General Information
    • ISSN: 1793-8201 (Print), 2972-4511 (Online)
    • Abbreviated Title: Int. J. Comput. Theory Eng.
    • Frequency: Quarterly
    • DOI: 10.7763/IJCTE
    • Editor-in-Chief: Prof. Mehmet Sahinoglu
    • Associate Editor-in-Chief: Assoc. Prof. Alberto Arteta, Assoc. Prof. Engin Maşazade
    • Managing Editor: Ms. Mia Hu
    • Abstracting/Indexing: Scopus (Since 2022), INSPEC (IET), CNKI,  Google Scholar, EBSCO, etc.
    • Average Days from Submission to Acceptance: 192 days
    • E-mail: ijcte@iacsitp.com
    • Journal Metrics:

Editor-in-chief
Prof. Mehmet Sahinoglu
Computer Science Department, Troy University, USA
I'm happy to take on the position of editor in chief of IJCTE. We encourage authors to submit papers concerning any branch of computer theory and engineering.

IJCTE 2017 Vol.9(6): 455-461 ISSN: 1793-8201
DOI: 10.7763/IJCTE.2017.V9.1186

Meteorological Visibility Evaluation on Webcam Weather Image Using Deep Learning Features

Shengyan Li, Hong Fu, and Wai-Lun Lo

Abstract—The estimation methods of meteorological visibility currently in use on digital image are mainly based on the meteorological laws and the corresponding features are extracted manually to calculate the visibility. However, besides the parameters of camera setting, the environmental and weather conditions will also affect the image quality and cause different kinds of noise to influence the evaluation accuracy. It is hard to extract all these factors manually and involve them into a certain equation to solve the visibility. Therefore, it is necessary to intelligently extract the useful factors and evaluate visibility. In this paper, an intelligent digital method is proposed to estimate the visibility on webcam weather image. In this method, a pre-trained convolutional neural network (CNN) is employed to automatically extract the visibility features instead of manual method and a generalized regression neural network (GRNN) is designed for intelligent visibility evaluation based on these deep learning features. A series of weather photo with the ground truth of visibility is used to train and test the proposed method. The result shows that the proposed automatic method with deep learning feature is workable for visibility prediction, whose accuracy is higher than that of the traditional method with hand-crafted features.

Index Terms—Meteorological visibility, weather photo, deep learning, feature extraction.

All authors are with the Department of Computer Science, Chu Hai College of Higher Education, Hong Kong (e-mail: syli@chuhai.edu.hk, hfu@chuhai.edu.hk, wllo@chuhai.edu.hk).

[PDF]

Cite:Shengyan Li, Hong Fu, and Wai-Lun Lo , "Meteorological Visibility Evaluation on Webcam Weather Image Using Deep Learning Features," International Journal of Computer Theory and Engineering vol. 9, no. 6, pp. 455-461, 2017.


Copyright © 2008-2024. International Association of Computer Science and Information Technology. All rights reserved.