Image segmentation based on shape constraints through the ultimate levelings

• Grant number: 18/15652-7
• Support Opportunities: Regular Research Grants
• Duration: November 01, 2018 - October 31, 2020
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Wonder Alexandre Luz Alves
• Grantee: Wonder Alexandre Luz Alves
• Host Institution: Universidade Nove de Julho (UNINOVE). Campus Vergueiro. São Paulo , SP, Brazil
• Associated researchers: Maria Lucia Cardillo Corrêa Giannella ; Ronaldo Fumio Hashimoto ; Sidnei Alves de Araújo

Abstract

Generally, a typical image analysis problem consists of five basic steps: image acquisition; preprocessing; segmentation; representation and description; recognition and interpretation. Among these, we highlight the segmentation of images, a step that consists of partitioning the image domain in order to demarcate the objects of interest in the image.Therefore, we have to keep in mind that an inaccurate segmentation can compromise the results of the analysis. In addition, the complexity of the analyzed scene and the particular characteristics of each object lead to the segmentation task extremely complex since the objects of interest, in many applications, represent organs, cells, characters, machine components, among others.In most practical problems of image analysis the object shape is known a priori. Thus, it is highly desirable to incorporate such knowledge into the models and algorithms but this is not a trivial task.In this context, we intend to explore a framework based on residual operators defined in the context of Mathematical Morphology to solve problems of shape analysis. Thus, following the regular research project "Shape analysis through the ultimate levelings" - financed by FAPESP (grant number: 2016/02547-5), the objective of this research project is to continue studies on methods of shape analysis based on ultimate levelings. (AU)

Scientific publications

(8)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

ALVES, WONDER A. L.; GOBBER, CHARLES F.; SILVA, DENNIS J.; MORIMITSU, ALEXANDRE; HASHIMOTO, RONALDO F.; MARCOTEGUI, BEATRIZ. Image segmentation based on ultimate levelings: From attribute filters to machine learning strategies. PATTERN RECOGNITION LETTERS, v. 133, p. 264-271, MAY 2020. (18/15652-7, 15/01587-0)

DA SILVA, DENNIS JOSE; LUZ ALVES, WONDER ALEXANDRE; MORIMITSU, ALEXANDRE; GOBBER, CHARLES FERREIRA; HASHIMOTO, RONALDO FUMIO; BURGETH, B; KLEEFELD, A; NAEGEL, B; PASSAT, N; PERRET, B. Incremental Bit-Quads Count in Tree of Shapes. MATHEMATICAL MORPHOLOGY AND ITS APPLICATIONS TO SIGNAL AND IMAGE PROCESSING, ISMM 2019, v. 11564, p. 12-pg., 2019-01-01. (18/15652-7, 15/01587-0)

MORIMITSU, ALEXANDRE; LUZ ALVES, WONDER ALEXANDRE; SILVA, DENNIS JOSE; GOBBER, CHARLES FERREIRA; HASHIMOTO, RONALDO FUMIO; COUPRIE, M; COUSTY, J; KENMOCHI, Y; MUSTAFA, N. Minimal Component-Hypertrees. DISCRETE GEOMETRY FOR COMPUTER IMAGERY, DGCI 2019, v. 11414, p. 12-pg., 2019-01-01. (18/15652-7)

MORIMITSU, ALEXANDRE; LUZ ALVES, WONDER ALEXANDRE; DA SILVA, DENNIS JOSE; GOBBER, CHARLES FERREIRA; HASHIMOTO, RONALDO FUMIO; BURGETH, B; KLEEFELD, A; NAEGEL, B; PASSAT, N; PERRET, B. Incremental Attribute Computation in Component-Hypertrees. MATHEMATICAL MORPHOLOGY AND ITS APPLICATIONS TO SIGNAL AND IMAGE PROCESSING, ISMM 2019, v. 11564, p. 12-pg., 2019-01-01. (18/15652-7, 15/01587-0)

SILVA, DENNIS J.; ALVES, WONDER A. L.; HASHIMOTO, RONALDO FUMIO. Incremental bit-quads count in component trees: Theory, algorithms, and optimization. PATTERN RECOGNITION LETTERS, v. 129, p. 33-40, JAN 2020. (18/15652-7, 15/01587-0)

LUZ ALVES, WONDER ALEXANDRE; GOBBER, CHARLES FERREIRA; DA SILVA, DENNIS JOSE; MORIMITSU, ALEXANDRE; HASHIMOTO, RONALDO FUMIO; MARCOTEGUI, BEATRIZ; BURGETH, B; KLEEFELD, A; NAEGEL, B; PASSAT, N; et al. Ultimate Levelings with Strategy for Filtering Undesirable Residues Based on Machine Learning. MATHEMATICAL MORPHOLOGY AND ITS APPLICATIONS TO SIGNAL AND IMAGE PROCESSING, ISMM 2019, v. 11564, p. 13-pg., 2019-01-01. (18/15652-7)

MORIMITSU, ALEXANDRE; PASSAT, NICOLAS; ALVES, WONDER A. L.; HASHIMOTO, RONALDO F.. Efficient component-hypertree construction based on hierarchy of partitions. PATTERN RECOGNITION LETTERS, v. 135, p. 30-37, JUL 2020. (18/15652-7, 15/01587-0)

GOBBER, CHARLES F.; HASHIMOTO, RONALDO F.; ALVES, WONDER A. L.. An efficient algorithm to update non-flat and incremental attributes in morphological trees. PATTERN RECOGNITION LETTERS, v. 163, p. 8-pg., 2022-10-01. (15/22308-2, 18/15652-7)