An interactive machine for algae image retrieval

Hassan Tabout (1) , Abdelmoghit Souissi (1) , Youssef Chahir (2) , Abderrahman Sbihi (3)

1 LASTID
2 Equipe Image - Laboratoire GREYC - UMR6072
3 UAE - Abdelmalek Essaadi University [Tétouan] = Université Abdelmalek Essaadi [Tétouan]

We study in this paper the problem of using multiple-instance semi-supervised learning to solve image relevance feedback problem. Many multiple-instance learning algorithms have been proposed to tackle this problem; most of them only have a global representation of images. In this paper, we present a semi-supervised version of multiple instance learning. By taking into account both the multiple-instance and the semi-supervised properties simultaneously, a novel graph-based algorithm is developed, in which global and local information are used. Experimental results show promising results of the proposed method for a test database containing more than 2000 color seaweed images.

Mots clés: Multi Instance Learning; Relevance Feedback; Seaweed Images; Semi-Supervised Learning

Domaines: Traitement des images [eess.IV]

Hassan Tabout, Abdelmoghit Souissi, Youssef Chahir, Abderrahman Sbihi. An interactive machine for algae image retrieval. IEEE ICCSIT 2009, Aug 2009, Beijing, China. pp.176 - 180, ⟨10.1109/ICCSIT.2009.5234597⟩. ⟨hal-00815331⟩