Development of Computational Method for Systematic Identification of Paths on not easily delimited Manifolds: Application to Nearby Spiral Galaxies

• Grant number: 25/25461-8
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: December 01, 2025
• End date: November 30, 2026
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Claudia Lucia Mendes de Oliveira
• Grantee: Branco Fernandes Peixoto
• Host Institution: Instituto de Astronomia, Geofísica e Ciências Atmosféricas (IAG). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 19/26492-3 - Science with the Brazilian robotic telescope, AP.ESP

Abstract

This study proposes the development of an image analysis method to map the arms of spiral galaxies and computationally characterize the unwinding phenomenon, generally caused by drag pressure in galaxy cluster environments. The project is extremely relevant given the fact that there are no methods for identifying this phenomenon, other than direct observation by researchers. In particular, it is worth highlighting the lack of systematic identification, possibly on a large scale, and with mathematical parameters (and respective ranges) that can be objectively measured, presented, and discussed in the literature as indicators of unwinding.Unsurprisingly, the unwinding phenomenon still has few records in the literature. The image analysis method developed could also be of great use for various applications in astronomy and other fields that focus on identifying paths (patterns) in images with high variability, whose surfaces of interest are erratically delimited and of low or intermediate sharpness. This study is based on data already collected by the student who will be awarded the scientific initiation scholarship and that was obtained via a previous project discussed in this proposal and whose reports are attached. The results obtained and the discussion presented in these past reports strongly suggest that iterative convergence methods, equipped with predefined mathematical reward functions, do not guarantee the adequate identification of the desired patterns for the spiral arms, which further emphasizes the importance of another, more contemplative and reliable alternative, which is the analysis proposed here. (AU)