DEEP GENERATIVE DISCRETE COSINE TRANSFORM FOR SPECTRAL IMAGE PROCESSING
Abstract
vioft2nntf2t|tblJournal|Abstract_paper|0xf4ff99612c0000004015060001000400
The ever-increasing number of publications and applications in the field of cross-spectral image processing has led to the area receiving greater focus than it previously had. In cross-spectral frameworks, the data from hyperspectral bands is blended with the data from other spectral bands in order to provide responses that are more robust to particular obstacles. Cross-spectral processing could be useful for a variety of applications, including dehazing, segmentation, calculating the vegetation index, and face identification, to name just a few of them. The availability of cross-and multi-spectral camera sets on the market, such as smartphones, multispectral cameras for remote sensing, or multi-spectral cameras for automotive systems or drones, has spawned an increased number of applications for these cameras. In this paper, we develop a deep generative discrete cosine transform for possible image processing for the enhancing the quality of images. This is conducted to improve the prediction or classification ability by the classifiers on hyperspectral images. The models are validated with various machine learning classifiers. The results of simulation shows that the proposed method higher degree of accuracy than the existing methods.

Authors
Mary Florida1, S.Vadhana Kumar2, A.J. Deepa3, M.L. Sworna Kokila4, S. Brilly Sangeetha5
St. Xavier’s Catholic College of Engineering, India 1, ILAHIA College of Engineering and Technology, India 2, Karunya Institute of Technology and Science, India3,Ponjsely College of Engineering, India4,IES College of Engineering, India 5

Keywords
Deep Generative Model, Discrete Cosine Transform, Spectral Image Processing
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
000000000000
Published By :
ICTACT
Published In :
ICTACT Journal on Image and Video Processing
( Volume: 12 , Issue: 4 , Pages: 2746-2749 )
Date of Publication :
May 2022
Page Views :
83
Full Text Views :

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.