Automatic design of image operators: extension and contextualization to not necessarily boolean lattices

• Grant number: 11/23310-0
• Support Opportunities: Scholarships in Brazil - Doctorate (Direct)
• Start date: April 01, 2012
• End date: March 31, 2017
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Roberto Hirata Junior
• Grantee: Igor dos Santos Montagner
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated scholarship(s): 14/21692-0 - Exploring high-level representations in image operator learning, BE.EP.DD

Abstract

The research on automatic design of image operators is motivated by the fact that the concatenation of image operators and the fine tuning of their parameters to solve a problem are not easy tasks. Significant advances were made on the realm of boolean lattices, that is, on the design of binary image operators. The same did not occur with non boolean lattices. The main reason is that, in this case, the challenges imposed are much larger, mainly because the size of the hypothesis space (number of operators) is $k^{k^{|W|}}$, where |W| is the size of neighborhood analyzed for the estimation of the operator and k is the number of gray levels considered. The objective of this research project is to advance the state-of-art on gray level operator design and apply the method to real world problems of great importance, like image segmentation and detection of markers in images. A second equally important objective is to make available to the scientific community an open source library containing the methods researched and implemented during this project. This library will be integrated to Python to increase the impact of this work, so the community will be able to use it to solve its image processing problems. (AU)

Scientific publications

(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)

HIRATA, NINA S. T.; MONTAGNER, IGOR S.; HIRATA, ROBERTO, JR.; ACM. Comics image processing: learning to segment text. PROCEEDINGS OF THE 1ST INTERNATIONAL WORKSHOP ON COMICS ANALYSIS, PROCESSING AND UNDERSTANDING (MANPU 2016), v. N/A, p. 6-pg., 2016-01-01. (11/23310-0, 15/01587-0, 15/17741-9, 14/21692-0)

MONTAGNER, IGOR S.; HIRATA, NINA S. T.; HIRATA, ROBERTO, JR.; CANU, STEPHANE; IEEE. Kernel approximations for W-operator learning. 2016 29TH SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI), v. N/A, p. 8-pg., 2016-01-01. (11/50761-2, 15/01587-0, 11/23310-0, 14/21692-0, 15/17741-9)

MONTAGNER, IGOR S.; HIRATA, NINA S. T.; HIRATA, JR., ROBERTO. Staff removal using image operator learning. PATTERN RECOGNITION, v. 63, p. 310-320, MAR 2017. (11/23310-0, 15/17741-9, 11/00325-1, 14/21692-0)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

MONTAGNER, Igor dos Santos. W-operator learning using linear models for both gray-level and binary inputs. 2017. Doctoral Thesis - Universidade de São Paulo (USP). Instituto de Matemática e Estatística (IME/SBI) São Paulo.