{"title":"A Generic Interval of Linguistic Variable based Genetic Fuzzy Inference System; A utility in Forestry Application","authors":"Sadaf Jabeen, M. Awais, Basit Shafiq","doi":"10.1109/AECT47998.2020.9194207","DOIUrl":null,"url":null,"abstract":"A standard fuzzy rule relates variables with different linguistic labels where each linguistic label is defined through a membership function having membership value in a range from 0-1. The research presented in this paper extends this concept by associating each linguistic variable with intervals within the scope of its membership value to different classes. Thus, making a fuzzy rule more comprehensive and complete. The introduction of this concept has resulted in achieving better and at least comparable results with the standard fuzzy rule generation systems. The real task in implementing the proposed algorithm has been to determine these intervals. The present paper proposes the use of genetic algorithm with extended chromosome encoding to determine the interval of linguistic variables automatically. One of the main applications in which the proposed algorithm has been tested, is the forest inventory management and estimation. The forest inventory measurement includes vegetation cover, deforestation rate, crop degradation rate or vegetation index calculation. The key measurement in this regard is the amount of vegetation present. Generally, expensive equipment such as LIDAR and multispectral cameras are employed. With the use of the proposed approach vegetation estimation has been achieved using simple RGB cameras that are much cheaper. The proposed algorithm is not just limited to vegetation segmentation problem but is generic enough to be applied to datasets of different types and complexities. In order to establish this claim multiple datasets from UCI machine learning repository have been used to evaluate the proposed algorithm.","PeriodicalId":331415,"journal":{"name":"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Advances in the Emerging Computing Technologies (AECT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AECT47998.2020.9194207","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
A standard fuzzy rule relates variables with different linguistic labels where each linguistic label is defined through a membership function having membership value in a range from 0-1. The research presented in this paper extends this concept by associating each linguistic variable with intervals within the scope of its membership value to different classes. Thus, making a fuzzy rule more comprehensive and complete. The introduction of this concept has resulted in achieving better and at least comparable results with the standard fuzzy rule generation systems. The real task in implementing the proposed algorithm has been to determine these intervals. The present paper proposes the use of genetic algorithm with extended chromosome encoding to determine the interval of linguistic variables automatically. One of the main applications in which the proposed algorithm has been tested, is the forest inventory management and estimation. The forest inventory measurement includes vegetation cover, deforestation rate, crop degradation rate or vegetation index calculation. The key measurement in this regard is the amount of vegetation present. Generally, expensive equipment such as LIDAR and multispectral cameras are employed. With the use of the proposed approach vegetation estimation has been achieved using simple RGB cameras that are much cheaper. The proposed algorithm is not just limited to vegetation segmentation problem but is generic enough to be applied to datasets of different types and complexities. In order to establish this claim multiple datasets from UCI machine learning repository have been used to evaluate the proposed algorithm.