{"title":"Color classification of solid wood floor based on LGBM","authors":"Yutu Yang, Binli Gou","doi":"10.1109/FENDT54151.2021.9749636","DOIUrl":null,"url":null,"abstract":"With the development of economy, solid wood flooring has become a new favorite of indoor decoration. The color of solid wood floor plays an important role in interior decoration quality. For 108 three types of solid wood floors provided by enterprises, the statistical characteristics of different color spaces of solid wood floor were extracted, and the LGBM model was used to classify and sort the importance of different characteristics to the classification results. Finally, 94.44% accuracy was obtained in 36 images of the test set, and find the effective characteristics of color feature representation of solid wood floor were the first-order moment of a, b channel in Lab color space and the first moment and second moment of H channel in HSV color space. (Abstract)","PeriodicalId":425658,"journal":{"name":"2021 IEEE Far East NDT New Technology & Application Forum (FENDT)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 IEEE Far East NDT New Technology & Application Forum (FENDT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/FENDT54151.2021.9749636","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
With the development of economy, solid wood flooring has become a new favorite of indoor decoration. The color of solid wood floor plays an important role in interior decoration quality. For 108 three types of solid wood floors provided by enterprises, the statistical characteristics of different color spaces of solid wood floor were extracted, and the LGBM model was used to classify and sort the importance of different characteristics to the classification results. Finally, 94.44% accuracy was obtained in 36 images of the test set, and find the effective characteristics of color feature representation of solid wood floor were the first-order moment of a, b channel in Lab color space and the first moment and second moment of H channel in HSV color space. (Abstract)