Nowadays, we are witnessing a steady increase in the number of available images, whether for experts or the general public, particularly in the age of Big Data, where data are distinguished by their volume, variety, authenticity and speed. Often, these vast databases contain numerous images, accompanied by texts and annotations more or less detailed The aim of this article is to respond to the growing need for high-performance tools for storing, research and exploring these vast image databases. In our context, the analysis of this massive data mainly aims at finding images based on their visual content. This approach enables the search of images based on some visual characteristics such as color patterns (RGB and HSV), shape and texture. The main aim of this research is to group images of crop diseases according to their similarities in a suitable storage system.
Big data, data lake, image, visual search, visual descriptor