Weather Data For The Prevention Of Agricultural Production With Convolutional Neural Networks

We present in this document, an approach that uses artificial intelligence in the service of agriculture. We use convolutional neural networks to prevent the amount of agricultural production per hectare. First, we started with the massive collection of weather and meteorological data from the study area during the last 57 years. Secondly, we have moved to the data processing phase: normalization, filtering and segmentation of data. Thirdly, we design the learning and testing database, and then we looked for the most efficient neural architecture in terms of learning time and recognition rate. Finally, we tested the proposed neural model and presented our results and perspectives.