Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2609.39997

Global existence for Consensus-Based Kinetic Models for Image Segmentation

Abramo Agosti·Pierluigi Colli·Elisabetta Rocca·Mattia Zanella
View original source ↗Read source document ↗Export metadata ↓

Abstract

In this article, we analyze a recently introduced kinetic model based on Hegselmann-Krause-type interaction dynamics describing consensus-based image segmentation in the mean-field regime, where particle density depends on spatial coordinates and the normalized gray level (feature) value. The model is formulated as an evolutive nonlocal partial differential equation featuring a non-local spatial drift term involving a bounded confidence interaction kernel, depending on spatial and feature proximity, and a spatial diffusion term degenerating in the feature variable representing aleatoric uncertainties. Establishing well-posedness on a bounded domain, essential for practical image processing, presents significant analytical challenges: the lack of uniform control of space derivatives due to kernels with jump discontinuities and degenerate spatial diffusion, complex non-standard boundary terms under homogeneous Neumann conditions, and a lack of compactness arising from the absence of derivatives with respect to the feature variable. To address these difficulties, we introduce a three-level regularization scheme. Specifically, we smooth the interaction kernel, apply structural boundary regularization to the non-local interaction term to simplify boundary conditions, and incorporate artificial diffusion in spatial and feature variables to resolve degeneracy. We prove the well-posedness and regularity of the regularized system via a finite difference scheme and study the asymptotic limit as regularization parameters vanish, proving the existence of weak/distributional solutions of the original model. This approach advances the mathematical theory of kinetic models for image segmentation by extending the framework of distributional solutions to bounded domains and non-smooth interaction kernels with jump discontinuities.

Explore related subjects

math.AP

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Abramo Agosti, Pierluigi Colli, Elisabetta Rocca, Mattia Zanella. 2026-09-30. Global existence for Consensus-Based Kinetic Models for Image Segmentation. https://arxiv.org/abs/2609.39997

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Positive solutions to general semilinear overdetermined boundary problems

We establish the existence of positive solutions to a general class of overdetermined semilinear elliptic boundary problems on suitable bounded open sets $Ω\subset\mathbb{R}^n$. Specifically, for $n\leq 4$ and under mild technical hypotheses on the coefficients and the nonlinearity, we show that there exist open sets $Ω\subset\mathbb{R}^n$ with smooth boundary and of any prescribed volume where the overdetermined problem admits a positive solution. The proof builds on ideas of Alt and Caffarelli on variational problems for functions defined on a bounded region. In our case, we need to consider functions defined on the whole $\mathbb{R}^n$, so the key challenge is to obtain uniform bounds for the minimizer and for the diameter of its support. Our methods extend to higher dimensions, although in this case the free boundary $\partialΩ$ could have a singular set of codimension 5. The results are new even in the case of the Poisson equation $-Δv =g(x)$ with constant Neumann data.

math.AP↗
arXiv2026

The fuzzy Landau equation: global well-posedness and Fisher information

We investigate a fuzzy variant of the inhomogeneous Landau equation and establish global-in-time existence and uniqueness of smooth solutions for a range of soft potentials. The spatial delocalization introduced in the collision operator not only prevents singularity formation and enhances regularity, but also reveals additional structural properties of the model. In particular, we show that several forms of the Fisher information decay monotonically or remain uniformly bounded in time.

math.AP↗
arXiv2026

Sharp boundary asymptotics and monotonicity for singular quasilinear elliptic equations with gradient terms in half-spaces

We establish sharp boundary asymptotics for directional derivatives and monotonicity of positive weak solutions to $-Δ_p u+\vartheta|\nabla u|^q=u^{-γ}+f(u)$ in the half-space $\mathbb R^N_+$ with zero Dirichlet data, where $p>1$, $γ>0$, $0 1),\\ (1-\log t)^{-\frac1p}\,\partial_ηu(x',t) &\longrightarrow \left(\frac{p}{p-1}\right)^{\frac1p} \langleη,e_N\rangle && (γ=1), \end{aligned} \] where $C_{γ,p}=\bigl((γ+p-1)^p/[p^{p-1}(p-1)(γ-1)]\bigr)^{1/(γ+p-1)}$. The convergence is uniform in the tangential variable; in particular, the appropriately rescaled tangential derivatives vanish and the leading constants depend only on $p$ and $γ$, not on the gradient term or $f$. The normal-derivative limits imply the corresponding exact leading-order boundary profiles of $u$ by integration. In the weakly singular regime $0<γ<1$, we obtain linear boundary bounds and positive inward directional-derivative estimates. We also establish local counterparts of the boundary results for solutions merely locally bounded up to the boundary. Finally, under additional structural assumptions, the moving-plane method yields normal monotonicity for strip-bounded solutions in arbitrary dimension and for locally bounded solutions in dimension two. The boundary asymptotics hold throughout the range $0<q\le p$.

math.AP↗