Yasmine Benyettou* and Hadria Fizazi
University of Sciences and Technology of Oran, USTO, Algeria
*Corresponding Author: Yasmine Benyettou, University of Sciences and Technology of Oran, USTO, Algeria.
Received: September 24, 2020; Published: November 18, 2020
This paper presents a method to increase the classification performance of satellite images by swarm intelligence. Traditional statistical classifiers have limitations in solving complex classification problems due to their harsh assumptions because these methods only examine spectral variance by ignoring the spatial distribution of pixels corresponding to land cover classes and the correlation between the different bands. An optimization algorithm inspired by the behavior of pigeons is applied and has been used in various fields such as image restoration, planning of the trajectories of aerial robots. In our case, the basic idea is: Davies-Bouldin (DBI) is used as a fitness function. The iterative optimization process is carried out by the pigeon optimization algorithm. In this process, the fitness function matches the coordinate of the pigeon in optimizing the problem. The best result is obtained when the pigeon finds the best overall position. This method converts the problem of finding the optimal solution to the problem of solving multidimensional variables and efficiently optimizes the result. In order to verify the feasibility and accuracy of the supervised classification, the K-means bisecting technique and the deep learning method were implemented. The results of the comparison indicate that the method based on the inspired pigeon optimization is effective with a good classification rate equal to 95.60%, an accuracy rate of 84.70% in a reduced execution time of 19.15 dry. The results of the calculation also show that the proposed PIO algorithm can effectively improve the speed of convergence, and the superiority of the overall search.
Keywords: Pigeon Inspired Optimization (PIO) Algorithm; K-Means Bisecting; Deep Learning; Satellite Image; Classification
Citation: Yasmine Benyettou and Hadria Fizazi. “Application of the Pigeon Method to the Classification of Captured Data". Acta Scientific Computer Sciences 2.12 (2020): 27-35.
Copyright: © 2020 Yasmine Benyettou and Hadria Fizazi. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
+91 9182824667
Open Access by
Acta Scientific is licensed under a Creative Commons Attribution 4.0 International License
Based on a work at https://actascientific.com
ff
