This project implements a Convolutional Neural Network (CNN) to recognize different weather conditions based on photos. The CNN is trained on a dataset containing images of various weather conditions such as cloudy, rainy, shining, and sunrise conditions.
The dataset consists of a collection of labeled images representing different weather conditions. Each image is labeled with the corresponding weather condition it represents. Dataset is available under url: http://www.kasprowski.pl/datasets/weather.zip
The dataset is organized into different folders, each representing a weather condition category. For instance:
cloudy/: Images depicting cloudy weatherrain/: Images depicting rainy weathershine/: Images depicting shining weathersunrise/: Images depicting sunrise scenes
The CNN architecture used for weather condition recognition involves several convolutional and pooling layers followed by fully connected layers.
The model is trained on the labeled dataset using appropriate preprocessing techniques, data augmentation (if required), and optimization algorithms.
The trained model's performance is evaluated on a separate test set to measure its accuracy in recognizing different weather conditions.
- Data Preparation: Ensure the dataset is structured correctly with labeled images in respective folders.
- Model Training: Run the training script to train the CNN on the dataset.
- Evaluation: Evaluate the model's performance on a separate test set using the provided evaluation script.
- TensorFlow
- Jupyter to run notebooks
- Data manipulation and visualization libraries (NumPy, Matplotlib, Opencv)
This project is licensed under the MIT License.