Multiclass Weather Classification on Single Image Using Artificial Intelligence

Abstract

Weather classification through artificial intelligence (AI) has gained substantial attention due to its significance in various applications like autonomous driving, smart agriculture, and environmental monitoring. This study presents a multi-class weather classification approach specifically designed for single images utilizing AI techniques. The proposed model leverages deep learning architectures, primarily convolutional neural networks (CNNs), to discern between different weather conditions encompassing sunny, cloudy, rainy, snowy, and foggy scenarios. The dataset employed for training and evaluation comprises diverse and annotated images capturing various weather conditions under different settings. Preprocessing techniques involving data augmentation and normalization are applied to enhance the model's robustness and generalization capability. The CNN-based model is trained using a sizable dataset to learn intricate patterns and features representative of each weather category.