Weather classification is an essential tool for meteorologists and weather forecasters to predict weather patterns and communicate them to the public.The use of a pre-trained model on a new problem is known as transfer learning in machine learning.
In this project, the system classifies the different types of weather. These weathers are majorly classified into 5 categories namely Cloudy, Shine, Rain, Foggy, Sunrise. Deep-learning (DL) methods in artificial intelligence (AI) play a dominant role as high-performance classifiers.
Transfer learning has become one of the most common techniques that has achieved better performance in many areas, especially in image analysis and classification.Transfer Learning techniques like Inception V3, VGG19, Xception V3 that are more widely used as a transfer learning method in image analysis and they are highly effective which are used here.
You can download the dataset used in this project using the link : https://www.kaggle.com/datasets/vijaygiitk/multiclass-weather-dataset
The user interacts with the UI to choose an image. The chosen image is processed by a VGG19 deep learning model. The VGG19 model is integrated with a Flask application. The VGG19 model analyzes the image and generates predictions. The predictions are displayed on the Flask UI for the user to see. This process enables users to input an image and receive accurate predictions quickly.
- NumPy
- Pandas
- Matplotlib
- Keras
- TensorFlow
- Flask
- CNN
- Vgg19
- Xception
- Inception-V3
- Resnet
- Flask