Résumé:
The work presented in this manuscript treats the problem of computer vision and image classification, recognition of objects and objects in images. Among the most used methods in the field of deep learning DL, we find Convolutional Neural Networks (CNNs) which can be considered as the best techniques currently used in the field and on which scientists are currently working.
Within this framework, we have developed an automated classifier that allows the classification of some high resolution images that represent a grouping of images from 10 categories (a combination of objects including animals and transportation). We proposed a new CNN model, consisting of several convolutional layers. We also explained and interpreted the obtained results.