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
 
   
      07 April 2023, Volume 38 Issue 2   
    Article

    KALMAN BUCY REGION ADJACENT AND BIVARIATE CORRELATED CLASSIFICATION FOR DISASTER MANAGEMENT WITH SATELLITE IMAGES
    M.Nirmala, V.Saravanan
    Journal of Data Acquisition and Processing, 2023, 38 (2): 4278-4297 . 

    Abstract

    Over the recent few years, satellite systems and image analysis mechanisms have evolved to a scope where commercial Earth-observation instruments come up with notably to aid the management of crucial natural and technical disasters. Upon comparison with today’s accessibility of satellite imagery to the circumstances about ten years ago, the quantity, promptness and obtain-ability of satellite imagery covering a definite calamity state of affairs has boosted in a significant manner. In this work, a novel Kalman Bucy Region Adjacent and Bivariate Correlated Classification (KBRA-BCC) method is proposed for image classification to perform an efficient disaster management event. The Kalman Bucy Region Adjacent and Bivariate Correlated Classification (KBRA-BCC) method is split into three sections. They are, image denoising, image segmentation and image classification for disastrous area identification. With the input image obtained from satellite image database, first, a Kalman–Bucy Image Denoising process is carried out to eliminate noisy pixels therefore improving the PSNR. With the processed image, second, Region Segmentation process is performed by employing Region Adjacency Gray Level Image Segmentation that splits the processed image into different segments with respect to two distinct features, color and intensity. Finally, with the segmented images, Bivariate Correlated Classification is employed to perform correlation between input and training image (i.e., disastrous image) that in turn classifies the segmented image into disastrous image or non-disastrous image. By employing Bivariate Correlated Classification assists in performing efficient disaster management with better accuracy and minimal time consumption. The subjective and objective evaluation, as well as the peak signal to noise ratio (PSNR) along with the segmentation time and classification accuracy, is compared, respectively, showing that the KBRA-BCC method can effectively enhance the disaster management with satellite images.

    Keyword

    Disaster Management, Kalman–Bucy, Image Denoising, Region Adjacency, Gray Level, Image Segmentation, Bivariate Correlated Classification


    PDF Download (click here)

SCImago Journal & Country Rank

ISSN 1004-9037

         

Home
Editorial Board
Author Guidelines
Subscription
Journal of Data Acquisition and Processing
Institute of Computing Technology, Chinese Academy of Sciences
P.O. Box 2704, Beijing 100190 P.R. China
E-mail: info@sjcjycl.cn
 
  Copyright ©2015 JCST, All Rights Reserved