|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
In this research, a method is proposed to effectively compress hyper-spectral images in a lossless manner using machine learning techniques. The goal of this method is to enable the storage and transfer of larger amounts of hyper-spectral image data while preserving image details in both the spectral and spatial domains. The proposed system utilizes two consecutive neural layers to predict pixel values efficiently, minimize prediction errors, and keep computational complexity low. This approach has a wide range of applications in remote sensing and image manipulation. . In the proposed system, two consecutive neural layers are defined for performing an efficient compression without any loss of image details. This neural layer has an objective of minimize prediction errors with minimal computational complexity.
Keyword
Concatenation process, hyper spectral image, Neural network, Image manipulation.
PDF Download (click here)
|
|
|
|
|