[Home ] [Archive]   [ فارسی ]  
:: Main :: About :: Current Issue :: Archive :: Search :: Submit :: Contact ::
Main Menu
Home::
Journal Information::
Articles archive::
For Authors::
For Reviewers::
Registration::
Contact us::
Site Facilities::
::
Search in website

Advanced Search
..
Receive site information
Enter your Email in the following box to receive the site news and information.
..
:: ::
Back to the articles list Back to browse issues page
Improving K-Means clustering algorithm using genetic algorithm to spatial analysis of oil spill detection in SAR polarimetric images
Mehrdad Kaveh , Yasser Ebrahimian Ghajari *
Babol Noshirvani University of Technology
Abstract:   (250 Views)
The existence of oil spills under ocean and sea surface is one of the big concerns for researchers in marine ecosystem terrains. In this study, K-Means clustering based genetic algorithm (GA) has been used in order to detect oil spills on sea surface. The main objective of developed K-Means with GA is to cause an intelligent search which not only randomly determines the initial clusters centers but also tries to find out the optimal centers for clusters. To achieve this goal, firstly required preprocessing steps for SAR images such as radiometric correction, speckle reduction and POLSAR feature extraction were done. Then the optimal cluster centers were determined by the developed K-Means by GA. Finally, to find out the final optimal centers, K-Means algorithm was used in way that the maximum similarity inner-classes was considered as cost function. The ground truth digitized from Pauli RGB image of POLSAR image was utilized to evaluate the performance of algorithms. According to the results, the results gained by the improved K-Means have more validity and precision than K-Means algorithm. As well, feature Entropy could obtain the overall accuracy by 83.24% that get success to provide the highest validity although it is the lowest accuracy in comparison with other features. Features ODD of Yamagouchi, ODD of Freeman decompositions and C11 of covariance matrix are succeeded to reach the overall accuracy by 90% but have the noticeable values of second type error by 18%, 11% and 12% which demonstrates the lowest validity in comparison with other features.
 
Keywords: Oil spill, SAR images, feature selection, K-Means, GA.
     
Type of Study: Research | Subject: GIS
Received: 2022/06/16 | Accepted: 2023/03/6 | ePublished ahead of print: 2024/08/6
Send email to the article author


XML   Persian Abstract   Print



Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Back to the articles list Back to browse issues page
نشریه علمی-پژوهشی مهندسی فناوری اطلاعات مکانی Engineering Journal of Geospatial Information Technology
Persian site map - English site map - Created in 0.04 seconds with 36 queries by YEKTAWEB 4660