lot issues need to overcome, such as downlink speed for
bandwidth image and on-board satellite database storage are limited. The distance and
downlink speed for image transmission from Earth Observation (EO) Satellites to the
ground station need to be measured. The developed image processing system need to
optimize the compression performance of image data in order to maximize the
transmission of image data as well as high speed downloading capacity. The compression
and decompression algorithm include in Integer Karhunen-Loève Theorem (Integer KLT)
which efficient in spectral decorrelation and OpenJPEG2000 is a suitable for spatial
decorrelation. The lossless compression ratio (CR) and execution time (ET) of
hyperspectral image in *.rawl file from OpenJPEG2000 will be compared to get better
result which will lead the improvement of EO Technology for reduction of data size
which allows more images to be stored in the on-board satellite database storage and also
shorter the time required to transfer the images from EO satellites to the ground station
within a single pass.