A population-based automatic clustering algorithm for image segmentation

• Clustering is one of the prominent approaches for image segmentation. Conventional algorithms such as k-means, while extensively used for image segmentation, suffer from problems such as sensitivity to initialisation and getting stuck in local optima. To overcome these, population-based metaheuristic algorithms can be employed. This paper proposes a novel clustering algorithm for image segmentation based on the human mental search (HMS) algorithm, a powerful population-based algorithm to tackle optimisation problems. One of the advantages of our proposed algorithm is that it does not require any information about the number of clusters. To verify the effectiveness of our proposed algorithm, we present a set of experiments based on objective function evaluation and image segmentation criteria to show that our proposed algorithm outperforms existing approaches.

Ämnesord

NATURVETENSKAP  -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Computer Sciences (hsv//eng)

NATURVETENSKAP  -- Data- och informationsvetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences (hsv//eng)

Nyckelord

automatic clustering

human mental search

image segmentation

optimisation

population-based algorithms

Evolutionary algorithms

Optimization

Automatic clustering algorithm

Conventional algorithms

K-means

Local optima

Meta heuristic algorithm

Number of clusters

Optimisation problems

Population-based algorithm

K-means clustering

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)