This repository contains the code of the paper "An optimum multi-level image thresholding segmentation using non-local means 2D histogram and exponential Kbest gravitational search algorithm". The paper presents a novel two-dimensional (2D) histogram-based segmentation method for efficient multi-level image thresholding segmentation. The proposed method includes:
- A new non-local means based 2D histogram,
- A novel variant of gravitational search algorithm (exponential Kbest gravitational search algorithm), and
- A 2D Rényi entropy is redefined.
Experiments demonstrate the potential of the proposed method through subjective and objective evaluation of the Berkeley Segmentation Dataset and Benchmark (BSDS300).
This code is available for non-commercial purposes. Incase you find this useful for your work, do cite our work.
Mittal, Himanshu, and Mukesh Saraswat. "An optimum multi-level image thresholding segmentation using non-local means 2D histogram and exponential Kbest gravitational search algorithm." Engineering Applications of Artificial Intelligence, pp. 226-235, Vol. 71, 2018.
To run the code, execute the following file:
- main.m