Road operatorsare using Intelligent Transport Systems composed of road side camerasfor traffic management. The artificial vision algorithmsused for automatic detection may be impacted by adverse weather conditions. Therefore, itis necessary to improve these algorithms in such conditions. In addition, the applications developed to operate inroad context impose a perfect reliability of operation including in adverse weather conditions. There are many works that allow weather classification but they do not take into consideration all the degraded conditions. In this paper, we proposea method based on convolutional neural networks to classify adverse weatherconditionsfrom a road camera. This method could beextendedto on-boardcamerasused by autonomous vehicles. We also present the weather image databasesthat we use to evaluate our learning