# Comparison of Possibilistic Fuzzy Local Information C-Means and Possibilistic K-Nearest Neighbors for Synthetic Aperture Sonar Image Segmentation

## TL;DR

This paper compares two possibilistic segmentation algorithms, PFLICM and PKNN, for high-resolution synthetic aperture sonar images, evaluating their effectiveness in seafloor environment segmentation.

## Contribution

It provides a comparative analysis of semi-supervised and supervised possibilistic segmentation methods applied to SAS imagery, including quantitative performance assessment.

## Key findings

- PFLICM effectively detects outlier seafloor environments.
- PKNN performs well with labeled training data.
- Both algorithms show promise for automated SAS image segmentation.

## Abstract

Synthetic aperture sonar (SAS) imagery can generate high resolution images of the seafloor. Thus, segmentation algorithms can be used to partition the images into different seafloor environments. In this paper, we compare two possibilistic segmentation approaches. Possibilistic approaches allow for the ability to detect novel or outlier environments as well as well known classes. The Possibilistic Fuzzy Local Information C-Means (PFLICM) algorithm has been previously applied to segment SAS imagery. Additionally, the Possibilistic K-Nearest Neighbors (PKNN) algorithm has been used in other domains such as landmine detection and hyperspectral imagery. In this paper, we compare the segmentation performance of a semi-supervised approach using PFLICM and a supervised method using Possibilistic K-NN. We include final segmentation results on multiple SAS images and a quantitative assessment of each algorithm.