Integer Klt For Lossless Hyperspectral Image Compression on BeagleBone Black Board
PDF

Keywords

integer KLT
BeagleBone
Black Board implementation
execution time

How to Cite

Mohd Ridzuan Tan, N. D., Mat Noor, N. R., Othman, W. A. F. W., Bakar, E. A., & Hawary, A. F. (2019). Integer Klt For Lossless Hyperspectral Image Compression on BeagleBone Black Board. ROBOTIKA, 1(2), 24-31. Retrieved from http://technical-journals.org/index.php/ROBOTIKA/article/view/37

Abstract

This research concerned with the lossless hyperspectral image compression for satellite imagery using Integer Karhunen–Loève Transform (KLT) on BeagleBone Black Board. The Integer KLT is selected as the transform for compression due to it showing superior performance in decorrelating the spectral component in hyperspectral images compared to other algorithms. The objective of this research is to develop an Integer KLT algorithm to implement it into the embedded system. The performance of the algorithm on the board in terms of execution time is investigated. The implementation of the Integer KLT algorithm is executed into the BeagleBone Black board and tested for its performance, profiling the execution time and comparing it with other embedded platforms which are a low-power DSP platform. The implementation shows a faster execution time compared to the low-power DSP platform.

PDF