Satellite Data Investigation for Change Estimation During COVID Era by Fusing Pixel and Object-Based Technique
International conference on Advanced Computing and Intelligent Technologies, ICACIT 2022
; 914:417-427, 2022.
Article
in English
| Scopus | ID: covidwho-2048179
ABSTRACT
In this investigation, an innovative combination of pixel-based change detection technique and object-based change detection technique is explored with the satellite images of Holy Masjid al-Haram, Saudi Arabia. The gray-level co-occurrence matrix (GLCM) method is used to quantify the texture of the remote sensing data through the texture classification approach on the satellite data in this work. GLCM produces results of the texture quantification in normalized form. Thus, applying a texture classification scheme on the satellite data is impressive to observe. Later maximum likelihood image classification approach is used for classification purposes. The classified information is categorized into four different classes. The kappa coefficient’s value and the overall accuracy for the pre- COVID classified study area are 0.6532 and 76.38%, respectively. During COVID, the classified study area presents the kappa coefficient and the overall accuracy of 0.7631 and 82.18%, respectively. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
COVID; GLCM; Maximum likelihood classification; Satellite; Texture; Change detection; Classification (of information); Maximum likelihood estimation; Object detection; Pixels; Remote sensing; Textures; Classification approach; Classifieds; Gray-level co-occurrence matrix; Grey-level co-occurrence matrixes; Kappa coefficient; Maximum-likelihood classification; Overall accuracies; Satellite data; Texture classification; Satellites
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Language:
English
Journal:
International conference on Advanced Computing and Intelligent Technologies, ICACIT 2022
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS