Open the 'weatherclassification.ipynb' python notebook with jupyter and run all the code. All the other python files represent classes that are used inside the notebook.
The purpose of this project is to classify a set of images based on the weather conditions. For doing this we developed an algorithm that computes a descriptor for each image (based on the image colors, sharpness, gradient, and luminance) and then classifies the images exploiting those descriptors and a neural network.
See the 'report.pdf' file for detailed informations about the project