Plant Bounding Box Detection from Desirable Residues of the Ultimate Levelings

Autor(es)

Alves, Wonder A. L. ; Gobber, Charles F. ; Hashimoto, Ronaldo F. ; Campilho, A ; Karray, F ; Romeny, BT

• Tipo de documento: Artigo Científico
• Fonte: IMAGE ANALYSIS AND RECOGNITION (ICIAR 2018); v. 10882, p. 8-pg., 2018-01-01.

Resumo

Ultimate levelings are operators that extract important image contrast information from a scale space based on levelings. During the residual extraction process, it is very common that some residues are extracted from undesirable regions, but they should be filtered out. In order to attend this problem it can be used some strategies to filter residues extracted by ultimate levelings. In this paper, we selected desirable regions from the residual extraction process through a binary classifier. The selected regions are used later in a solution to the bounding box detection problem applied in a plant images dataset. (AU)

• Processo FAPESP: 16/02547-5 - Análise de formas por meio dos últimos levelings
• Beneficiário: Wonder Alexandre Luz Alves
• Modalidade de apoio: Auxílio à Pesquisa - Regular