MobileNetV2 based deep neural network for automated rain condition detection in UAV imagery

Unmanned Aerial Vehicles (UAVs) are used rapidly in different fields. Few of the important areas where UAVs are essential in disaster management and agriculture owing to their cost-effectiveness and accessibility to remote areas. However, adverse weather conditions like rain hinder their navigation. This study is a deep learning approach using MobileNetV2 to detect rainy conditions from UAV captured images. It aims to enhance the operational safety and efficiency. The balanced dataset of 245Kimages across seven rain classes was used. The dataset was divided into training, testing, and validation sets in the ratio 70:15:15. This convolutional neural network model achieved a test accuracy of 95.35%. This suggests that the model is reliable and robust and can be further researched for real-time deployment.