International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
Volume 186 - Issue 30 |
Published: July 2024 |
Authors: Sachin Kumar, Shiwalika Sambyal, Sourabh Shastri, Vibhakar Mansotra |
![]() |
Sachin Kumar, Shiwalika Sambyal, Sourabh Shastri, Vibhakar Mansotra . Enhancing Satellite Remote Sensing Image Classification using Two Layer Convolutional Neural Network. International Journal of Computer Applications. 186, 30 (July 2024), 18-23. DOI=10.5120/ijca2024923839
@article{ 10.5120/ijca2024923839, author = { Sachin Kumar,Shiwalika Sambyal,Sourabh Shastri,Vibhakar Mansotra }, title = { Enhancing Satellite Remote Sensing Image Classification using Two Layer Convolutional Neural Network }, journal = { International Journal of Computer Applications }, year = { 2024 }, volume = { 186 }, number = { 30 }, pages = { 18-23 }, doi = { 10.5120/ijca2024923839 }, publisher = { Foundation of Computer Science (FCS), NY, USA } }
%0 Journal Article %D 2024 %A Sachin Kumar %A Shiwalika Sambyal %A Sourabh Shastri %A Vibhakar Mansotra %T Enhancing Satellite Remote Sensing Image Classification using Two Layer Convolutional Neural Network%T %J International Journal of Computer Applications %V 186 %N 30 %P 18-23 %R 10.5120/ijca2024923839 %I Foundation of Computer Science (FCS), NY, USA
Due to the recent advancements in deep learning techniques, image classification based on Convolutional Neural Networks (CNN) has acquired an important place in various applications of remote sensing data. Tracking clouds for weather prediction, vegetation, wildlife, prior prediction of hurricanes, storms, and environment monitoring are a few of remote sensing applications in different domains. Still, all these applications require the identification and classification of images. Researchers have been implementing various strategies for classifying the images so that more accurate and appropriate techniques in hand for different applications. The experiment focuses on enhancing the existing accuracy by proposing a novel architecture for classification. In this paper, a novel CNN architecture for classifying 5631 images (Cloudy, Desert, Water, Green area) has been proposed. The proposed model has achieved an overall accuracy of 99.11%, which is better than any traditional approach.