High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Deep Learning Based Weather Classification System

Author(s):

V. R. Patil , Department of CSE, Ashokrao Mane Group of Institutions, Vathar, MH, India.; P S Powar, Department of CSE, Ashokrao Mane Group of Institutions, Vathar, MH, India.

Keywords:

Weather Image Classification, Deep Learning, CNN, Prediction

Abstract

The recent work using Deep Learning has shown promising results in various domains. The present scenario of weather detection is laborious and time consuming. Therefore, it's crucial to include a preprocessing step that determines the weather for an incoming picture, followed by the appropriate deweathering procedures. This research considers the most prevalent weather scenarios to offer a deep learning-based environmental image classification system. The proposed system utilized four DCN pretrained networks for weather image classification. The best optimized ResNet attained with accuracy of 97.31%, precision of 86.99%, sensitivity of 86.44% and f1-score of 86.70%. Further, the proposed weather classification will be helpful in deployment of real time environment.

Other Details

Paper ID: IJSRDV11I90051
Published in: Volume : 11, Issue : 9
Publication Date: 01/12/2023
Page(s): 66-69

Article Preview

Download Article