This project implements a web-app weather recognition classifier using deep learning with EfficientNetB4 architecture. The classifier is trained to classify images into 11 weather classes, including hail, snow, glaze, lightning, fog smog, frost, dew , rain, rainbow, rime, and sandstorm.
The classifier is trained on the Weather Dataset available on Kaggle. This dataset contains labeled images representing various weather conditions.
The classifier utilizes the EfficientNetB4 architecture, a state-of-the-art convolutional neural network (CNN) known for its efficiency and effectiveness in image classification tasks. EfficientNetB4 offers a good balance between model size and performance, making it suitable for deployment on resource-constrained devices.
The model is deployed using a Flask web application, allowing users to interact with it by uploading images and receiving predictions on weather conditions in real-time. The deployed application achieves 91% accuracy on the test set, demonstrating its effectiveness in recognizing weather patterns.
Project Link: Weather Classification Website