| MACHINE LEARNING-BASED CLASSIFICATION OF AIRGLOW IMAGES FOR GRAVITY WAVE DETECTION FROM ALL-SKY CAMERA DATA |
| Paper ID : 1070-ISCHU |
| Authors |
|
Mostafa Nasser Sayed Ahmed *1, Mohamed Youssef Omar1, Hala El-Dosoqy1, Ayman Mohamed Mahrous2 1Helwan University 2Helwan University E-JUST |
| Abstract |
| Airglow phenomena provide valuable insights into the dynamics and processes occurring in the Earth's upper atmosphere. The detection of gravity waves within airglow images captured by all-sky cameras is of great interest for understanding atmospheric disturbances. In this study, we present a novel approach using machine learning techniques for the classification of airglow images obtained from the MacDonald Observatory's all-sky camera. Our objective is to distinguish between images containing gravity wave signatures and those that do not exhibit such features. We developed and trained a classification model on a comprehensive dataset spanning an 11-year period, from 2010 to 2020. The dataset encompasses a wide range of atmospheric conditions and gravity wave occurrences. Our methodology involves preprocessing the images, extracting relevant features, and employing advanced machine learning algorithms to classify them. The trained model aims to accurately differentiate between airglow images with gravity waves and those without. The results of our study demonstrate the potential of utilizing machine learning techniques to enhance the detection and classification of gravity waves within airglow images. This approach contributes to a deeper understanding of atmospheric dynamics and offers a valuable tool for researchers studying upper atmospheric phenomena. The extended timespan of the dataset allows for comprehensive analysis of gravity wave occurrences, leading to insights that could have broader implications for atmospheric science and space weather research. |
| Keywords |
| Gravity waves, Machine learning, All-sky camera. |
| Status: Abstract Accepted (Poster Presentation) |