Unsupervised nonparametric detection of unknown objects in noisy images based on percolation theory

TL;DR

This paper introduces an unsupervised, nonparametric detection method for unknown objects in noisy images, leveraging percolation and random graph theories, suitable for real-time applications with irregular noise.

Contribution

It presents a scalable, linear-complexity algorithm capable of detecting objects of arbitrary shapes and sizes in highly irregular, unknown noise conditions, with proven consistency.

Findings

Algorithm has linear complexity and exponential accuracy.

Method is strongly consistent and scalable.

Effective for real-time detection in irregular noise environments.

Abstract

We develop an unsupervised, nonparametric, and scalable statistical learning method for detection of unknown objects in noisy images. The method uses results from percolation theory and random graph theory. We present an algorithm that allows to detect objects of unknown shapes and sizes in the presence of nonparametric noise of unknown level. The noise density is assumed to be unknown and can be very irregular. The algorithm has linear complexity and exponential accuracy and is appropriate for real-time systems. We prove strong consistency and scalability of our method in this setup with minimal assumptions.