Titre : | Processing And Extracting Data From A Dataset For Artificial Intelligence IN Neural Radiance Fields.Traitement et extractiondes données d’un dataset via l’IA pour le Neural Radiance Fields | Type de document : | document multimédia | Auteurs : | Ferial Demana, Auteur ; Zineb Guesmia, Auteur ; Mohamed Redha Bouzidi, Directeur de thèse | Editeur : | Laghouat : Université Amar Telidji - Département d'électronique | Année de publication : | 2024 | Importance : | 60p. | Accompagnement : | 1 CD ROM Optique Némérique | Note générale : | Telecommunications and Networks
| Langues : | Anglais | Mots-clés : | Artificial intelligence, image processing, background removal, deep learning, model, convolutional neural networks, transfer learning.. | Résumé : | In this works the use of artificial intelligence techniques in image processing, with a focus on background removal in complex images such as trees. It analyzes the Roboflow platform as a powerful tool for data management and enhancement, demonstrating its use in transfer learning to improve model performance. Tools like Photoshop and deep learning techniques such as U-Net and GANs are also discussed, highlighting how Google Colab can be used effectively for prototyping models. The results show that AI significantly improves the accuracy and efficiency of background removal, reducing the manual effort required. The thesis works the continuous development of AI techniques to enhance image processing, paving the way for new and improved applications in the future. | note de thèses : | Memoire de Master en Electronique |
Processing And Extracting Data From A Dataset For Artificial Intelligence IN Neural Radiance Fields.Traitement et extractiondes données d’un dataset via l’IA pour le Neural Radiance Fields [document multimédia] / Ferial Demana, Auteur ; Zineb Guesmia, Auteur ; Mohamed Redha Bouzidi, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'électronique, 2024 . - 60p. + 1 CD ROM Optique Némérique. Telecommunications and Networks
Langues : Anglais Mots-clés : | Artificial intelligence, image processing, background removal, deep learning, model, convolutional neural networks, transfer learning.. | Résumé : | In this works the use of artificial intelligence techniques in image processing, with a focus on background removal in complex images such as trees. It analyzes the Roboflow platform as a powerful tool for data management and enhancement, demonstrating its use in transfer learning to improve model performance. Tools like Photoshop and deep learning techniques such as U-Net and GANs are also discussed, highlighting how Google Colab can be used effectively for prototyping models. The results show that AI significantly improves the accuracy and efficiency of background removal, reducing the manual effort required. The thesis works the continuous development of AI techniques to enhance image processing, paving the way for new and improved applications in the future. | note de thèses : | Memoire de Master en Electronique |
|