Remote sensing techniques using satellite imagery are widely applied in various fields of agricultural sciences due to the ease of obtaining real-time information, enabling data acquisition from a region without the need for physical displacement, thus avoiding costs. It also allows for the exploration of more efficient methods for crop monitoring tasks. This article presents a software platform that collects images from publicly available sources on the internet, groups these images by category and geographic region, and allows for the application of different artificial intelligence algorithms for image classification. The experiments conducted achieved an accuracy of 89.54% for the Random Forest (RF) classifier. A neural network with three hidden layers and one output layer achieved an accuracy of 86.66%. It is believed that the lower performance is related to the small number of samples used for training the network. The results demonstrate the potential benefits involved in applying this platform in precision agriculture, considering the ability to acquire, organize, and apply artificial intelligence to image classification.